From 72344fce93b920b6d3241793b68735d944cdef98 Mon Sep 17 00:00:00 2001 From: Yichen Jiang Date: Mon, 10 Aug 2026 22:40:57 +0800 Subject: [PATCH 01/16] fix(web): edit a declared provider's name and protocol MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The Models editor card curated its pi-ai fields by what every route has, so `displayName` and `api` — the two a hand-declared route names for itself — were asked for at creation and then reachable only through settings.yaml. The editor now renders both for a route the directory reports as declared, from the same namespace schema the create card reads. A catalog route gets neither: it defaults its name from its catalog entry, and its models each carry their own protocol, so a route-level one could only override all of them. Clearing the name unsets it and the route falls back to its id, which is what the field's placeholder shows; storing the empty string would be refused by the adapter. A declared profile naming no protocol selects nothing rather than reading as if it had picked the first choice. The Provider ID stays fixed: it is the settings dict key, it is referenced from `agent-default-model` and every logged request header, and it is the stem of a credential reference the page can never read back to move. Fixes #2204 --- ...-a-provider-from-the-models-page.i18n.yaml | 4 +- ...claring-a-provider-from-the-models-page.md | 12 ++- ...ring-a-provider-from-the-models-page.zh.md | 12 ++- apps/web/tests/models-settings.e2e.ts | 36 ++++++- .../models-settings/declared-edit.expected.md | 65 +++++++++++++ docs/user/guide/providers.i18n.yaml | 4 +- docs/user/guide/providers.md | 2 + docs/user/guide/providers.zh.md | 2 + packages/client/ui-models/README.i18n.yaml | 4 +- packages/client/ui-models/README.md | 4 +- packages/client/ui-models/README.zh.md | 4 +- .../ui-models/src/client/ModelsSection.tsx | 7 ++ .../ui-models/src/client/ProviderEditor.tsx | 72 +++++++++++++- .../ui-models/tests/provider-form.spec.tsx | 93 +++++++++++++++++++ 14 files changed, 300 insertions(+), 21 deletions(-) create mode 100644 apps/web/tests/snapshots/models-settings/declared-edit.expected.md diff --git a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml index c34594bd36..bff5cc88fc 100644 --- a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml +++ b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md -2026-08-04-declaring-a-provider-from-the-models-page.md: 44441da5427dbcd3f0651c2c137a5132ff6d968b -2026-08-04-declaring-a-provider-from-the-models-page.zh.md: 974c4fa5be0893e8fdf6a9142877a8b966708dab +2026-08-04-declaring-a-provider-from-the-models-page.md: 3c63254364e3803e9bddab2a5aad4c092e4f0426 +2026-08-04-declaring-a-provider-from-the-models-page.zh.md: c38b0387b3c7778b93fb43232c22a7acb24dc281 diff --git a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md index 44441da542..3c63254364 100644 --- a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md +++ b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md @@ -22,22 +22,30 @@ Fetching asks about the endpoint **the form currently shows** — a base URL edi The protocol choices come from the namespace's **own schema**, read through the settings descriptor the page already fetches (`providers.*.api` is a union of the adapter's `supportedProtocols()`). No new wire field, no constant in the client, and no way for the offered choices to drift from the accepted ones. +The editor reaches the two fields a route the directory reports as **declared** names for itself — its display name and that protocol. A create card asking for a field no editor can change leaves that field reachable only through `settings.yaml`, which is the posture this note set out to end. Both render in the fold beside the endpoint, the protocol from the same schema read; clearing the name unsets it and the route falls back to its id, which is what the field's placeholder shows, while the protocol has no fallback to clear to. A catalog route gets neither: it defaults its name from its catalog entry, and each of its models carries its own protocol, so a route-level one could only override every one of them. + +The **Provider ID** is the one create-card field that stays fixed, and not for want of a control. It is the `providers.` dict key, so changing it is a move rather than an edit, and the editor is addressed by the `settingsPath` that move would invalidate. It is referenced from outside this namespace — `agent-default-model` stores a `provider` string, and every `request/header` in every session log already records one — so a rename would silently strip meaning from referents this page cannot see. And it is the stem of the derived credential reference: the page writes keys but can never read one back, so it cannot move `OLD_API_KEY` to `NEW_API_KEY`, leaving a rename to either orphan the stored key or point the profile at a reference under the previous name. Declaring the new route and deleting the old one does all three explicitly, and the page already offers both halves. + ## Alternatives considered **Declare a provider through `ProviderEditor` with extra fields.** One card instead of two, but the editor is addressed by `settingsPath`, and a route being named has no path yet. Recomputing the path per keystroke would remount the card and discard the draft; deferring it would mean the editor's whole write path no longer described what it was editing. **Add a wire field for the protocol list.** Explicit. But the settings schema already crosses the wire and already contains the union, so a second copy could disagree with the first — and the one the adapter enforces is the schema. +**Let the Provider ID be edited, with the page performing the move.** The card would unset the old key and set the new profile in one `settings.mutate`, and the rest is a rename. But the credential cannot travel with it — the page holds a redacted descriptor, never a value — and the referents in other namespaces and in logged sessions have no rename path at all, so the honest version of this feature is the create-then-delete the page already has. + +**Offer the protocol on every pi-ai route, with an inherit choice.** Symmetric with the base URL beside it, and repointing a catalog route at a gateway speaking another wire protocol is a real thing to want. But no consumer asks for it, one wrong pick silently repoints every model on the route, and the inherit choice would be the only way to write a declared route into a profile the adapter refuses. `settings.yaml` still expresses the repoint for a deployment that means it. + **Fetch against the stored profile instead of the live form.** No key would leave the form for an unsaved provider. But the flow that needs fetching most is the one where nothing is stored yet, and a form whose endpoint was edited would quietly interrogate the old one. **Write adopted candidates straight into the list.** Fewer clicks, but a fetch would then overwrite capacities the user had corrected, and a listing that discloses only ids would replace real numbers with nothing. ## Consequences -A gateway, a self-hosted server, or a model newer than the installed catalog is now configurable without leaving the browser, and the endpoint itself supplies the model ids where it can. The page grew two components and one shared list editor; the editor card's pi-ai fold grew from two fields to a list. +A gateway, a self-hosted server, or a model newer than the installed catalog is now configurable without leaving the browser, and the endpoint itself supplies the model ids where it can. The page grew two components and one shared list editor; the editor card's pi-ai fold grew from two fields to a list, plus a name and a protocol on a declared route. What it costs: only pi-ai routes can be hand-declared, because `llm-pi-ai` is the one namespace whose profiles describe a whole provider — a `llm-deepseek` route stays a composition fact. Interrogation reaches only OpenAI-compatible endpoints, so a gateway speaking another protocol reports that it cannot be asked and its models are typed in. And the page now holds a key in component state for the duration of a fetch, which is the same exposure `credentials.set` already has and no longer than the card lives. ## Testing -`packages/client/ui-models/tests/provider-form.spec.tsx` drives the rendered page over a scripted wire face: adding, editing, and removing rows; a cleared optional field leaving the profile and a non-integer capacity never entering it; the interrogation carrying the edited endpoint, the unsaved key, and the profile's protocol; the picker's default selection, toggling, cancel, and adopt-keeps-tuned-rows; the empty, refused, and rejected-transport paths; the create writing one profile plus its credential; every gate on the create button; and the read-only posture. `protocolChoices` is covered against a schema that declares the union and one that does not. +`packages/client/ui-models/tests/provider-form.spec.tsx` drives the rendered page over a scripted wire face: adding, editing, and removing rows; a cleared optional field leaving the profile and a non-integer capacity never entering it; the interrogation carrying the edited endpoint, the unsaved key, and the profile's protocol; the picker's default selection, toggling, cancel, and adopt-keeps-tuned-rows; the empty, refused, and rejected-transport paths; the create writing one profile plus its credential; every gate on the create button; and the read-only posture. `protocolChoices` is covered against a schema that declares the union and one that does not. The editor's own field inventory is asserted per route kind — a catalog route stops at the key and the endpoint, a declared one also carries the protocol — along with the protocol edit travelling as a single `api` path op, a rename travelling as a single `displayName` one, a cleared name unsetting rather than storing the empty string the adapter refuses, and a declared profile naming no protocol selecting nothing rather than the first choice. `apps/web/tests/models-settings.e2e.ts` reopens the declared route through the real wire, captures the card, and asserts the chosen protocol and the new name both reach `settings.yaml` and the row re-registers under the rename. diff --git a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.zh.md b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.zh.md index 974c4fa5be..c38b0387b3 100644 --- a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.zh.md +++ b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.zh.md @@ -22,22 +22,30 @@ Status: implemented 协议选项来自该 namespace **自己的 schema**，经页面本就会获取的 settings 描述符读出（`providers.*.api` 是适配器 `supportedProtocols()` 的一个 union）。没有新增协议字段，客户端里没有常量，提供的选项也无从与被接受的集合发生漂移。 +对于目录报告为**已声明**的路由，编辑器够得着它为自己命名的那两个字段——显示名称与该协议。创建卡片索要一个编辑器改不了的字段，等于把该字段留在只有 `settings.yaml` 才能触及的位置，而那正是本记录要终结的姿态。两者都渲染在折叠区里、紧挨着端点，协议读的是同一份 schema；清空名称即取消设置，路由退回自己的 id——字段占位符显示的就是它，而协议没有可退回的兜底。内置目录路由两个都不给：它的名称由目录条目兜底，它的每个模型各自带着自己的协议，路由级协议只可能把它们全部覆盖掉。 + +**Provider ID** 是创建卡片上唯一保持固定的字段，原因不是没做控件。它是 `providers.` 这个字典键，因此改它是一次搬移而非一次编辑，而编辑器正是由那次搬移会作废的 `settingsPath` 寻址的。它还被本 namespace 之外引用——`agent-default-model` 存着一个 `provider` 字符串，每条会话日志里的每个 `request/header` 也都已经记下了一个——因此重命名会悄悄抽空这个页面看不见的那些引用。它同时是派生凭据引用的词干：页面写得了密钥却永远读不回来，因此无法把 `OLD_API_KEY` 搬到 `NEW_API_KEY`，重命名要么让已存密钥成为孤儿，要么让 profile 指向一个仍带旧名的引用。声明新路由再删掉旧的，把这三件事都显式做了一遍，而页面本就提供这两半。 + ## Alternatives considered **在 `ProviderEditor` 上加字段来声明提供方。** 两张卡片变一张，但编辑器由 `settingsPath` 寻址，而正在被命名的路由还没有路径。逐次按键重算路径会让卡片重新挂载并丢掉草稿；推迟计算则意味着编辑器的整条写入路径不再描述它正在编辑的东西。 **为协议列表新增一个协议字段。** 显式。但 settings schema 本来就会跨越协议层、本来就含有那个 union，因此第二份副本可能与第一份不一致——而适配器强制执行的是 schema 那一份。 +**开放 Provider ID 编辑，由页面来完成这次搬移。** 卡片可以在一次 `settings.mutate` 里取消旧键、设置新 profile，剩下的就只是改名。但凭据没法跟着走——页面手里只有脱敏描述符，从来没有值——而其他 namespace 与已记录会话里的引用根本没有重命名通路，因此这个功能诚实的版本，就是页面已经具备的「先建后删」。 + +**给每条 pi-ai 路由都提供协议，并附一个「继承」选项。** 与紧挨着的 API 地址对称，而且把内置目录路由指向讲另一种协议的网关确实是有人会想要的事。但目前没有消费方提出这个诉求，一次选错就会静默地把该路由上每个模型都重指，而「继承」选项还会成为把已声明路由写成适配器拒绝的 profile 的唯一途径。真要这么做的部署，`settings.yaml` 仍然表达得了。 + **针对已存 profile 而非实时表单发起获取。** 对尚未保存的提供方来说，密钥就不会离开表单。但最需要获取的恰恰是「什么都还没存」的那条流程，而端点已修改的表单会悄悄去询问旧地址。 **把采纳的候选直接写进列表。** 点击更少，但一次获取就会覆盖用户已更正的容量，而只公布 id 的列表会把真实数字替换成空。 ## Consequences -网关、自建服务，或比已安装 catalog 更新的模型，如今无需离开浏览器就能配置，而模型 id 在端点能提供时由端点自己给出。页面多了两个组件和一个共用的列表编辑器；编辑卡片的 pi-ai 折叠区从两个字段长成了一个列表。 +网关、自建服务，或比已安装 catalog 更新的模型，如今无需离开浏览器就能配置，而模型 id 在端点能提供时由端点自己给出。页面多了两个组件和一个共用的列表编辑器；编辑卡片的 pi-ai 折叠区从两个字段长成了一个列表，已声明路由上还多了一个名称输入框和一个协议选择框。 代价是：只有 pi-ai 路由可以手工声明，因为 `llm-pi-ai` 是唯一一个其 profile 描述整个提供方的 namespace——`llm-deepseek` 路由仍是组合面的事实。询问只覆盖 OpenAI 兼容端点，因此讲其他协议的网关会报告自己无法被询问，其模型需手工键入。另外，页面在一次获取期间会把密钥保存在组件状态里，这与 `credentials.set` 已有的暴露面相同，且不长于卡片的存活时间。 ## Testing -`packages/client/ui-models/tests/provider-form.spec.tsx` 在脚本化的协议面之上驱动渲染后的页面：添加、编辑与移除行；被清空的可选字段离开 profile、非整数容量从不进入；询问携带已修改的端点、未保存的密钥，以及 profile 自身的协议；选择框的默认选中、勾选切换、取消，以及「采纳保留已调优的行」；空列表、被拒、传输被拒三条路径；创建写入一份 profile 加其凭据；创建按钮上的每一道门控；以及只读姿态。`protocolChoices` 针对「声明了该 union」与「没有声明」两种 schema 都有覆盖。 +`packages/client/ui-models/tests/provider-form.spec.tsx` 在脚本化的协议面之上驱动渲染后的页面：添加、编辑与移除行；被清空的可选字段离开 profile、非整数容量从不进入；询问携带已修改的端点、未保存的密钥，以及 profile 自身的协议；选择框的默认选中、勾选切换、取消，以及「采纳保留已调优的行」；空列表、被拒、传输被拒三条路径；创建写入一份 profile 加其凭据；创建按钮上的每一道门控；以及只读姿态。`protocolChoices` 针对「声明了该 union」与「没有声明」两种 schema 都有覆盖。编辑器自身的字段清单按路由种类各有断言——内置目录路由止于密钥与端点，已声明路由还带着协议——同时覆盖协议改动只以单条 `api` path op 传出、改名只以单条 `displayName` path op 传出、清空名称是取消设置而不是存入适配器会拒绝的空串，以及不写协议的已声明 profile 什么都不选中、而非选中第一个候选。`apps/web/tests/models-settings.e2e.ts` 经真实协议层重新打开这条已声明路由，捕获该卡片，并断言选定的协议与新名称都抵达了 `settings.yaml`、该行也以新名重新注册。 diff --git a/apps/web/tests/models-settings.e2e.ts b/apps/web/tests/models-settings.e2e.ts index 364b00f9e5..6afdfa4429 100644 --- a/apps/web/tests/models-settings.e2e.ts +++ b/apps/web/tests/models-settings.e2e.ts @@ -28,6 +28,7 @@ const SNAPSHOT_DIR = fileURLToPath(new URL('./snapshots/models-settings', import const EMPTY_EXPECTED = join(SNAPSHOT_DIR, 'empty.expected.md') const CONFIGURED_EXPECTED = join(SNAPSHOT_DIR, 'configured.expected.md') const DECLARED_EXPECTED = join(SNAPSHOT_DIR, 'declared.expected.md') +const DECLARED_EDIT_EXPECTED = join(SNAPSHOT_DIR, 'declared-edit.expected.md') const NATIVE_DELETE_EXPECTED = join(SNAPSHOT_DIR, 'native-delete.expected.md') const DELETE_EXPECTED = join(SNAPSHOT_DIR, 'delete.expected.md') const MODE = webSnapshotMode() @@ -209,6 +210,37 @@ describe('web e2e: Models settings page configures a dormant provider', () => { expect(tripwire.pageErrors).toEqual([]) }, 60_000) + it('reopens the name and protocol a declared route was created with', async () => { + onTestFailed(() => saveFailureShot(page, 'web-e2e-models-declared-identity')) + const dialog = page.getByRole('dialog', { name: '设置' }) + await dialog.getByRole('button', { name: '编辑 Acme Gateway (acme-gateway)' }).click() + await dialog.getByText('自定义设置').click() + // The create card asked this route for a name and a protocol because + // nothing can default them; the editor reaches the same two fields rather + // than sending the user to settings.yaml for what only this route names. + const protocol = dialog.getByLabel('API 协议') + await protocol.waitFor({ timeout: 10_000 }) + expect(await protocol.inputValue()).toBe('openai-completions') + const name = dialog.getByLabel('显示名称', { exact: true }) + expect(await name.inputValue()).toBe('Acme Gateway') + const snapshot = await captureStableAria(page, '[role="dialog"]', scaffold.workspaceCwd) + await compareOrRefreshGolden(DECLARED_EDIT_EXPECTED, snapshot, MODE) + + await protocol.selectOption('anthropic-messages') + await name.fill('Acme 网关') + await dialog.getByRole('button', { name: '保存', exact: true }).click() + await expect.poll(async () => dialog.getByLabel('API 协议').count(), { timeout: 10_000 }).toBe(0) + // The adapter re-resolved the route under the new protocol and re-registered + // it under the new name: an unserviceable profile would have been refused + // at the write instead, and a rename that did not re-register would leave + // the old label on the row. + await dialog.getByText('Acme 网关', { exact: true }).first().waitFor({ timeout: 10_000 }) + const document = await readFile(join(scaffold.harnessHome, 'settings.yaml'), 'utf8') + expect(document).toContain('api: anthropic-messages') + expect(document).toContain('displayName: Acme 网关') + expect(tripwire.pageErrors).toEqual([]) + }, 60_000) + it('confirms an identified provider deletion before removing its profile and key', async () => { onTestFailed(() => saveFailureShot(page, 'web-e2e-models-delete')) const settingsDialog = page.getByRole('dialog', { name: '设置' }) @@ -243,8 +275,8 @@ describe('web e2e: Models settings page configures a dormant provider', () => { it.skipIf(MODE === 'record')('keeps the fixture inventory closed', async () => { await assertFixtureInventory(SNAPSHOT_DIR, [ - 'configured.expected.md', 'declared.expected.md', 'delete.expected.md', - 'empty.expected.md', 'native-delete.expected.md', + 'configured.expected.md', 'declared-edit.expected.md', 'declared.expected.md', + 'delete.expected.md', 'empty.expected.md', 'native-delete.expected.md', ]) }) }) diff --git a/apps/web/tests/snapshots/models-settings/declared-edit.expected.md b/apps/web/tests/snapshots/models-settings/declared-edit.expected.md new file mode 100644 index 0000000000..e36c7ff2e8 --- /dev/null +++ b/apps/web/tests/snapshots/models-settings/declared-edit.expected.md @@ -0,0 +1,65 @@ +- dialog "设置": + - navigation: + - text: 设置 + - button "通用设置": + - img + - text: 通用设置 + - button "模型": + - img + - text: 模型 + - button "Agent 预设": + - img + - text: Agent 预设 + - button "打开配置文件" + - button "关闭": + - img + - text: 关闭 + - heading "模型" [level=2] + - paragraph: 填入各提供方的 API 密钥即可使用其模型。 + - list: + - listitem: + - text: minimax-cn + - img "API 密钥已配置" + - button "编辑 minimax-cn": 编辑 + - button "删除 minimax-cn": 删除 + - listitem: + - text: Acme Gateway 自定义 + - button "编辑 Acme Gateway (acme-gateway)": 编辑 + - button "删除 Acme Gateway (acme-gateway)": 删除 + - text: Acme Gateway acme-gateway API 密钥 + - textbox "API 密钥": + - /placeholder: 输入 API 密钥，或留空使用环境认证 + - group: + - text: 自定义设置 显示名称 + - textbox "显示名称": + - /placeholder: acme-gateway + - text: Acme Gateway + - text: API 地址 + - textbox "API 地址": + - /placeholder: https://gateway.acme.example/v1 + - text: https://gateway.acme.example/v1 + - text: API 协议 + - combobox "API 协议": + - option "openai-completions" [selected] + - option "openai-responses" + - option "anthropic-messages" + - region "模型目录": + - text: 模型目录 已自定义模型目录 + - button "恢复默认模型" + - button "获取可用模型" + - textbox "模型 ID 1": + - /placeholder: 模型 ID + - text: acme-large + - textbox "显示名称 1": + - /placeholder: 显示名称 + - button "容量 1" + - button "删除模型 1" + - button "添加模型" + - button "取消" + - button "保存" + - button "添加提供方": + - img + - text: 添加提供方 + - button "添加自定义提供方": + - img + - text: 添加自定义提供方 diff --git a/docs/user/guide/providers.i18n.yaml b/docs/user/guide/providers.i18n.yaml index c95a0db6c0..e482a07615 100644 --- a/docs/user/guide/providers.i18n.yaml +++ b/docs/user/guide/providers.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write docs/user/guide/providers.md -providers.md: 4667e54161e77a62d454f3f78a8164d5c79b78c8 -providers.zh.md: 15e0ccc826a5c285c71d4776793f8048c11f6254 +providers.md: fff2e9f66b1dcec2b549420ee43f908043983338 +providers.zh.md: a27adefdbb76c43d099982b0503a9adde139597f diff --git a/docs/user/guide/providers.md b/docs/user/guide/providers.md index 4667e54161..fff2e9f66b 100644 --- a/docs/user/guide/providers.md +++ b/docs/user/guide/providers.md @@ -29,6 +29,8 @@ That holds for providers that authenticate with an API key. The catalog also car ![The custom provider form: Provider ID, display name, base URL, API protocol, and API key](providers-custom-form.png) +Every field but the Provider ID stays editable afterwards: **Edit** on the row reopens the same fields, with the display name and the protocol under 自定义设置 beside the base URL. Clearing the display name falls back to the Provider ID. The Provider ID itself is fixed: it names the route in requests, in `agent-default-model`, and in every session already logged, and it is the stem of the credential reference the page can never read back — so renaming a route means declaring a new provider and deleting the old one. + **Let the endpoint report its models.** Expand **Model catalog** and choose **Fetch available models**: the interrogation asks the endpoint **the form currently shows** — including a base URL edited but not yet saved and a key typed but not yet stored — and offers what it reports as candidates to pick from. A route the installed catalog describes is answered from that catalog with no network call. Adopting a candidate only writes rows into the draft; nothing is stored until you save. Keys are write-only: the page only ever holds a redacted descriptor, never the literal secret. A key you enter is stored in `$DSH_HOME/.credentials.yaml`, and the profile records only the variable name that references it. diff --git a/docs/user/guide/providers.zh.md b/docs/user/guide/providers.zh.md index 15e0ccc826..a27adefdbb 100644 --- a/docs/user/guide/providers.zh.md +++ b/docs/user/guide/providers.zh.md @@ -29,6 +29,8 @@ Harness 出厂自带 DeepSeek，同时预装了一个通用的多提供方适配 ![自定义提供方表单：Provider ID、显示名称、API 地址、API 协议、API 密钥](providers-custom-form.zh.png) +除 Provider ID 外的每个字段之后都还能改：行上的**编辑**会重新打开这些字段，显示名称和协议在「自定义设置」里、紧挨着 API 地址；显示名称清空即退回 Provider ID。Provider ID 本身固定不可改：它在请求里、在 `agent-default-model` 里、在每一条已记录的会话里点名这条路由，同时还是凭据引用的词干，而页面永远读不回凭据值——因此重命名一条路由等于声明一个新提供方再把旧的删掉。 + **让端点自己报模型。** 展开**模型目录**后点**获取可用模型**，会按你**当前表单里**的地址与密钥去问端点（地址改了但没保存、密钥刚输入还没存下，都算数），把它报告的模型列成候选让你勾选。内置目录里的路由直接由目录作答，不联网。采纳只是把行写进草稿，最终还是你点保存才落盘。 密钥是只写的：页面拿到的永远是脱敏描述符，不是明文。写入的密钥存进 `$DSH_HOME/.credentials.yaml`，profile 里只记录引用它的变量名。 diff --git a/packages/client/ui-models/README.i18n.yaml b/packages/client/ui-models/README.i18n.yaml index c03e091a5e..bb7212c7ff 100644 --- a/packages/client/ui-models/README.i18n.yaml +++ b/packages/client/ui-models/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/client/ui-models/README.md -README.md: e0c5728d47e053df1934ef9eb69df3f8d985a4ec -README.zh.md: fe11e6cdd190e19d5b5dac6dc95950ba59a3172b +README.md: f86564151a3c3788c3002421cac465ef98177281 +README.zh.md: f94438db18a2034b4180f5b79ea521ad7037754c diff --git a/packages/client/ui-models/README.md b/packages/client/ui-models/README.md index e0c5728d47..f86564151a 100644 --- a/packages/client/ui-models/README.md +++ b/packages/client/ui-models/README.md @@ -4,7 +4,7 @@ English | [中文](README.zh.md) Models settings plugin: the provider configuration page and official-DeepSeek conditional onboarding step. It joins three wire domains into one shared snapshot — `llm.providers` (the configurable-provider directory with each route's live/dormant state), `settings.describe` (serialized schemas, layered redacted values, secret slots), and `credentials.describe` (value-free configured/source/writable badges) — and renders provider rows with one editor card at a time, without presenting route liveness as provider status. -Rows are the *configured* providers (their profile resolves in the owning namespace); a whole-section provider whose key is not configured anywhere (the first-run DeepSeek posture) renders as its open setup card instead of a row, and the add flow is a card carrying the dormant-directory provider select — a bare-mounted `llm-pi-ai` offers its whole installed catalog before any route exists. The pi-ai card additionally edits that route's **model list** and can ask the provider what it serves. A row labels API-key state with a green solid dot only when a referenced credential is confirmed configured, and with a red solid dot only when a named reference is confirmed missing; reference-free provider-native authentication and unavailable credential enrichment remain unmarked. The editor is a hand-written card per adapter family: the primary field is a single **API key** input — the page never asks for an environment-variable name; a typed key stores **write-only** through `credentials.set` under the profile's reference, deriving `_API_KEY` when the profile has none, and the pi-ai profile records that derivation as `apiKeyEnv`, so `settings.yaml` never carries a key value. Leaving a new pi-ai provider's key blank saves a reference-free profile and therefore preserves provider-native authentication such as the Bedrock credential chain or Vertex ADC. A successful Apply emits a local accessible status message without echoing secret material. The collapsed 自定义设置 fold carries the curated extras — `baseURL` for both families (the deepseek placeholder shows the public endpoint) and each adapter's model catalog. Reasoning effort is deliberately NOT among them: it is a per-model capability and the models under one provider disagree about which levels they accept, so a provider-scoped control could only be set to a value some of them reject — which would hide even the models that support the level. The composer's model picker offers each model its own levels, and a switch there records provider, model, and effort together as the default for the next session. The profile field stays in `settings.yaml` for a deployment that knows its route. Each DeepSeek row edits `id`, optional display `name`, and optional `contextWindow`; existing fields outside that curated set survive edits, while every other profile field stays owned by `settings.yaml`. A row is deletable only when the user layer alone carries it (removal restores the composition base), and its localized confirmation dialog names the provider in the title, description, and final action. A row is tagged **Custom** when the directory entry says the owning adapter ships nothing under that key. The tag follows that answer alone: having a stored profile does not make a route custom — narrowing a shipped provider's models stores one too — and an adapter that reports nothing leaves its rows untagged rather than being read as shipped. +Rows are the *configured* providers (their profile resolves in the owning namespace); a whole-section provider whose key is not configured anywhere (the first-run DeepSeek posture) renders as its open setup card instead of a row, and the add flow is a card carrying the dormant-directory provider select — a bare-mounted `llm-pi-ai` offers its whole installed catalog before any route exists. The pi-ai card additionally edits that route's **model list** and can ask the provider what it serves. A row labels API-key state with a green solid dot only when a referenced credential is confirmed configured, and with a red solid dot only when a named reference is confirmed missing; reference-free provider-native authentication and unavailable credential enrichment remain unmarked. The editor is a hand-written card per adapter family: the primary field is a single **API key** input — the page never asks for an environment-variable name; a typed key stores **write-only** through `credentials.set` under the profile's reference, deriving `_API_KEY` when the profile has none, and the pi-ai profile records that derivation as `apiKeyEnv`, so `settings.yaml` never carries a key value. Leaving a new pi-ai provider's key blank saves a reference-free profile and therefore preserves provider-native authentication such as the Bedrock credential chain or Vertex ADC. A successful Apply emits a local accessible status message without echoing secret material. The collapsed 自定义设置 fold carries the curated extras — `baseURL` for both families (the deepseek placeholder shows the public endpoint), each adapter's model catalog, and the **display name** and **API protocol** of a pi-ai route the adapter does not ship. Those two are what a hand-declared route names for itself: the create card asks for both because nothing can default them, so the editor reaches both rather than leaving them to `settings.yaml`. Clearing the name unsets it and the route falls back to its id, which is what the placeholder shows; the protocol has no such fallback. A catalog route gets neither — it defaults its name from its catalog entry, and its models each carry their own protocol, so a route-level one could only override every one of them. The Provider ID stays fixed: it is the settings key, the name every other namespace and every logged session references, and the stem of a credential reference the page cannot read back to move. Reasoning effort is deliberately NOT among them: it is a per-model capability and the models under one provider disagree about which levels they accept, so a provider-scoped control could only be set to a value some of them reject — which would hide even the models that support the level. The composer's model picker offers each model its own levels, and a switch there records provider, model, and effort together as the default for the next session. The profile field stays in `settings.yaml` for a deployment that knows its route. Each DeepSeek row edits `id`, optional display `name`, and optional `contextWindow`; existing fields outside that curated set survive edits, while every other profile field stays owned by `settings.yaml`. A row is deletable only when the user layer alone carries it (removal restores the composition base), and its localized confirmation dialog names the provider in the title, description, and final action. A row is tagged **Custom** when the directory entry says the owning adapter ships nothing under that key. The tag follows that answer alone: having a stored profile does not make a route custom — narrowing a shipped provider's models stores one too — and an adapter that reports nothing leaves its rows untagged rather than being read as shipped. The DeepSeek step projects `deepseek-official` readiness from that same joined snapshot after earlier onboarding pages complete. It recognizes the official adapter through its `llm-deepseek` configurable-provider declaration, so an undeclared live route with the same provider id is not treated as repairable configuration. A configured credential reference completes the step without rendering, including a read-only launch-environment credential. Only a mounted, active adapter with a missing writable reference shows the page that opens Settings on Models, whose existing setup card exclusively owns key input and `credentials.set`; the step never holds a secret. An absent adapter, inactive route, failed join, read-only deployment, or unusable settings or credential capability completes the step without rendering so onboarding cannot block the product; Models remains the diagnostic surface. @@ -28,7 +28,7 @@ None; this package neither assembles nor sends a provider request. ## Known Limitations and Deferred Work -- **Only the API key and curated fold fields are editable on the card** — the hand-written editor traded schema-generic field coverage for the mockup layout ([Agent Note](../../../.agents/notes/implemented/architecture/2026-07-30-web-config-plane.md)). DeepSeek exposes `baseURL`, `reasoningEffort`, and model `id`/`name`/`contextWindow`/`maxTokens`; pi-ai exposes `baseURL` and `reasoning`. Retry policy, timeouts, DeepSeek model descriptions, and other advanced fields remain in `settings.yaml`; existing model fields the editor does not show are preserved. A profile schema without the conventional fields renders the hint alone, and the two curated layouts key on the `llm-deepseek`/`llm-pi-ai` namespaces by name. +- **Only the API key and curated fold fields are editable on the card** — the hand-written editor traded schema-generic field coverage for the mockup layout ([Agent Note](../../../.agents/notes/implemented/architecture/2026-07-30-web-config-plane.md)). Both families expose `baseURL` and model `id`/`name`/`contextWindow`/`maxTokens`; a hand-declared pi-ai route also exposes `displayName` and `api`. Retry policy, timeouts, DeepSeek model descriptions, and other advanced fields remain in `settings.yaml`; existing model fields the editor does not show are preserved. A profile schema without the conventional fields renders the hint alone, and the two curated layouts key on the `llm-deepseek`/`llm-pi-ai` namespaces by name. - **Credential cleanup is intentionally narrow** — deleting a row removes the configured, writable credential only when its reference is the exact `_API_KEY` target this page derives. Custom references, environment credentials, and unidentifiable targets are retained because the row cannot prove ownership of them. - **Only pi-ai routes can be hand-declared** — the custom-provider card writes into `llm-pi-ai`, the one namespace whose profiles describe a whole provider. A `llm-deepseek` route is a composition fact, not something this page can create. - **Interrogation covers OpenAI-compatible endpoints** — the adapter reads only that model-list response format, so a gateway speaking another protocol reports that it cannot be asked and its models are entered by hand. diff --git a/packages/client/ui-models/README.zh.md b/packages/client/ui-models/README.zh.md index fe11e6cdd1..f94438db18 100644 --- a/packages/client/ui-models/README.zh.md +++ b/packages/client/ui-models/README.zh.md @@ -4,7 +4,7 @@ 模型设置插件：提供方配置页和按条件显示的 DeepSeek 官方首次使用引导步骤。它把三个协议领域汇聚为一个共享快照：`llm.providers`（可配置提供方目录，含每条路由的存活／休眠状态）、`settings.describe`（序列化 schema、分层脱敏值、secret 槽位）与 `credentials.describe`（不含值的 configured/source/writable 徽标）；页面据此渲染提供方行，一次只展开一张编辑卡片，且不把路由存活状态呈现为提供方状态。 -行是*已配置*的提供方（其 profile 在所属 namespace 中解析得出）；密钥未在任何地方配置的整分节提供方（DeepSeek 的首次运行姿态）会渲染为其展开的设置卡片而非一行，「新增」流程则是一张承载休眠目录提供方选择框的卡片——裸挂载的 `llm-pi-ai` 在任何路由存在之前就能提供其完整的已安装 catalog。pi-ai 卡片还会编辑该路由的**模型列表**，并可查询提供方所提供的模型。只有确认引用的凭据已配置时，行才会以绿色实心点标示 API 密钥状态；只有确认具名引用缺失时，才会以红色实心点标示。无引用的提供方原生认证以及无法取得凭据补充信息时都不显示状态点。编辑器是每个适配器家族各一张的手写卡片：主字段是单独一个 **API 密钥**输入框——页面从不询问环境变量名；键入的密钥经 `credentials.set` 以**只写**方式存入 profile 的引用之下，profile 没有引用时便派生 `_API_KEY`，pi-ai profile 会把这次派生记录为 `apiKeyEnv`，因此 `settings.yaml` 从不携带密钥值。为新的 pi-ai 提供方留空密钥会保存一个不带引用的 profile，因此能保留提供方原生认证，例如 Bedrock 凭据链或 Vertex ADC。「应用」成功后会发出本地无障碍状态消息，且绝不回显任何机密内容。收起的「自定义设置」折叠区承载精选的额外字段——两个家族都有 `baseURL`（deepseek 的占位符显示公共端点），以及各适配器自己的模型目录。推理等级刻意**不在**其中：它是按模型的能力，而同一提供方下各模型接受的档位并不一致，因此提供方级的控件只可能被设成其中一些模型会拒绝的值——那会连支持该档位的模型也一并隐藏。输入框的模型选择器为每个模型提供它自己的档位，在那里切换会把提供方、模型、推理等级一并记为下一个会话的默认值。profile 字段仍留在 `settings.yaml`，供清楚自己路由的部署使用。每条 DeepSeek 模型行可编辑 `id`、可选的显示名称 `name` 与可选的 `contextWindow`；精选集合以外的现有字段会在编辑后保留，其余每个 profile 字段仍归 `settings.yaml` 所有。只有当某行仅由用户层承载时它才可删除（删除会还原组合 base），其本地化确认对话框会在标题、说明和最终操作中点名该提供方。当目录条目表明拥有该路由的适配器在这个键下什么都没有时，该行会带上 **自定义** 标签。标签只跟随这个答案：存了 profile 并不使一条路由成为自定义——收窄一个内置提供方的模型同样会存下 profile——而什么都不回答的适配器，其路由保持无标签，不会被当成内置。 +行是*已配置*的提供方（其 profile 在所属 namespace 中解析得出）；密钥未在任何地方配置的整分节提供方（DeepSeek 的首次运行姿态）会渲染为其展开的设置卡片而非一行，「新增」流程则是一张承载休眠目录提供方选择框的卡片——裸挂载的 `llm-pi-ai` 在任何路由存在之前就能提供其完整的已安装 catalog。pi-ai 卡片还会编辑该路由的**模型列表**，并可查询提供方所提供的模型。只有确认引用的凭据已配置时，行才会以绿色实心点标示 API 密钥状态；只有确认具名引用缺失时，才会以红色实心点标示。无引用的提供方原生认证以及无法取得凭据补充信息时都不显示状态点。编辑器是每个适配器家族各一张的手写卡片：主字段是单独一个 **API 密钥**输入框——页面从不询问环境变量名；键入的密钥经 `credentials.set` 以**只写**方式存入 profile 的引用之下，profile 没有引用时便派生 `_API_KEY`，pi-ai profile 会把这次派生记录为 `apiKeyEnv`，因此 `settings.yaml` 从不携带密钥值。为新的 pi-ai 提供方留空密钥会保存一个不带引用的 profile，因此能保留提供方原生认证，例如 Bedrock 凭据链或 Vertex ADC。「应用」成功后会发出本地无障碍状态消息，且绝不回显任何机密内容。收起的「自定义设置」折叠区承载精选的额外字段——两个家族都有 `baseURL`（deepseek 的占位符显示公共端点）、各适配器自己的模型目录，以及适配器未提供的那类 pi-ai 路由的**显示名称**与 **API 协议**。这两个字段是手工声明路由为自己命名的东西：创建卡片之所以索要它们，正因为没有东西能为它们兜底，因此编辑器也够得着这两个，而不是把它们留给 `settings.yaml`。清空名称即取消设置，路由退回自己的 id——占位符显示的就是它；协议没有这样的兜底。内置目录路由两个都不给：它的名称由目录条目兜底，它的每个模型各自带着自己的协议，路由级协议只可能把它们全部覆盖掉。Provider ID 保持固定：它是 settings 的键、是其他每个 namespace 与每一条已记录会话引用的名字，也是页面读不回、因而搬不走的凭据引用词干。推理等级刻意**不在**其中：它是按模型的能力，而同一提供方下各模型接受的档位并不一致，因此提供方级的控件只可能被设成其中一些模型会拒绝的值——那会连支持该档位的模型也一并隐藏。输入框的模型选择器为每个模型提供它自己的档位，在那里切换会把提供方、模型、推理等级一并记为下一个会话的默认值。profile 字段仍留在 `settings.yaml`，供清楚自己路由的部署使用。每条 DeepSeek 模型行可编辑 `id`、可选的显示名称 `name` 与可选的 `contextWindow`；精选集合以外的现有字段会在编辑后保留，其余每个 profile 字段仍归 `settings.yaml` 所有。只有当某行仅由用户层承载时它才可删除（删除会还原组合 base），其本地化确认对话框会在标题、说明和最终操作中点名该提供方。当目录条目表明拥有该路由的适配器在这个键下什么都没有时，该行会带上 **自定义** 标签。标签只跟随这个答案：存了 profile 并不使一条路由成为自定义——收窄一个内置提供方的模型同样会存下 profile——而什么都不回答的适配器，其路由保持无标签，不会被当成内置。 前序首次使用引导页面完成后，DeepSeek 步骤会从同一个联接快照得出 `deepseek-official` 的就绪状态。它通过 `llm-deepseek` 的可配置提供方声明识别官方适配器，因此同 id 但未声明的存活路由不属于可修复配置。凭据引用已配置时，该步骤会直接完成而不渲染，其中包括来自启动环境且只读的凭据。只有已挂载且活跃、引用可写但尚未配置的适配器才会显示前往「设置」Models 分区的页面；密钥输入和 `credentials.set` 仅由该分区已有的设置卡片负责，该步骤绝不持有 secret。适配器缺失、路由不活跃、联接失败、部署只读或设置／凭据能力不可用时，该步骤均不渲染并直接完成，以免首次使用引导阻塞产品；Models 页仍是诊断界面。 @@ -28,7 +28,7 @@ pi-ai profile 的 `models` 列表就在卡片上编辑：一行一个模型， ## 已知限制与暂缓事项 -- **卡片上可编辑的只有 API 密钥与精选折叠区字段**：手写编辑器用 schema 通用的字段覆盖面换来了设计稿上的布局（[Agent Note](../../../.agents/notes/implemented/architecture/2026-07-30-web-config-plane.md)）。DeepSeek 公开 `baseURL`、`reasoningEffort` 与模型的 `id`/`name`/`contextWindow`/`maxTokens`；pi-ai 公开 `baseURL` 与 `reasoning`。重试策略、超时、DeepSeek 模型说明及其他进阶字段仍留在 `settings.yaml` 中；编辑器未展示的现有模型字段会予以保留。不带这些约定字段的 profile schema 只渲染该提示，两套精选布局则以 `llm-deepseek`/`llm-pi-ai` 这两个 namespace 的名字为键。 +- **卡片上可编辑的只有 API 密钥与精选折叠区字段**：手写编辑器用 schema 通用的字段覆盖面换来了设计稿上的布局（[Agent Note](../../../.agents/notes/implemented/architecture/2026-07-30-web-config-plane.md)）。两个家族都公开 `baseURL` 与模型的 `id`/`name`/`contextWindow`/`maxTokens`；手工声明的 pi-ai 路由还公开 `displayName` 与 `api`。重试策略、超时、DeepSeek 模型说明及其他进阶字段仍留在 `settings.yaml` 中；编辑器未展示的现有模型字段会予以保留。不带这些约定字段的 profile schema 只渲染该提示，两套精选布局则以 `llm-deepseek`/`llm-pi-ai` 这两个 namespace 的名字为键。 - **凭据清理范围刻意保持狭窄**：删除一行时，仅当其引用与页面派生的 `_API_KEY` 目标完全一致，才会清除已配置且可写的凭据。自定义引用、环境凭据和无法识别的目标会保留，因为该行无法证明自己拥有它们。 - **只有 pi-ai 路由可以手工声明**：自定义提供方卡片写入 `llm-pi-ai`——唯一一个其 profile 描述整个提供方的 namespace。`llm-deepseek` 路由是组合面的事实，不是本页能创建的东西。 - **询问只覆盖 OpenAI 兼容端点**：适配器只读这种模型列表响应格式，因此讲其他协议的网关会报告自己无法被询问，其模型需手工填写。 diff --git a/packages/client/ui-models/src/client/ModelsSection.tsx b/packages/client/ui-models/src/client/ModelsSection.tsx index 5551b847fd..be4532444c 100644 --- a/packages/client/ui-models/src/client/ModelsSection.tsx +++ b/packages/client/ui-models/src/client/ModelsSection.tsx @@ -54,6 +54,8 @@ interface EditorTarget extends ProviderIdentity { settingsPath: readonly string[] /** Writable credential identified under this page's conventional reference. */ credentialRef?: string + /** The adapter reports this route as one it does not ship (see {@link ProviderEditorProps.declared}). */ + declared?: boolean } /** Values that vary around the shared provider-editor rendering. */ @@ -71,6 +73,7 @@ function renderProviderEditor({ target, ...props }: ProviderEditorRenderProps): provider={target.provider} displayName={target.displayName} settingsPath={target.settingsPath} + {...target.declared === true ? { declared: true } : {}} {...props} /> ) @@ -135,6 +138,10 @@ function targetOf(row: ProviderRow): EditorTarget { settingsNs: row.entry.settingsNs, settingsPath: row.entry.settingsPath, ...credentialRef === undefined ? {} : { credentialRef }, + // Absent is not "shipped": an adapter that answers nothing leaves the + // route-level fields only a declared route owns off the card, exactly as + // it leaves the custom tag off the row. + ...row.entry.declared === true ? { declared: true } : {}, } } diff --git a/packages/client/ui-models/src/client/ProviderEditor.tsx b/packages/client/ui-models/src/client/ProviderEditor.tsx index 01978273ae..afae84a89f 100644 --- a/packages/client/ui-models/src/client/ProviderEditor.tsx +++ b/packages/client/ui-models/src/client/ProviderEditor.tsx @@ -7,7 +7,9 @@ * a key is entered; a blank key materializes a reference-free profile for * provider-native authentication); * the collapsed 自定义设置 area carries the per-family extras (`baseURL` for - * both families and DeepSeek's id/name/context-window model catalog). + * both families, DeepSeek's id/name/context-window model catalog, and the + * wire protocol of a pi-ai route the adapter does not ship — the field the + * create card asked that route for, editable here for the same reason). * Reasoning effort is deliberately absent: it is a per-MODEL capability, and * the models under one provider disagree about it, so a provider-scoped * control can only be set to a value some of them reject. The composer's @@ -30,7 +32,7 @@ import { import { apiKeyFailure } from './apiKey.ts' import { EditorFooter } from './EditorFooter.tsx' import { ModelListEditor } from './ModelListEditor.tsx' -import { deriveKeyRef, messageOf } from './store.ts' +import { deriveKeyRef, messageOf, protocolChoices } from './store.ts' import type { en } from './locales.ts' import styles from './ModelsSection.module.css' @@ -48,6 +50,14 @@ export interface ProviderEditorProps { displayName: string /** Hide the title row (the add card renders its own provider select). */ hideTitle?: boolean + /** + * Whether the adapter reports this route as hand-declared — absent from its + * installed catalog. Such a route carries its own wire protocol, chosen when + * it was created and editable here for the same reason; a catalog route's + * models each carry theirs, so a route-level protocol there could only + * override every one of them and the card does not offer it. + */ + declared?: boolean /** The owning namespace view (schema, layers, secrets). */ namespace: SettingsNamespaceView /** Path from the section root to this provider's profile. */ @@ -139,6 +149,9 @@ export function ProviderEditor(props: ProviderEditorProps): ReactNode { const disabled = props.readOnly || busy const layout = layoutOf(namespace.ns) const keyRef = refFor(namespace, settingsPath, props.provider) + // The same schema read the create card makes, so the choices offered here + // and there cannot drift apart: both come from the adapter's own `Config`. + const protocols = useMemo(() => protocolChoices(namespace), [namespace]) useEffect(() => { let stale = false @@ -289,11 +302,15 @@ export function ProviderEditor(props: ProviderEditorProps): ReactNode { } /** - * The curated fields of one known adapter family. Taking the narrowed - * family as a parameter is what makes `EFFORT_FIELD` total here: an - * unknown namespace never reaches this body. + * The curated fields of one known adapter family. The family arrives + * narrowed so the per-family branches below are total: an unknown namespace + * renders the hint instead and never reaches this body. */ const curatedFields = (family: 'deepseek' | 'pi-ai'): ReactNode => { + // What a hand-declared route names for itself and nothing else can supply. + // A whole-section `llm-deepseek` profile is a composition fact with no + // per-route identity for its schema to carry, hence the family test. + const ownsIdentity = family === 'pi-ai' && props.declared === true const customModels = getPath(draft, ['models']) const modelsOverridden = hasPath(draft, ['models']) const models = modelDrafts(modelsOverridden ? customModels : inheritedModels()) @@ -334,6 +351,28 @@ export function ProviderEditor(props: ProviderEditorProps): ReactNode {
{t('customized')}
+ {/* The name and the protocol are the create card's two remaining + profile fields; a route the adapter ships defaults both from + its catalog entry and neither belongs on its card. */} + {ownsIdentity + ? ( +
+ {t('customDisplayName')} + { setField('displayName', event.target.value) }} + /> +
+ ) + : null}
{t('baseUrl')}
+ {/* The protocol sits beside the endpoint it describes, as it does + on the create card. */} + {ownsIdentity + ? ( +
+ {t('customApi')} + +
+ ) + : null} {/* Both families edit the same rows through the same contract; only the extras differ — DeepSeek's inherited capacities, pi-ai's endpoint interrogation. */} diff --git a/packages/client/ui-models/tests/provider-form.spec.tsx b/packages/client/ui-models/tests/provider-form.spec.tsx index 9ff96cdebc..696e2258be 100644 --- a/packages/client/ui-models/tests/provider-form.spec.tsx +++ b/packages/client/ui-models/tests/provider-form.spec.tsx @@ -719,10 +719,103 @@ describe('hand-declared providers', () => { expect(fields()).toEqual([en.customRoute, en.customDisplayName, en.baseUrl, en.customApi, en.keyInput]) cleanup() + // A shipped route's models each carry their own protocol, so its editor + // offers no route-level protocol to override them with. await mountSection({ providers: { openai: { apiKeyEnv: 'OPENAI_API_KEY' } } }) openEditor('openai') fireEvent.click(screen.getByText(en.customized)) expect(fields()).toEqual([en.keyInput, en.baseUrl]) + cleanup() + + // A hand-declared route named its own protocol at creation, so editing it + // reaches the same field the create card asked for. + await mountSection({ + providers: { 'acme-gateway': { api: 'openai-completions', baseURL: 'https://gateway.acme.example/v1' } }, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + expect(fields()).toEqual([en.keyInput, en.customDisplayName, en.baseUrl, en.customApi]) + }) + + it('renames a declared route and falls back to its id when the name is cleared', async () => { + const { mutate } = await mountSection({ + providers: { + 'acme-gateway': { displayName: 'Acme Gateway', api: 'openai-completions', baseURL: 'https://acme.test/v1' }, + }, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + + const name = screen.getByLabelText(en.customDisplayName) + expect(name.value).toBe('Acme Gateway') + // The route id, not the stored name: it is what the route will be called + // the moment the field is cleared. + expect(name.placeholder).toBe('acme-gateway') + fireEvent.change(name, { target: { value: 'Acme 网关' } }) + fireEvent.click(screen.getByText(en.apply)) + + await waitFor(() => { expect(mutate).toHaveBeenCalledTimes(1) }) + expect(firstMutate(mutate).ops) + .toEqual([{ op: 'set', path: ['providers', 'acme-gateway', 'displayName'], value: 'Acme 网关' }]) + }) + + it('drops the stored name rather than storing an empty one the adapter refuses', async () => { + // `llm-pi-ai` rejects an empty displayName outright, so clearing the field + // must unset it — which is also what the user means: use the route id. + const { mutate } = await mountSection({ + providers: { 'acme-gateway': { displayName: 'Acme Gateway', api: 'openai-completions' } }, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + + fireEvent.change(screen.getByLabelText(en.customDisplayName), { target: { value: ' ' } }) + fireEvent.click(screen.getByText(en.apply)) + + await waitFor(() => { expect(mutate).toHaveBeenCalledTimes(1) }) + expect(firstMutate(mutate).ops) + .toEqual([{ op: 'unset', path: ['providers', 'acme-gateway', 'displayName'] }]) + }) + + it('edits the protocol a declared route was created with', async () => { + const { mutate } = await mountSection({ + providers: { + 'acme-gateway': { + apiKeyEnv: 'ACME_GATEWAY_API_KEY', + api: 'openai-completions', + baseURL: 'https://gateway.acme.example/v1', + models: [{ id: 'acme-large' }], + }, + }, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + + const protocol = screen.getByLabelText(en.customApi) + expect(protocol.value).toBe('openai-completions') + fireEvent.change(protocol, { target: { value: 'anthropic-messages' } }) + fireEvent.click(screen.getByText(en.apply)) + + await waitFor(() => { expect(mutate).toHaveBeenCalledTimes(1) }) + // Only the protocol travels: every other stored field is unchanged, so no + // op restates it. + expect(firstMutate(mutate)).toEqual({ + ns: 'llm-pi-ai', + ops: [{ op: 'set', path: ['providers', 'acme-gateway', 'api'], value: 'anthropic-messages' }], + expectedRevision: 3, + }) + }) + + it('selects nothing for a declared route whose profile names no protocol', async () => { + // A route hand-written into settings.yaml with no model needs no protocol + // to resolve, so the card can be opened over one. The select must not read + // as if that route had picked its first choice. + await mountSection({ + providers: { 'acme-gateway': { baseURL: 'https://gateway.acme.example/v1' } }, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + + expect(screen.getByLabelText(en.customApi).value).toBe('') }) it('retries only the key after the profile landed, and reports the provider on cancel', async () => { From 23995f88cec8fef41a7545e67c3c22b26ed97eef Mon Sep 17 00:00:00 2001 From: Yichen Jiang Date: Mon, 10 Aug 2026 23:18:26 +0800 Subject: [PATCH 02/16] fix(web): give the protocol dropdowns the shared chevron MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit `select.input` caps the control at 240px, and a `` that takes `.input` without `.selectInput`, so the next one cannot be added bare. --- ...-a-provider-from-the-models-page.i18n.yaml | 4 ++-- ...claring-a-provider-from-the-models-page.md | 2 +- ...ring-a-provider-from-the-models-page.zh.md | 2 +- .../src/client/CustomProviderCard.tsx | 2 +- .../ui-models/src/client/ProviderEditor.tsx | 2 +- .../client/ui-models/tests/styles.spec.ts | 23 +++++++++++++++++++ 6 files changed, 29 insertions(+), 6 deletions(-) diff --git a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml index bff5cc88fc..79aba9caa4 100644 --- a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml +++ b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md -2026-08-04-declaring-a-provider-from-the-models-page.md: 3c63254364e3803e9bddab2a5aad4c092e4f0426 -2026-08-04-declaring-a-provider-from-the-models-page.zh.md: c38b0387b3c7778b93fb43232c22a7acb24dc281 +2026-08-04-declaring-a-provider-from-the-models-page.md: dc0f6826be4c878c0ae5ffbca076f2b3037f1182 +2026-08-04-declaring-a-provider-from-the-models-page.zh.md: 5f25eff196869008d4e0d58776283c70de30e26a diff --git a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md index 3c63254364..dc0f6826be 100644 --- a/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md +++ b/.agents/notes/implemented/architecture/2026-08-04-declaring-a-provider-from-the-models-page.md @@ -48,4 +48,4 @@ What it costs: only pi-ai routes can be hand-declared, because `llm-pi-ai` is th ## Testing -`packages/client/ui-models/tests/provider-form.spec.tsx` drives the rendered page over a scripted wire face: adding, editing, and removing rows; a cleared optional field leaving the profile and a non-integer capacity never entering it; the interrogation carrying the edited endpoint, the unsaved key, and the profile's protocol; the picker's default selection, toggling, cancel, and adopt-keeps-tuned-rows; the empty, refused, and rejected-transport paths; the create writing one profile plus its credential; every gate on the create button; and the read-only posture. `protocolChoices` is covered against a schema that declares the union and one that does not. The editor's own field inventory is asserted per route kind — a catalog route stops at the key and the endpoint, a declared one also carries the protocol — along with the protocol edit travelling as a single `api` path op, a rename travelling as a single `displayName` one, a cleared name unsetting rather than storing the empty string the adapter refuses, and a declared profile naming no protocol selecting nothing rather than the first choice. `apps/web/tests/models-settings.e2e.ts` reopens the declared route through the real wire, captures the card, and asserts the chosen protocol and the new name both reach `settings.yaml` and the row re-registers under the rename. +`packages/client/ui-models/tests/provider-form.spec.tsx` drives the rendered page over a scripted wire face: adding, editing, and removing rows; a cleared optional field leaving the profile and a non-integer capacity never entering it; the interrogation carrying the edited endpoint, the unsaved key, and the profile's protocol; the picker's default selection, toggling, cancel, and adopt-keeps-tuned-rows; the empty, refused, and rejected-transport paths; the create writing one profile plus its credential; every gate on the create button; and the read-only posture. `protocolChoices` is covered against a schema that declares the union and one that does not. The stylesheet gate reads the package's own sources and fails any `` 都会失败——否则它保留的系统箭头会紧贴 `select.input` 所设 240px 上限的右边缘。编辑器自身的字段清单按路由种类各有断言——内置目录路由止于密钥与端点，已声明路由还带着协议——同时覆盖协议改动只以单条 `api` path op 传出、改名只以单条 `displayName` path op 传出、清空名称是取消设置而不是存入适配器会拒绝的空串，以及不写协议的已声明 profile 什么都不选中、而非选中第一个候选。`apps/web/tests/models-settings.e2e.ts` 经真实协议层重新打开这条已声明路由，捕获该卡片，并断言选定的协议与新名称都抵达了 `settings.yaml`、该行也以新名重新注册。 diff --git a/packages/client/ui-models/src/client/CustomProviderCard.tsx b/packages/client/ui-models/src/client/CustomProviderCard.tsx index b8b4459c77..f3c43fcc5a 100644 --- a/packages/client/ui-models/src/client/CustomProviderCard.tsx +++ b/packages/client/ui-models/src/client/CustomProviderCard.tsx @@ -236,7 +236,7 @@ export function CustomProviderCard(props: CustomProviderCardProps): ReactNode {
{t('customApi')} { expect(block('.rowCard')).not.toMatch(/\bbackground\s*:/) }) + it('gives every dropdown the shared chevron instead of the OS arrow', () => { + // `select.input` caps the control at 240px, and the OS arrow is painted + // flush inside that shrunk right edge — visibly tighter than every other + // control on the page. `.selectInput` is what removes it, reserves the + // right pad, and paints the shared chevron; a `
diff --git a/packages/client/ui-models/src/client/locales.ts b/packages/client/ui-models/src/client/locales.ts index 7d75e1de11..561a5df494 100644 --- a/packages/client/ui-models/src/client/locales.ts +++ b/packages/client/ui-models/src/client/locales.ts @@ -80,6 +80,7 @@ export const en = { customRouteTaken: 'A provider already uses this ID.', customDisplayName: 'Display name', customApi: 'API protocol', + customApiUnset: 'Not selected', customNeedsBaseUrl: 'A custom provider needs a base URL.', customNeedsModels: 'A custom provider needs at least one model.', create: 'Create provider', @@ -173,6 +174,7 @@ export const zh: typeof en = { customRouteTaken: '已有提供方使用了这个 ID。', customDisplayName: '显示名称', customApi: 'API 协议', + customApiUnset: '未选择', customNeedsBaseUrl: '自定义提供方需要填写 API 地址。', customNeedsModels: '自定义提供方至少需要一个模型。', create: '创建提供方', diff --git a/packages/client/ui-models/tests/provider-form.spec.tsx b/packages/client/ui-models/tests/provider-form.spec.tsx index 696e2258be..be0f198df7 100644 --- a/packages/client/ui-models/tests/provider-form.spec.tsx +++ b/packages/client/ui-models/tests/provider-form.spec.tsx @@ -5,7 +5,7 @@ import { afterEach, describe, expect, it, vi } from 'vitest' import Schema from 'schemastery' import { bindSnapshotSelector } from '@deepseek-ai/dsh-client-web-react' import type { RpcResponse, SettingsNamespaceView } from '@deepseek-ai/dsh-client-connection/client' -import { ModelsSection } from '../src/client/ModelsSection.tsx' +import { ModelsSection, providerCopy } from '../src/client/ModelsSection.tsx' import type { ModelsSectionInjected } from '../src/client/ModelsSection.tsx' import { CustomProviderCard } from '../src/client/CustomProviderCard.tsx' import { formatCapacity, parseCapacity } from '../src/client/DeepSeekModelsEditor.tsx' @@ -47,6 +47,7 @@ function fail(message: string, code: string): RpcResponse { function piAiNamespace( providers: Record, userProviders: Record = providers, + baseProviders: Record = {}, ): SettingsNamespaceView { return { ns: 'llm-pi-ai', @@ -54,7 +55,7 @@ function piAiNamespace( // `value` is the effective section; `user` is only the layer this page // writes. They differ whenever a composition `base` supplies something. value: { providers }, - base: {}, + base: { providers: baseProviders }, user: { providers: userProviders }, applies: 'live', secrets: [], @@ -66,6 +67,8 @@ function scriptedFace(options: { providers?: Record /** User layer, when it differs from the effective section. */ userProviders?: Record + /** Composition layer, for a route a `cordis.yml` pins rather than the page. */ + baseProviders?: Record /** Routes the adapter reports as hand-declared; the rest come back as shipped. */ declaredRoutes?: readonly string[] discover?: ReturnType @@ -75,7 +78,7 @@ function scriptedFace(options: { const providers = options.providers ?? { openai: { apiKeyEnv: 'OPENAI_API_KEY', baseURL: 'https://proxy.example/v1' }, } - const namespace = piAiNamespace(providers, options.userProviders ?? providers) + const namespace = piAiNamespace(providers, options.userProviders ?? providers, options.baseProviders ?? {}) const discover = options.discover ?? vi.fn(() => Promise.resolve(ok({ models: [] }))) const mutate = options.mutate ?? vi.fn(() => Promise.resolve(ok(namespace))) const set = options.set ?? vi.fn(() => Promise.resolve(ok({}))) @@ -759,6 +762,55 @@ describe('hand-declared providers', () => { .toEqual([{ op: 'set', path: ['providers', 'acme-gateway', 'displayName'], value: 'Acme 网关' }]) }) + it('offers the composition name as what a cleared field falls back to', async () => { + // A `cordis.yml` can pin a route the catalog does not ship, so a declared + // route's profile is not always the page's own. The field edits the user + // layer alone, and clearing it restores the layer beneath — the + // composition name here, not the route id — so that is what it offers. + await mountSection({ + providers: { 'acme-gateway': { displayName: 'Acme (pinned)', api: 'openai-completions' } }, + baseProviders: { 'acme-gateway': { displayName: 'Acme (pinned)', api: 'openai-completions' } }, + userProviders: {}, + declaredRoutes: ['acme-gateway'], + }) + openEditor('acme-gateway') + + const name = screen.getByLabelText(en.customDisplayName) + expect(name.value).toBe('') + expect(name.placeholder).toBe('Acme (pinned)') + }) + + it('names the provider as the refreshed directory reports it after a rename', async () => { + // The status line used to echo the target captured when the card opened, + // which never lied while the name could not change. It can now. + const { face } = await mountSection({ + providers: { 'acme-gateway': { displayName: 'Acme Gateway', api: 'openai-completions' } }, + declaredRoutes: ['acme-gateway'], + }) + // The reload after the write answers with the renamed route, exactly as + // the adapter re-registers it. + face.llm.providers = vi.fn(() => Promise.resolve(ok({ + providers: [{ + provider: 'acme-gateway', + displayName: 'Acme 网关', + settingsNs: 'llm-pi-ai', + settingsPath: ['providers', 'acme-gateway'], + active: true, + declared: true, + }], + }))) + openEditor('acme-gateway') + + fireEvent.change(screen.getByLabelText(en.customDisplayName), { target: { value: 'Acme 网关' } }) + fireEvent.click(screen.getByText(en.apply)) + + const notice = await screen.findByRole('status') + expect(notice.textContent).toBe(providerCopy(en.savedProvider, { + provider: 'acme-gateway', + displayName: 'Acme 网关', + })) + }) + it('drops the stored name rather than storing an empty one the adapter refuses', async () => { // `llm-pi-ai` rejects an empty displayName outright, so clearing the field // must unset it — which is also what the user means: use the route id. From 76cf6cbd0b8d8959c7aefc7d9789d53a74a2ce39 Mon Sep 17 00:00:00 2001 From: Hypatia May Date: Tue, 11 Aug 2026 11:50:39 +0800 Subject: [PATCH 04/16] feat(subagent): make final reports the continuable return contract A continuable child could finish without giving its parent a usable result: the report tool was described as optional and its default delivery stayed quiet. Install the child-scoped report obligation (tool description plus the tool:report prompt section), default reportDelivery to wakeup, bind shipped subagent_fork delegations to one-shot so fork keeps its prefix reuse, and extend the ACP snapshot harness to pin child system prompts alongside child tool schemas. --- ...8-10-fork-children-stay-one-shot.i18n.yaml | 6 + .../2026-08-10-fork-children-stay-one-shot.md | 49 ++++++ ...26-08-10-fork-children-stay-one-shot.zh.md | 49 ++++++ ...continuable-subagent-report-tool.i18n.yaml | 4 +- ...-07-30-continuable-subagent-report-tool.md | 20 +-- ...-30-continuable-subagent-report-tool.zh.md | 20 +-- ...tinuable-child-report-obligation.i18n.yaml | 6 + ...-06-continuable-child-report-obligation.md | 60 +++++++ ...-continuable-child-report-obligation.zh.md | 60 +++++++ docs/config-catalog.i18n.yaml | 4 +- docs/config-catalog.md | 9 +- docs/config-catalog.zh.md | 9 +- docs/module-graph.i18n.yaml | 4 +- docs/module-graph.md | 3 +- docs/module-graph.zh.md | 3 +- docs/tool-catalog.i18n.yaml | 4 +- docs/tool-catalog.md | 12 +- docs/tool-catalog.zh.md | 12 +- examples/acp-agent/cordis.yml | 6 +- examples/acp-agent/tests/acp.snapshot.ts | 8 +- .../advanced-toolchain/session.1.jsonl | 2 +- .../advanced-toolchain/session.2.jsonl | 2 +- .../system-prompt.expected.md | 4 +- .../tool-schemas.expected.json | 6 +- .../both-mode-turn/tool-schemas.expected.json | 6 +- .../code-mode-turn/system-prompt.expected.md | 4 +- .../lsp-definition/tool-schemas.expected.json | 6 +- .../tool-schemas.expected.json | 6 +- .../tool-schemas.expected.json | 6 +- .../pty-tools/tool-schemas.expected.json | 6 +- .../tool-schemas.expected.json | 6 +- .../tool-schemas.expected.json | 6 +- .../system-prompt.1.expected.md | 24 +++ .../tool-schemas.1.expected.json | 10 +- .../system-prompt.1.expected.md | 24 +++ .../tool-schemas.1.expected.json | 10 +- .../session.1.jsonl | 2 +- .../session.2.jsonl | 2 +- .../system-prompt.1.expected.md | 24 +++ .../tool-schemas.1.expected.json | 10 +- .../snapshots/subagent-mixed/session.1.jsonl | 2 +- .../snapshots/subagent-mixed/session.2.jsonl | 2 +- .../snapshots/subagent-multi/session.1.jsonl | 2 +- .../snapshots/subagent-multi/session.2.jsonl | 2 +- .../snapshots/subagent-report/input.json | 7 + .../snapshots/subagent-report/session.1.jsonl | 16 +- .../snapshots/subagent-report/session.jsonl | 54 ++++--- .../subagent-report/stdout.expected.jsonl | 1 + .../system-prompt.1.expected.md | 24 +++ .../tool-schemas.1.expected.json | 10 +- .../text-turn/tool-schemas.expected.json | 6 +- .../web-fetch/tool-schemas.expected.json | 6 +- examples/headless-agent/cordis.yml | 6 +- .../advanced-toolchain/session.1.jsonl | 10 +- .../advanced-toolchain/session.2.jsonl | 10 +- .../advanced-toolchain/session.jsonl | 28 ++-- .../compaction-recovery/session.jsonl | 14 +- .../tests/snapshots/pty-tools/session.jsonl | 34 ++-- packages/bundle/base/cordis.patch.yml | 6 +- .../subagent/subagent-fork/README.i18n.yaml | 4 +- packages/subagent/subagent-fork/README.md | 3 +- packages/subagent/subagent-fork/README.zh.md | 3 +- packages/subagent/subagent-fork/src/index.ts | 6 + .../tool-subagent-report/README.i18n.yaml | 4 +- .../subagent/tool-subagent-report/README.md | 20 +-- .../tool-subagent-report/README.zh.md | 20 +-- .../tool-subagent-report/package.json | 2 + .../tool-subagent-report/src/index.ts | 147 ++++++++++++------ .../tests/tool-subagent-report.spec.ts | 126 ++++++++++++++- .../tool-subagent-report/tsconfig.json | 3 + .../support/acp-snapshot/README.i18n.yaml | 4 +- packages/support/acp-snapshot/README.md | 2 + packages/support/acp-snapshot/README.zh.md | 2 + packages/support/acp-snapshot/src/suite.ts | 87 +++++++++-- .../fixtures/suite/plain-turn/behavior.json | 2 +- .../plain-turn/system-prompt.1.expected.md | 3 + .../support/acp-snapshot/tests/suite.spec.ts | 33 ++++ pnpm-lock.yaml | 3 + scripts/gen-tool-catalog.ts | 14 +- 79 files changed, 916 insertions(+), 326 deletions(-) create mode 100644 .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml create mode 100644 .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md create mode 100644 .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md create mode 100644 .agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml create mode 100644 .agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md create mode 100644 .agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md create mode 100644 examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md create mode 100644 examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md create mode 100644 examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md create mode 100644 examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md create mode 100644 packages/support/acp-snapshot/tests/fixtures/suite/plain-turn/system-prompt.1.expected.md diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml new file mode 100644 index 0000000000..ef424471d8 --- /dev/null +++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.i18n.yaml @@ -0,0 +1,6 @@ +# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each +# side as of the last confirmed-consistent state. Both languages carry equal authority; +# after editing either side, bring the other along and re-record with: +# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md +2026-08-10-fork-children-stay-one-shot.md: 030bed3c1b516a54afd5f97b00000ed7ceaf64d8 +2026-08-10-fork-children-stay-one-shot.zh.md: 4b033a70c3674315c491e85be67a1d357aeac51b diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md new file mode 100644 index 0000000000..030bed3c1b --- /dev/null +++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md @@ -0,0 +1,49 @@ +# Agent Note: Forked children stay one-shot + +Status: implemented + +English | [中文](2026-08-10-fork-children-stay-one-shot.zh.md) + +## Problem + +Fork's only difference from spawn is that the child Session is seeded with the parent's completed-turn prefix ([subagent-fork](../../../../packages/subagent/subagent-fork/README.md)). That seed costs real tokens — the inherited history is re-sent in every child request — and its one concrete payoff is provider-side prefix reuse: under the same provider and model, a child request whose leading bytes are identical to the parent's re-prefills none of the shared span. Anything a child scope adds *ahead* of the inherited history spends that payoff, because reuse stops at the first differing byte. + +The child-scoped `report` return channel is now the largest such addition, and since [the report obligation](../feature/2026-08-06-continuable-child-report-obligation.md) it is two deltas rather than one: the `report` tool schema and the `tool:report` system-prompt section. Both live in the request head — the system block and the tool block precede every message — so a continuable forked child invalidates reuse before the first inherited turn and re-prefills the whole transcript it was forked to reuse. That composition pays fork's duplication cost and collects none of its benefit, while the parent still holds a reusable prefix the child could have shared. + +## Decision + +Every shipped composition binds the fork delegation tool to `backgroundMode: one-shot`: [the base bundle](../../../../packages/bundle/base/cordis.patch.yml), [the ACP example](../../../../examples/acp-agent/cordis.yml), and [the headless example](../../../../examples/headless-agent/cordis.yml). The base bundle leaves `run_in_background` available, because it mounts a task service; the two examples set `enableRunInBackground: false`, because they mount none and a one-shot background start would otherwise fail at call time on a missing `tasks` service. + +One-shot children — foreground and background alike — are created through `SubagentService.start()`, which never enters the continuable activation-setup registry, so neither `report` nor its prompt section is installed. A forked one-shot child's system prompt and tool schemas therefore equal its parent's, apart from the `persona` and `toolFilter` deltas a deployment opts into per delegation tool. + +`spawn` keeps `backgroundMode: continuable`. Continuable children and the report obligation ship unchanged for the provider whose child starts with no inherited prefix to protect, so this decision costs the report channel nothing. + +### The restriction is composition, not code + +`ForkProvider.prepareContinuable` stays implemented and `ctx.subagents.startContinuable()` still accepts `fork`; only the shipped `cordis.yml` rows changed. `tool-subagent` knows both the provider's `inheritsParentContext` and its own `backgroundMode` at mount, so a load-time rejection of the pair was available and is deliberately not added: the pair is not wrong in general. It is wrong only while a child-scope delta precedes inherited history, and the package that creates that delta — [`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md) — is separately installable and, by its own design, invisible to `tool-subagent`. A deployment that omits the report package can run continuable forked children with the prefix intact. Encoding one roster's consequence as a delegation-tool invariant would make the tool assert something it cannot observe. + +The reintroduction condition is recorded as a `TODO(fork-continuable-prefix-reuse)` marker on `prepareContinuable` itself, the one method the shipped compositions do not call, and tracked as issue #2124: continuable fork reopens when a child's system prompt and tool schemas can match its parent's byte for byte. + +## Alternatives considered + +**Reject `inheritsParentContext` + `continuable` at mount.** A loud load-time failure would prevent silent reintroduction, which is what the configuration change cannot do. Rejected because the delegation tool cannot see the report package and the combination is legitimate without it; the invariant would be false for a deployment that never installs a child-scope delta, and `tool-subagent` would be asserting a fact owned by the roster. + +**Stop mounting the fork provider at all.** This was the broader form of the restriction. Rejected because foreground fork *is* the prefix-reusing case and is untouched by the report channel, so a full ban gives up the capability without buying anything the one-shot binding does not already buy — and would leave no shipped composition exercising session seeding. + +**Ship continuable forked children and accept the loss.** Rejected because the loss is total rather than marginal: reuse breaks ahead of the inherited history, so the child pays full prefill on a transcript it duplicated for the sole purpose of not paying it. A deployment that wants a long-lived child with no inherited context already has `spawn`. + +**Make `report` visible to every Agent.** A global registration would restore byte-identical prefixes by giving parent and child the same schema and section. Rejected because roots, one-shot children, remote children, and agentless callers would advertise a tool with no derivable recipient, and execution-time rejection would make schema visibility disagree with authority — the scope-local decision the [report tool Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.md) already settled. + +**Install the child-scope deltas after the inherited history.** Rejected as unrepresentable: the system prompt and the tool schemas are request-head structures in every provider's wire format, so no ordering within them can place a child-only addition behind the message list. + +## Consequences + +- No shipped composition creates a continuable forked child; `subagent_fork` returns a result to its caller's turn, and `send_message` addresses only spawned children. +- A forked child's request prefix stays byte-identical to its parent's unless the deployment configures `persona` or `toolFilter` on the fork delegation tool, so the token cost of seeding buys provider-side reuse again. +- The fork provider's continuable path has no production caller and no assembled-composition coverage. It keeps its package-level tests, and the seam still accepts it, so a bundle or `--patch` overlay can reintroduce it with no code change and no warning. +- `subagent_fork`'s model-visible schema changes: the continuable background wording is replaced by the one-shot task wording in the base bundle, and disappears entirely from the two examples. The affected keyless snapshot tool-schema sidecars are re-recorded in the same change. +- The report obligation's reach narrows to spawned children in shipped deployments. Its default `wakeup` scheduling, authority model, and coverage are unchanged. + +### Accepted risks + +The constraint lives in three configuration files and a code comment, not in a gate. A future bundle row or profile patch can set `backgroundMode: continuable` on a fork tool and silently reintroduce the prefix loss; nothing fails loud. That is the accepted cost of not encoding one roster's consequence into `tool-subagent`. diff --git a/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md new file mode 100644 index 0000000000..4b033a70c3 --- /dev/null +++ b/.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.zh.md @@ -0,0 +1,49 @@ +# Agent Note: fork 出的 child 保持 one-shot + +Status: implemented + +[English](2026-08-10-fork-children-stay-one-shot.md) | 中文 + +## 问题 + +fork 与 spawn 的唯一区别是 child 的 Session 会以 parent 已完成轮次的前缀作为初始内容（见 [subagent-fork](../../../../packages/subagent/subagent-fork/README.md)）。这份初始内容有实打实的 token 成本——继承的历史会在 child 的每次请求中重新发送——而它唯一确定的回报是提供方侧的前缀复用：在提供方与模型相同的前提下，起始字节与 parent 逐字节相同的 child 请求，无需为这段共享区间重新预填充。任何由 child 作用域添加在继承历史*之前*的内容都会消耗掉这份回报，因为复用在第一个不同字节处即告停止。 + +作用域局部的 `report` 返回通道现在是此类添加中最大的一项，而自[report 义务](../feature/2026-08-06-continuable-child-report-obligation.md)起它是两项而非一项增量：`report` 工具 schema，以及 `tool:report` 系统提示词 section。两者都位于请求头部——系统块与工具块先于所有消息——因此一个可继续的 fork child 会在第一条继承轮次之前就使复用失效，并重新预填充它当初 fork 就是为了复用的整份 transcript（文本记录）。这种组合付出了 fork 的复制成本却收不到它的收益，而 parent 手上仍握着一份 child 本可共享的可复用前缀。 + +## 决策 + +所有随附组合都把 fork 委派工具绑定为 `backgroundMode: one-shot`：[base 组合包](../../../../packages/bundle/base/cordis.patch.yml)、[ACP 示例](../../../../examples/acp-agent/cordis.yml)与[headless 示例](../../../../examples/headless-agent/cordis.yml)。base 组合包保留 `run_in_background`，因为它挂载了 task 服务；两个示例设置 `enableRunInBackground: false`，因为它们都不挂载 task 服务，否则一次 one-shot 后台启动会在调用时因缺少 `tasks` 服务而失败。 + +one-shot child——前台与后台皆然——经由 `SubagentService.start()` 创建，该路径从不进入可继续的 activation setup 注册表，因此 `report` 与它的提示词 section 都不会被安装。于是一个 fork 出的 one-shot child 的系统提示词与工具 schema 与其 parent 相同，只差部署逐个委派工具主动选择的 `persona` 与 `toolFilter` 增量。 + +`spawn` 保持 `backgroundMode: continuable`。对于 child 起步时本就没有继承前缀需要保护的那个提供方，可继续 child 与 report 义务随附行为不变，因此本决策没有让 report 通道付出任何代价。 + +### 该限制在于组合，不在于代码 + +`ForkProvider.prepareContinuable` 仍然实现完好，`ctx.subagents.startContinuable()` 也仍接受 `fork`；改动的只有随附的 `cordis.yml` 行。`tool-subagent` 在挂载时同时知道提供方的 `inheritsParentContext` 与自身的 `backgroundMode`，因此一个加载期拒绝该组合的检查是可行的，而这里刻意不加：该组合并非普遍错误。它只在某个 child 作用域增量位于继承历史之前时才是错的，而产生该增量的包——[`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md)——是独立安装的，并且按其自身设计对 `tool-subagent` 不可见。一个不安装 report 包的部署可以在前缀完好的前提下运行可继续的 fork child。把某一份插件清单的后果写成委派工具的不变量，会让该工具断言它无法观察到的事实。 + +重新开放的条件记录为 `prepareContinuable` 方法上的 `TODO(fork-continuable-prefix-reuse)` 标记——随附组合不调用这个方法——并由 issue #2124 跟踪：当 child 的系统提示词与工具 schema 能与其 parent 逐字节一致时，可继续 fork 即可重新开放。 + +## 备选方案 + +**在挂载时拒绝 `inheritsParentContext` 与 `continuable` 的组合。** 一次响亮的加载期失败可以阻止悄然的重新引入，而配置改动做不到这一点。否决的原因是委派工具看不到 report 包，且在没有它时该组合是合法的；对于从不安装任何 child 作用域增量的部署，这个不变量是假的，而 `tool-subagent` 会去断言一件由插件清单拥有的事实。 + +**干脆不挂载 fork 提供方。** 这是该限制更彻底的形式。否决的原因是前台 fork *正是*复用前缀的那种情形，且不受 report 通道影响，因此全面禁用会在不换来任何 one-shot 绑定尚未换来的东西的同时放弃该能力——并且随附组合将没有任何一个演练 session 初始内容。 + +**照常随附可继续的 fork child 并接受这份损失。** 否决的原因是这份损失是全额而非边际的：复用在继承历史之前就已中断，于是 child 为一份自己复制过来、目的恰恰是不必付费的 transcript 付了全额预填充。想要一个没有继承上下文的长期 child 的部署，本来就有 `spawn`。 + +**让 `report` 对每个 Agent 可见。** 全局注册会通过让 parent 与 child 拥有相同的 schema 与 section 来恢复逐字节相同的前缀。否决的原因是根 agent、one-shot child、远端 child 与无 agent 调用方都会宣告一件推导不出收件方的工具，而执行期拒绝会让 schema 可见性与权限彼此矛盾——这正是[report 工具 Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.md)已经定下的作用域局部决策。 + +**把 child 作用域增量安装到继承历史之后。** 否决的原因是它无法表达：在每个提供方的协议格式中，系统提示词与工具 schema 都是请求头部结构，因此它们内部的任何排序都无法把仅属于 child 的添加放到消息列表之后。 + +## 后果 + +- 没有任何随附组合会创建可继续的 fork child；`subagent_fork` 把结果返回给调用方的轮次，而 `send_message` 只寻址 spawn 出的 child。 +- 除非部署在 fork 委派工具上配置了 `persona` 或 `toolFilter`，fork child 的请求前缀与其 parent 逐字节相同，因此初始内容的 token 成本重新换来了提供方侧的复用。 +- fork 提供方的可继续路径没有生产调用方，也没有整体组装层面的覆盖。它保留自己的包内测试，seam 也仍然接受它，因此某个组合包或 `--patch` 覆盖层可以无需改动代码、也不会有任何警告地把它重新引入。 +- `subagent_fork` 面向模型的 schema 发生变化：base 组合包中可继续的后台措辞被 one-shot 的 task 措辞取代，在两个示例中则完全消失。受影响的无密钥快照工具 schema 伴随文件在同一次改动中重新记录。 +- 在随附部署中，report 义务的覆盖范围收窄到 spawn 出的 child。它的 `wakeup` 默认调度、权限模型与覆盖均保持不变。 + +### 已接受的风险 + +该限制存在于三个配置文件与一处代码注释中，而不在门禁里。未来某个组合包行或 profile 补丁可以在 fork 工具上设置 `backgroundMode: continuable`，从而悄然重新引入前缀损失；没有任何东西会失败得很响亮。这就是不把某一份插件清单的后果写入 `tool-subagent` 所接受的代价。 diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml index 775794a381..d250012eb9 100644 --- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml +++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md -2026-07-30-continuable-subagent-report-tool.md: 6a607f1da48fdfde36b2d9351f131dd3ef67ba7f -2026-07-30-continuable-subagent-report-tool.zh.md: b4d65fc71695e9fca46f4e2c69c9749745de260a +2026-07-30-continuable-subagent-report-tool.md: d4681d0ea30335a5537525927ecad615e6edd8b8 +2026-07-30-continuable-subagent-report-tool.zh.md: 731225d2a5b788997704fafe9331901ca9b84393 diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md index 6a607f1da4..d4681d0ea3 100644 --- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md +++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.md @@ -12,7 +12,7 @@ Treating every final assistant message as an implicit result would conflate turn ## Decision -Add the independently installed `@deepseek-ai/dsh-tool-subagent-report` package. It contributes an ordinary model-facing `report` tool to each continuable in-process child Activation. A child may call it zero or multiple times in a turn. Success neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. +Add the independently installed `@deepseek-ai/dsh-tool-subagent-report` package. It contributes an ordinary model-facing `report` tool to each continuable in-process child Activation. The mechanism accepts zero or multiple calls in a turn; the child is separately instructed to call it once before finishing ([the report obligation](2026-08-06-continuable-child-report-obligation.md)). Success neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. The feature is a collaboration control, not a result-bearing execution wrapper. It adds no Task, `SubagentRun`, result promise, Activation state, delivery queue, or replay path. @@ -22,7 +22,7 @@ The feature is a collaboration control, not a result-bearing execution wrapper. `messageId` is the stable `MessageId` of the user-role message accepted by the parent. It is not an `InboxItemId`: quiet delivery creates no inbox occurrence, while waking delivery creates one occurrence for the same stable message. It is also not a read receipt, parent-log acknowledgement, turn-completion receipt, or persistence flush. -The description states that reporting is explicit, repeatable, direct-parent-only, and non-terminal. It warns that a failed tool result may still follow an accepted send because a later `tools/post-execute` failure can replace the result. Without an idempotency key, stronger wording would encourage duplicate retries after ambiguous failure. +The description states that reporting is required before finishing, repeatable, direct-parent-only, and non-terminal. It warns that a failed tool result may still follow an accepted send because a later `tools/post-execute` failure can replace the result. Without an idempotency key, stronger wording would encourage duplicate retries after ambiguous failure. The tool uses generic rendering with no locations. Its acknowledgement includes `messageId`. Scope-local registration keeps presentation and execution aligned: roots, one-shot children, remote providers, sibling scopes, and agentless execution neither see nor execute `report`. It installs after the child's global `toolFilter`, so a delegation allow-list cannot accidentally remove the structural return channel; deployments that require no return channel omit the package. @@ -36,7 +36,7 @@ Nested reporting crosses exactly one edge. A grandchild reports to its direct ch ### Delivery policy -The package validates `reportDelivery: 'quiet' | 'wakeup'`; the default is `quiet`. +The package validates `reportDelivery: 'quiet' | 'wakeup'`; the default is `wakeup` ([why the default reversed](2026-08-06-continuable-child-report-obligation.md)). Quiet delivery calls `parent.inject()`. It adds model-visible context without starting a parent model request: an idle parent appends before the call returns, while an admitting or running parent stages the report for the next safe log position. It creates no inbox occurrence and therefore no synthetic continuation-manager acceptance record. @@ -56,13 +56,13 @@ The subagent seam adds `registerContinuableSetup(contribution): () => void`, bac The registry owns registration, per-child installation records, setup rollback, child-scope cleanup, and immediate revocation. Applying a batch returns the Agent setup commit that revalidates provisioning after every setup await and immediately before Agent publication. A throwing or concurrently revoked contribution therefore rejects before either Agent or Session publication and rolls back the batch. New registrations affect a resident child only on its next Activation; removing a registration first closes it to new setup and then revokes every provisioning or resident installation immediately. Registration disposal and child-context disposal are idempotent and attempt every release before aggregating failures. -This seam keeps the continuation manager unaware of tool names. The report package installs only `report`; `@deepseek-ai/dsh-tool-subagent-control` independently installs parent-side `send_message` and `list_agents`. A deployment can install either direction, both, or neither. Providers remain data-only, durable descriptors do not snapshot report availability or delivery mode, and cold resume uses the deployment's current contributions and policy. +This seam keeps the continuation manager unaware of tool names. The report package installs only `report` and its child-scoped guidance section; `@deepseek-ai/dsh-tool-subagent-control` independently installs parent-side `send_message` and `list_agents`. A deployment can install either direction, both, or neither. Providers remain data-only, durable descriptors do not snapshot report availability or delivery mode, and cold resume uses the deployment's current contributions and policy. ### Snapshot coverage -The ACP snapshot harness adds `waitForSubagentTurnEnd`, selecting the Nth harvested child by the same order as `session.N.jsonl`. It waits for a closed child turn containing a request header so a continuable child's earlier descriptor-seed turn cannot satisfy the boundary. This lets the assembled quiet-mode scenario wait for the child-side report without inventing a parent-visible signal. +The ACP snapshot harness adds `waitForSubagentTurnEnd`, selecting the Nth harvested child by the same order as `session.N.jsonl`. It waits for a closed child turn containing a request header so a continuable child's earlier descriptor-seed turn cannot satisfy the boundary. This lets the assembled scenario wait for the child-side report without inventing a parent-visible signal. -The authored snapshot starts a continuable child, executes the real scope-local `report` tool, confirms that the idle parent is not woken, and then submits a later parent prompt that consumes the framed report. It declares child schema pin `1`, so the otherwise non-global `report` schema is checked against `tool-schemas.1.expected.json` while the root keeps the default schema pin. The generated tool catalog separately mints a child scope to include the same scope-local schema. +The authored snapshot starts a continuable child, executes the real scope-local `report` tool, observes the one ordinary parent turn the default waking delivery creates, and then submits a later parent prompt that consumes the framed report. It declares child pins `1`, so the otherwise non-global `report` schema and the child's own prompt are checked against `tool-schemas.1.expected.json` and `system-prompt.1.expected.md` while the root keeps the class pins. The generated tool catalog separately mints a child scope to include the same scope-local schema. ## Alternatives considered @@ -72,7 +72,7 @@ Automatic delivery cannot represent zero reports, progress reports, or several s ### Always wake the parent -Waking on every report creates unsolicited turns and can cascade through nested subagents. Quiet delivery matches background coordination better as the default, while deployments that require immediate action can select wakeup. +Waking on every report creates unsolicited turns and can cascade through nested subagents. Quiet delivery was chosen as the default on the assumption that the parent had another reason to read its context. [The report obligation](2026-08-06-continuable-child-report-obligation.md) supersedes that choice: a parked background coordinator has no such reason, so waking is the default and this paragraph now records why `quiet` still exists. ### Let the child choose the delivery mode @@ -103,16 +103,16 @@ A post-creation revocation check can reject the Activation only after the Agent - A continuable in-process child exposes exactly one scope-local `report` schema only while the report package's contribution is installed; unrelated Agents never expose it. - The tool returns the parent message's stable `MessageId`. Quiet delivery has no `InboxItemId`; waking delivery has a separate inbox occurrence. - Only the exact resident child may report, and only to the exact live direct parent derived from durable lineage. The service has no recipient parameter or offline fallback. -- Quiet delivery is the validated default and never starts a parent request. Wakeup creates exactly one later FIFO turn and never steers an open turn. +- Waking delivery is the validated default: it creates exactly one later FIFO turn and never steers an open turn. Quiet delivery never starts a parent request. - Child cancellation or disposal after parent acceptance does not retract the report. Before acceptance, child disposal, drain, parent loss, or caller cancellation rejects the operation. - Fresh and resumed Activations compose current setup contributions before publication. Grants wait for the next Activation; revocation is immediate for resident children. - Unit coverage pins visibility, allow-list behavior, both delivery modes, stable message and sender identities, nested routing, invalid senders, absent parents, cancellation, drain, revocation races, and the absence of Tasks or implicit final reporting. -- The keyless assembled snapshot proves the real child tool, quiet non-wakeup behavior, durable parent framing, and later parent consumption. +- The keyless assembled snapshot proves the real child tool, the one waking parent turn, durable parent framing, and later parent consumption. ### Accepted risks The acceptance boundary is weaker than durable end-to-end delivery. A crash can leave the result ambiguous, and retries may duplicate reports. -Wakeup mode can amplify model work when nested children report frequently. Deployment ownership and a quiet default limit but do not remove that risk. +Waking delivery can amplify model work when nested children report frequently. Deployment ownership through `reportDelivery` bounds but does not remove that risk. Registry presence is the parent liveness signal. A host-owned parent whose `AgentHandle.dispose()` has started but has not yet unwound its scope can still accept and append a report that it will not act on in this process. Closing that gap requires an Agent-level disposal-start signal rather than subagent-layer inference. diff --git a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md index b4d65fc716..731225d2a5 100644 --- a/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md +++ b/.agents/notes/implemented/feature/2026-07-30-continuable-subagent-report-tool.zh.md @@ -12,7 +12,7 @@ Status: implemented ## 决策 -新增可独立安装的 `@deepseek-ai/dsh-tool-subagent-report` 包。它会向每个可继续进程内 child Activation 贡献一个普通的面向模型 `report` 工具。child 在一个轮次中可调用零次或多次。调用成功既不会结束该轮次或结算 Activation，也不会阻止 parent 之后继续 follow-up；完成轮次也绝不会自动报告。 +新增可独立安装的 `@deepseek-ai/dsh-tool-subagent-report` 包。它会向每个可继续进程内 child Activation 贡献一个普通的面向模型 `report` 工具。机制本身接受一个轮次中调用零次或多次；child 会另行被要求在结束前调用一次（见[报告义务](2026-08-06-continuable-child-report-obligation.md)）。调用成功既不会结束该轮次或结算 Activation，也不会阻止 parent 之后继续 follow-up；完成轮次也绝不会自动报告。 该功能是协作控制，不是承载结果的执行包装层。它不新增 Task、`SubagentRun`、结果 promise、Activation 状态、投递队列或回放路径。 @@ -22,7 +22,7 @@ Status: implemented `messageId` 是 parent 接受的用户角色消息所对应的稳定 `MessageId`。它不是 `InboxItemId`：静默投递不创建 inbox 条目实例，唤醒投递则会为同一条稳定消息创建一个条目实例。它也不是已读回执、parent 日志确认、轮次完成回执或持久化 flush。 -工具描述会明确报告操作是显式、可重复、仅限直接 parent 且不会结束轮次的。它还会警告：发送被接受后，后续 `tools/post-execute` 失败可能替换工具结果，因此工具结果失败时内容仍可能已经送达。没有幂等键时，更强的表述会诱导调用方在结果不明确的失败后重复重试。 +工具描述会明确报告操作在结束前必须执行、可重复、仅限直接 parent 且不会结束轮次。它还会警告：发送被接受后，后续 `tools/post-execute` 失败可能替换工具结果，因此工具结果失败时内容仍可能已经送达。没有幂等键时，更强的表述会诱导调用方在结果不明确的失败后重复重试。 该工具使用不带 location 的通用渲染，其确认中包含 `messageId`。作用域局部注册使呈现与执行保持一致：root、one-shot child、远程提供方、同级作用域和无 agent（智能体）执行既不能看到，也不能执行 `report`。它会在 child 的全局 `toolFilter` 之后安装，因此委派 allow-list 不会意外移除这条结构性返回通道；不需要返回通道的部署不安装该包。 @@ -36,7 +36,7 @@ root、one-shot child、伪造对象、陈旧 Agent 和同 id 替换对象都以 ### 投递策略 -该包会校验 `reportDelivery: 'quiet' | 'wakeup'`，默认值为 `quiet`。 +该包会校验 `reportDelivery: 'quiet' | 'wakeup'`，默认值为 `wakeup`（见[默认值反转的理由](2026-08-06-continuable-child-report-obligation.md)）。 静默投递调用 `parent.inject()`。它会添加模型可见上下文，但不启动 parent 模型请求：若 parent 空闲，则在调用返回前追加消息；若 parent 正在准入或运行，则暂存报告，留到下一个安全日志位置。该模式不创建 inbox 条目实例，因此也不会产生虚构的继续执行管理器接受记录。 @@ -56,13 +56,13 @@ subagent seam 新增 `registerContinuableSetup(contribution): () => void`，由 注册表负责注册、每个 child 的安装记录、设置回滚、child 作用域清理和立即撤销。应用一个批次会返回 Agent setup 提交对象，用于在所有 setup 的 await 均结算后、紧邻 Agent 发布前重新校验配置状态。因此，某项贡献抛出异常或被并发撤销时，会在 Agent 与会话发布前拒绝操作并回滚该批次。新注册项只会在驻留 child 的下一个 Activation 生效；移除注册项时，会先将它对新设置关闭，再立即撤销为正在配置或驻留的每个 child 安装的实例。注册 dispose（资源释放）与 child 上下文 dispose 都是幂等的，两者都会先尝试每项释放，再聚合失败。 -该 seam 使继续执行管理器无需知道工具名。report 包只安装 `report`；`@deepseek-ai/dsh-tool-subagent-control` 则独立安装 parent 侧的 `send_message` 和 `list_agents`。部署时可安装任一方向、同时安装两者或两者均不安装。提供方仍只负责数据，持久化描述符不会对 report 可用性或投递模式建立快照，冷恢复则使用部署当前的贡献与策略。 +该 seam 使继续执行管理器无需知道工具名。report 包只安装 `report` 及其 child 作用域指引 section；`@deepseek-ai/dsh-tool-subagent-control` 则独立安装 parent 侧的 `send_message` 和 `list_agents`。部署时可安装任一方向、同时安装两者或两者均不安装。提供方仍只负责数据，持久化描述符不会对 report 可用性或投递模式建立快照，冷恢复则使用部署当前的贡献与策略。 ### 快照覆盖 -ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，按与 `session.N.jsonl` 相同的顺序选择第 N 个已收集 child。它会等待一个包含请求 header 的已闭合 child 轮次，以防可继续 child 早期播种描述符的轮次错误满足该边界。这样，整体组装的静默模式场景无需伪造 parent 可见信号，就能等待 child 侧报告。 +ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`，按与 `session.N.jsonl` 相同的顺序选择第 N 个已收集 child。它会等待一个包含请求 header 的已闭合 child 轮次，以防可继续 child 早期播种描述符的轮次错误满足该边界。这样，整体组装的场景无需伪造 parent 可见信号，就能等待 child 侧报告。 -手写快照会启动一个可继续 child，执行真实的作用域局部 `report` 工具，确认空闲 parent 未被唤醒，然后提交一条后续 parent 提示词，使其消费封装后的报告。它声明 child schema pin `1`，因此本不属于全局的 `report` schema 会与 `tool-schemas.1.expected.json` 比对，root 则继续使用默认 schema pin。生成的工具目录会另外铸造一个 child 作用域，以收录同一个作用域局部 schema。 +手写快照会启动一个可继续 child，执行真实的作用域局部 `report` 工具，观察默认唤醒投递所产生的那一个普通 parent 轮次，然后提交一条后续 parent 提示词，使其消费封装后的报告。它声明 child pin `1`，因此本不属于全局的 `report` schema 与该 child 自身的提示词会分别与 `tool-schemas.1.expected.json` 和 `system-prompt.1.expected.md` 比对，root 则继续使用类别 pin。生成的工具目录会另外铸造一个 child 作用域，以收录同一个作用域局部 schema。 ## 曾考虑的替代方案 @@ -72,7 +72,7 @@ ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`， ### 始终唤醒 parent -每次报告都唤醒 parent 会产生未经请求的轮次，还可能沿嵌套 subagent 级联扩散。静默投递更适合作为后台协调的默认值，而需要立即处理的部署可选择 wakeup。 +每次报告都唤醒 parent 会产生未经请求的轮次，还可能沿嵌套 subagent 级联扩散。当初选择静默投递作为默认值，前提是 parent 还有别的理由去读自己的上下文。[报告义务](2026-08-06-continuable-child-report-obligation.md)取代了该选择：已经停驻的后台协调者并没有这样的理由，因此唤醒成为默认值，而本段现在记录的是 `quiet` 为何仍然保留。 ### 允许 child 选择投递模式 @@ -103,16 +103,16 @@ ACP（Agent Client Protocol）快照 harness 新增 `waitForSubagentTurnEnd`， - 只有安装 report 包贡献时，可继续进程内 child 才会恰好暴露一个作用域局部 `report` schema；无关 Agent 永远不会暴露该 schema。 - 工具返回 parent 消息的稳定 `MessageId`。静默投递没有 `InboxItemId`；唤醒投递会产生一个单独的 inbox 条目实例。 - 只有确切的驻留 child 才能报告，且只能报告给根据持久化谱系推导的确切在线直接 parent。服务不接受接收方参数，也不提供离线 fallback。 -- 静默投递是校验后的默认模式，绝不会启动 parent 请求。wakeup 会恰好创建一个后续 FIFO 轮次，绝不 steering 已开始的轮次。 +- 唤醒投递是校验后的默认模式：它会恰好创建一个后续 FIFO 轮次，绝不 steering 已开始的轮次。静默投递则绝不会启动 parent 请求。 - parent 接受后取消或 dispose child 不会撤回报告。接受前，child dispose、drain、parent 丢失或调用方取消都会拒绝操作。 - 新建和恢复的 Activation 都会在发布前组合当前设置贡献。新授权等待下一个 Activation 才生效，而已驻留 child 的授权撤销立即生效。 - 单元覆盖固定可见性、allow-list 行为、两种投递模式、稳定的消息与发送方身份、嵌套路由、无效发送方、缺失的 parent、取消、drain、撤销竞争，以及不存在 Task 或隐式最终报告。 -- 无密钥整体组装快照证明真实 child 工具、静默且不唤醒的行为、持久化 parent 封装，以及 parent 后续消费。 +- 无密钥整体组装快照证明真实 child 工具、那一个被唤醒的 parent 轮次、持久化 parent 封装，以及 parent 后续消费。 ### 已接受的风险 该接受边界弱于持久化端到端投递。崩溃可能导致结果不明，重试则可能重复报告。 -wakeup 模式可能在嵌套 child 频繁报告时放大模型工作量。由部署所有者控制并默认静默，可以限制该风险，但无法完全消除。 +唤醒投递可能在嵌套 child 频繁报告时放大模型工作量。通过 `reportDelivery` 交由部署所有者控制，可以限制该风险，但无法完全消除。 注册表中的存在性就是 parent 在线信号。宿主拥有的 parent 如果已开始 `AgentHandle.dispose()` 但尚未完成其作用域清理，仍可能接受并追加一条本进程不会再处理的报告。要弥合这个缺口，需要 Agent 层面的 dispose 开始信号，不能由 subagent 层推断。 diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml new file mode 100644 index 0000000000..3295cd47d1 --- /dev/null +++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.i18n.yaml @@ -0,0 +1,6 @@ +# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each +# side as of the last confirmed-consistent state. Both languages carry equal authority; +# after editing either side, bring the other along and re-record with: +# pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md +2026-08-06-continuable-child-report-obligation.md: ba52c72a94350d637699f778342093e71f2ed442 +2026-08-06-continuable-child-report-obligation.zh.md: 5c4178e0e095dac5b1e50ad92ef75d0138940439 diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md new file mode 100644 index 0000000000..ba52c72a94 --- /dev/null +++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.md @@ -0,0 +1,60 @@ +# Agent Note: The continuable child return channel is an obligation + +Status: implemented + +English | [中文](2026-08-06-continuable-child-report-obligation.zh.md) + +## Problem + +A continuable background child owns its own Session, so nothing it writes there reaches the agent that started it. [The report tool](2026-07-30-continuable-subagent-report-tool.md) gave that child a return channel and then presented it as one option among several: the schema said "call this zero or more times", nothing in the child's prompt asked it to call the tool at all, and the accepted default scheduling (`quiet`) added the report to a parked parent's next request without waking it. + +Each of those choices is defensible alone. Together they made the return channel unusable as a delegation contract. A child that finished its work, wrote its answer into its own transcript, and stopped left the parent with nothing; a child that did report reached a parent that had already parked and would not read the report until something unrelated woke it. External reports of parents busy-polling `list_agents`, re-sending messages to settled children, and abandoning `subagent` for `workflow` all reduce to the same missing guarantee. + +## Decision + +The return channel is an instruction the child receives, not a capability it may discover. The report package installs two scope-local registrations into every continuable in-process child, and one disposer revokes both: + +- the `report` tool, whose description now states that the child calls it once before finishing with a self-contained final result, and earlier for progress that changes what the parent should do next; +- a `tool:report` system-prompt section at order 117 carrying the same obligation in the child's own voice, so a child that never reads tool descriptions closely still receives it. + +`reportDelivery` now defaults to `wakeup`. An accepted report creates exactly one ordinary later parent turn and wakes a parked parent driver; it still never steers an open turn. `quiet` remains available for deployments that prefer unread reports over turn amplification. + +### Why the section and the description both exist + +They address different failure modes. The tool description is read when the model is already considering `report`; the prompt section is read when it is deciding whether it is finished. The obligation belongs at both points because the failure this fixes — a child that simply stops — happens at the second one. + +The section is registered on the child's own scope, the same mechanism [child composition](../../../../packages/subagent/subagent/src/child-agent.ts) already uses for a shadowing persona, so the parent and every sibling see neither the tool nor the guidance. `installReportTool` rolls the section back if tool registration fails, and its returned disposer attempts both revocations before surfacing cleanup failures. + +### Instruction, not enforcement + +Nothing rejects a child that never reports. No runtime path inspects whether a report was sent, and `report` still accepts zero or many calls per turn. The change is model-facing wording plus a scheduling default; the service authority, acknowledgement, and recovery contracts are unchanged. + +That boundary is deliberate: prompt text can only reach a child that is still running its own loop. A child stopped by an error, a token ceiling, cancellation, or teardown never gets the chance to comply. Covering those cases needs an account the runtime keeps itself, which this package cannot supply and does not claim to. + +### Snapshot coverage + +The assembled ACP `subagent-report` scenario now exercises the shipped default: the child reports, the parked parent takes one ordinary turn on that report, and a later prompt still reads the report back out of the durable log. Because the child's scope now composes a prompt the class pin cannot describe, the snapshot harness gained `pinsChildSystemPrompts`, the exact counterpart of the existing `pinsChildToolSchemas`: it moves one child fixture's prompt into `system-prompt..expected.md`, leaves every other request-header field to the class pin, requires the sidecar exactly when declared, and rejects a sidecar identical to that class pin so a redundant copy cannot drift. + +## Alternatives considered + +**Keep `quiet` as the default and rely on the prompt alone.** This was the shipped position, and it supersedes nothing on its own: a report the parent never reads is indistinguishable from a report never sent. The [report-tool note's](2026-07-30-continuable-subagent-report-tool.md) rejection of always-waking assumed the parent had another reason to look at its context; a parked background coordinator does not. Turn amplification is the real cost, and it is now the reason `quiet` still exists rather than the reason it is the default. + +**Let the child choose the delivery mode per call.** Unchanged from the original rejection: the model would own scheduler pressure, and behavior would vary per call rather than per deployment. + +**Put the obligation only in the tool description.** A description is read while choosing among tools. The child this change targets is not choosing a tool; it believes it is done. Prompt guidance is the surface that reaches that decision. + +**Enforce the obligation at settlement by rejecting a silent child.** There is nothing to reject: by the time settlement is observable the child's loop is over, and failing its teardown would destroy work rather than deliver it. Delivering the terminal facts unconditionally from the runtime is the answer to that case, and it belongs to the continuation manager, not to this package. + +## Consequences + +- Every continuable in-process child with this package loaded carries one extra prompt section and a longer `report` description in every request; no other Agent's request changes. +- The default deployment wakes the parent once per accepted report. A nested tree that reports frequently consumes extra parent turns; `quiet` is the documented escape. +- `installReportTool` requires `ctx.systemPrompt` in the child scope, so the package declares `systemPrompt` in `inject` and fails at load rather than at the next child materialization. +- Unit coverage pins the new default, two load-bearing instruction phrases, the section's child-only scope against both the parent and a sibling, and rollback or revocation of both registrations. +- Three assembled ACP scenarios with continuable children pin the complete instruction text through the new sidecar; a future change to any child-scoped section fails those scenarios instead of passing silently. + +### Accepted risks + +Waking by default amplifies model work in deep trees. The deployment owns that through `reportDelivery`, and the amplification is bounded by one turn per accepted report. + +A child can still finish without reporting, and this change cannot detect it. Closing that case needs the runtime's own settlement account, which this package does not own. diff --git a/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md new file mode 100644 index 0000000000..5c4178e0e0 --- /dev/null +++ b/.agents/notes/implemented/feature/2026-08-06-continuable-child-report-obligation.zh.md @@ -0,0 +1,60 @@ +# Agent Note: 可继续 child 的返回通道是一项义务 + +Status: implemented + +[English](2026-08-06-continuable-child-report-obligation.md) | 中文 + +## 问题 + +可继续后台 child 拥有自己的 Session，因此它写在那里的任何内容都不会到达启动它的 agent。[report 工具](2026-07-30-continuable-subagent-report-tool.md)为该 child 提供了一条返回通道，却把它呈现为若干选项之一：schema 里写着「可调用零次或多次」，child 的提示词中没有任何地方要求它调用该工具，而已采纳的默认调度（`quiet`）会把报告加入已停驻 parent 的下一次请求，却不唤醒它。 + +这些选择单独看都站得住脚。合在一起，它们让这条返回通道无法作为委派契约使用。一个完成工作、把答案写进自己 transcript（文本记录）随后停止的 child，会让 parent 一无所获；而确实上报了的 child，面对的是一个已经停驻、要等到别的事件把它唤醒才会读到报告的 parent。外部反馈中的 parent 忙轮询 `list_agents`、反复向已结算 child 发送消息、以及放弃 `subagent` 改用 `workflow`，都可归结为同一处缺失的保证。 + +## 决策 + +返回通道是 child 收到的一条指令，而不是它需要自行发现的能力。report 包会向每个可继续进程内 child 安装两项作用域局部注册，并由同一个 disposer 撤销两者： + +- `report` 工具，其描述现在说明 child 要在结束前调用一次并给出自足的最终结果，并在部分进展会改变 parent 下一步动作时提前调用； +- 一个 order 为 117 的 `tool:report` 系统提示词 section，用 child 自己的语气承载同一条义务，使从不细读工具描述的 child 仍能收到它。 + +`reportDelivery` 的默认值现在是 `wakeup`。一条被接受的报告恰好创建一个普通的后续 parent 轮次并唤醒停驻的 parent 驱动；它仍然绝不 steering（中途引导）已开始的轮次。对于宁可让报告无人阅读也要避免轮次放大的部署，`quiet` 依旧可用。 + +### 为什么 section 与描述同时存在 + +两者针对不同的失效模式。工具描述是在模型已经在考虑 `report` 时被读到的；提示词 section 是在它判断自己是否已经完成时被读到的。这条义务必须同时出现在两处，因为本次修复的失效——child 直接停下——发生在第二处。 + +该 section 注册在 child 自己的作用域上，与[child 组合](../../../../packages/subagent/subagent/src/child-agent.ts)为遮蔽式 persona 已经使用的机制相同，因此 parent 与所有同级都看不到该工具与该指引。工具注册失败时，`installReportTool` 会回滚该 section；它返回的 disposer 会先尝试撤销两项注册，再抛出清理失败。 + +### 是指令，不是强制 + +没有任何东西会拒绝一个从不上报的 child。没有任何运行时路径会检查是否发送过报告，`report` 仍接受一个轮次中调用零次或多次。本次改动是面向模型的措辞加上一个调度默认值；服务权限、确认与恢复契约都保持不变。 + +这条边界是刻意划定的：提示词文本只能到达仍在运行自身循环的 child。被错误、token 上限、取消或拆卸终止的 child 根本没有机会遵守。覆盖这些情形需要由运行时自己保存的记账，而本包无法提供，也不声称提供。 + +### 快照覆盖 + +整体组装的 ACP `subagent-report` 场景现在演练随附的默认行为：child 上报，停驻的 parent 就该报告执行一个普通轮次，随后的提示词仍能从持久化日志中把报告读回来。由于该 child 的作用域现在组合出类别 pin 无法描述的提示词，快照 harness 新增了 `pinsChildSystemPrompts`，它与既有 `pinsChildToolSchemas` 完全对称：把一个 child fixture 的提示词移入 `system-prompt..expected.md`，其余请求 header 字段仍归类别 pin 所有，要求 sidecar 恰好在声明时存在，并拒绝与该类别 pin 完全相同的 sidecar，使冗余副本无法悄悄漂移。 + +## 备选方案 + +**保留 `quiet` 作为默认值，只依赖提示词。** 这曾是随附的立场，而它本身什么也没有解决：一条 parent 从不阅读的报告，与一条从未发送的报告无法区分。[report 工具 Agent Note](2026-07-30-continuable-subagent-report-tool.md)对「始终唤醒」的否决，前提是 parent 还有别的理由去查看自己的上下文；已停驻的后台协调者并没有。轮次放大才是真正的代价，而它现在是 `quiet` 仍然保留的理由，而不是它作为默认值的理由。 + +**让 child 按调用选择投递模式。** 与最初的否决相同：模型将掌握调度压力，行为也会随调用而非随部署变化。 + +**只把义务写在工具描述里。** 描述是在从多个工具中选择时被读到的。本次改动针对的 child 并不在选择工具，它认为自己已经做完了。提示词指引才是能触及该判断的界面。 + +**在结算时拒绝沉默的 child，以此强制该义务。** 没有什么可以拒绝：当结算可被观察时 child 的循环已经结束，让它的拆卸失败只会毁掉工作而不会送达结果。由运行时无条件投递终止事实才是这一情形的答案，而它属于继续执行管理器，不属于本包。 + +## 后果 + +- 加载本包后，每个可继续进程内 child 的每次请求都会多出一个提示词 section 和一段更长的 `report` 描述；其他任何 Agent 的请求都不变。 +- 默认部署会为每条被接受的报告唤醒 parent 一次。频繁上报的嵌套树会消耗额外的 parent 轮次；`quiet` 是有文档记载的退路。 +- `installReportTool` 需要 child 作用域中的 `ctx.systemPrompt`，因此本包在 `inject` 中声明 `systemPrompt`，从而在加载时失败，而不是等到下一次 child 物化时。 +- 单元覆盖固定了新默认值、两处关键指令措辞、该 section 相对 parent 与同级均仅限 child 的作用域，以及两项注册在安装回滚或撤销时的清理。 +- 三个带可继续 child 的整体组装 ACP 场景通过新的 sidecar 逐字固定完整的 child 提示词；今后任何对 child 作用域 section 的改动都会让这些场景失败，而不是悄悄通过。 + +### 已接受的风险 + +默认唤醒会在深层树中放大模型工作量。部署通过 `reportDelivery` 掌握该取舍，且放大幅度以每条被接受报告一个轮次为界。 + +child 仍可能不上报就结束，本次改动无法检测这一点。补上这一情形需要运行时自己的结算记账，而本包并不拥有它。 diff --git a/docs/config-catalog.i18n.yaml b/docs/config-catalog.i18n.yaml index bc972314e8..3ce68e5c14 100644 --- a/docs/config-catalog.i18n.yaml +++ b/docs/config-catalog.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write docs/config-catalog.md -config-catalog.md: cc50a2021cfc378481dc3e830a0973773bae1e02 -config-catalog.zh.md: 5ed4f54a60130927ba710cae1f041da397c2d77a +config-catalog.md: 1646498f379acd628b5c28463b0cabd57d0112a4 +config-catalog.zh.md: 9b8042c616a10da9236b9db0820257a60fd8049d diff --git a/docs/config-catalog.md b/docs/config-catalog.md index cc50a2021c..1646498f37 100644 --- a/docs/config-catalog.md +++ b/docs/config-catalog.md @@ -2324,14 +2324,15 @@ Source: [`packages/subagent/tool-subagent/src/index.ts:25`](../packages/subagent ## `@deepseek-ai/dsh-tool-subagent-report` -Requires: `subagents` · `tools` +Requires: `subagents` · `tools` · `systemPrompt` ```ts config-catalog /** Config: how accepted reports are scheduled on the parent. */ export interface Config { /** - * Parent scheduling (default `quiet`). `quiet` adds context without waking; - * `wakeup` creates one ordinary later parent turn. + * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later + * parent turn; `quiet` adds context without waking, so a parked parent learns + * of the report only when something else wakes it. */ reportDelivery?: SubagentReportDelivery } @@ -2339,7 +2340,7 @@ export interface Config { Depends on: [`SubagentReportDelivery`](subsystems/subagent.md) -Source: [`packages/subagent/tool-subagent-report/src/index.ts:22`](../packages/subagent/tool-subagent-report/src/index.ts) +Source: [`packages/subagent/tool-subagent-report/src/index.ts:27`](../packages/subagent/tool-subagent-report/src/index.ts) ## `@deepseek-ai/dsh-tool-tasks` diff --git a/docs/config-catalog.zh.md b/docs/config-catalog.zh.md index 5ed4f54a60..9b8042c616 100644 --- a/docs/config-catalog.zh.md +++ b/docs/config-catalog.zh.md @@ -2325,14 +2325,15 @@ export interface Config { ## `@deepseek-ai/dsh-tool-subagent-report` -需要：`subagents` · `tools` +需要：`subagents` · `tools` · `systemPrompt` ```ts config-catalog /** Config: how accepted reports are scheduled on the parent. */ export interface Config { /** - * Parent scheduling (default `quiet`). `quiet` adds context without waking; - * `wakeup` creates one ordinary later parent turn. + * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later + * parent turn; `quiet` adds context without waking, so a parked parent learns + * of the report only when something else wakes it. */ reportDelivery?: SubagentReportDelivery } @@ -2340,7 +2341,7 @@ export interface Config { 依赖：[`SubagentReportDelivery`](subsystems/subagent.md) -来源：[`packages/subagent/tool-subagent-report/src/index.ts:22`](../packages/subagent/tool-subagent-report/src/index.ts) +来源：[`packages/subagent/tool-subagent-report/src/index.ts:27`](../packages/subagent/tool-subagent-report/src/index.ts) ## `@deepseek-ai/dsh-tool-tasks` diff --git a/docs/module-graph.i18n.yaml b/docs/module-graph.i18n.yaml index e0a5822779..b2f838dfce 100644 --- a/docs/module-graph.i18n.yaml +++ b/docs/module-graph.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write docs/module-graph.md -module-graph.md: cbe1a2d75ef33c44b31ac3b84bb9a54def97d0e5 -module-graph.zh.md: 3c7511cfd391ec69548588df3ab597457a420334 +module-graph.md: f87070e3eda9ab729dd364cee239b0a4adfcca54 +module-graph.zh.md: 8338461ff4fb50e3eab0b0e68d995138881b8f45 diff --git a/docs/module-graph.md b/docs/module-graph.md index cbe1a2d75e..f87070e3ed 100644 --- a/docs/module-graph.md +++ b/docs/module-graph.md @@ -1044,6 +1044,7 @@ flowchart TD pkg_tool_subagent_report --> pkg_invariants pkg_tool_subagent_report --> pkg_llm pkg_tool_subagent_report --> pkg_subagent + pkg_tool_subagent_report --> pkg_system_prompt pkg_tool_subagent_report --> pkg_tools pkg_hooks_claude --> pkg_agent pkg_hooks_claude --> pkg_hook_protocol @@ -1428,7 +1429,7 @@ flowchart TD | [`subagent-inprocess`](../packages/subagent/subagent-inprocess) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) | | [`tool-subagent`](../packages/subagent/tool-subagent) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tasks`](../packages/tasks/tasks), [`tools`](../packages/core/tools) | | [`tool-subagent-control`](../packages/subagent/tool-subagent-control) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | -| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | +| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) | | [`hooks-claude`](../packages/hooks/hooks-claude) | `hooks` | [`agent`](../packages/core/agent), [`hook-protocol`](../packages/hooks/hook-protocol), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | | [`web-app`](../packages/bundle/web-app) | `bundle` | [`bash-env`](../packages/bash/bash-env), [`invariants`](../packages/support/invariants), [`system-prompt`](../packages/core/system-prompt) | | [`client-ui-conversation`](../packages/client/ui-conversation) | `client` | [`agent`](../packages/core/agent), [`attachment`](../packages/attachment/attachment), [`brand`](../packages/util/brand), [`client-connection`](../packages/client/connection), [`client-locale`](../packages/client/locale), [`client-runtime`](../packages/client/runtime), [`client-ui-primitives`](../packages/client/ui-primitives), [`client-ui-slash`](../packages/client/ui-slash), [`client-ui-slots`](../packages/client/ui-slots), [`commands`](../packages/interaction/commands), [`compact`](../packages/compact/compact), [`invariants`](../packages/support/invariants), [`llm-retry`](../packages/llm/llm-retry), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools) | diff --git a/docs/module-graph.zh.md b/docs/module-graph.zh.md index 3c7511cfd3..8338461ff4 100644 --- a/docs/module-graph.zh.md +++ b/docs/module-graph.zh.md @@ -1046,6 +1046,7 @@ flowchart TD pkg_tool_subagent_report --> pkg_invariants pkg_tool_subagent_report --> pkg_llm pkg_tool_subagent_report --> pkg_subagent + pkg_tool_subagent_report --> pkg_system_prompt pkg_tool_subagent_report --> pkg_tools pkg_hooks_claude --> pkg_agent pkg_hooks_claude --> pkg_hook_protocol @@ -1430,7 +1431,7 @@ flowchart TD | [`subagent-inprocess`](../packages/subagent/subagent-inprocess) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) | | [`tool-subagent`](../packages/subagent/tool-subagent) | `subagent` | [`agent`](../packages/core/agent), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tasks`](../packages/tasks/tasks), [`tools`](../packages/core/tools) | | [`tool-subagent-control`](../packages/subagent/tool-subagent-control) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | -| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | +| [`tool-subagent-report`](../packages/subagent/tool-subagent-report) | `subagent` | [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`subagent`](../packages/subagent/subagent), [`system-prompt`](../packages/core/system-prompt), [`tools`](../packages/core/tools) | | [`hooks-claude`](../packages/hooks/hooks-claude) | `hooks` | [`agent`](../packages/core/agent), [`hook-protocol`](../packages/hooks/hook-protocol), [`invariants`](../packages/support/invariants), [`llm`](../packages/llm/llm), [`session`](../packages/core/session), [`session-persistence`](../packages/session/session-persistence), [`subagent`](../packages/subagent/subagent), [`tools`](../packages/core/tools) | | [`web-app`](../packages/bundle/web-app) | `bundle` | [`bash-env`](../packages/bash/bash-env), [`invariants`](../packages/support/invariants), [`system-prompt`](../packages/core/system-prompt) | | [`client-ui-conversation`](../packages/client/ui-conversation) | `client` | [`agent`](../packages/core/agent), [`attachment`](../packages/attachment/attachment), [`brand`](../packages/util/brand), [`client-connection`](../packages/client/connection), [`client-locale`](../packages/client/locale), [`client-runtime`](../packages/client/runtime), [`client-ui-primitives`](../packages/client/ui-primitives), [`client-ui-slash`](../packages/client/ui-slash), [`client-ui-slots`](../packages/client/ui-slots), [`commands`](../packages/interaction/commands), [`compact`](../packages/compact/compact), [`invariants`](../packages/support/invariants), [`llm-retry`](../packages/llm/llm-retry), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools) | diff --git a/docs/tool-catalog.i18n.yaml b/docs/tool-catalog.i18n.yaml index b74b7fa8ea..794985a9e5 100644 --- a/docs/tool-catalog.i18n.yaml +++ b/docs/tool-catalog.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write docs/tool-catalog.md -tool-catalog.md: f61b6daeb7209d0fa81606bda7718fd42c7f22dc -tool-catalog.zh.md: cbecdba084fa64ec78913706221a1558d206cde8 +tool-catalog.md: 58d560832ac8de5dc660a3c3ed75ab829cd262e8 +tool-catalog.zh.md: f2e330251db8a1be842dc05d89014a7be65369c6 diff --git a/docs/tool-catalog.md b/docs/tool-catalog.md index f61b6daeb7..58d560832a 100644 --- a/docs/tool-catalog.md +++ b/docs/tool-catalog.md @@ -31,9 +31,9 @@ This table connects model-visible tool names to the plugin package and service s | `@deepseek-ai/dsh-tool-ralph` | `ralph` | `ctx.tools`, `ctx.workflows`, `ctx.subagents`, `ctx.systemPrompt`, `a calling Agent (exec.agent parents every fresh round)` | `tool/call`, `tool/result`, `workflow and child session events during execution` | - | A fixed foreground workflow starts one fresh structured child per round; the model selects only the immutable objective and an optional round cap. | | `@deepseek-ai/dsh-tool-skill` | `skill` | `ctx.tools`, `ctx.agents`, `ctx.skills` | `tool/call`, `tool/result`, `user/message replacement catalogs via agent.inject()` | - | - | | `@deepseek-ai/dsh-tool-session-query` | `session_event_read`, `session_event_search`, `session_event_trace`, `session_search`, `session_trace` | `ctx.tools`, `ctx.systemPrompt`, `ctx.sessionQuery`, `a calling Agent for workspace authority` | `tool/call`, `tool/result` | - | The five read-only tools hide provider cursors and authorize every result from the immutable calling agent session. The package is opt-in; compositions that need enforced deadlines or bounded inline output also mount the generic timeout or spill policies. | -| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`, `ctx.subagents` | `tool/call`, `tool/result`, `child session events through the chosen provider` | `subagent`, `subagent_fork` | The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped example agents load this package once per subagent backend, so the model additionally sees `subagent_fork` (bound to the fork backend) with an identical schema — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. | +| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`, `ctx.subagents` | `tool/call`, `tool/result`, `child session events through the chosen provider` | `subagent`, `subagent_fork` | The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped compositions load this package once per subagent backend, so the model additionally sees `subagent_fork` bound to the fork backend. Each instance's description and `run_in_background` parameter follow its own `backgroundMode` and `enableRunInBackground`, so the two shipped schemas are not identical: `subagent` is `continuable`, while `subagent_fork` stays `one-shot` — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. | | `@deepseek-ai/dsh-tool-subagent-control` | `interrupt_agent`, `list_agents`, `send_message` | `ctx.tools`, `ctx.subagents`, `ctx.agents and ctx.sessionProjections (list_agents only)` | `tool/call`, `tool/result`, `child session events through ctx.subagents` | - | The globally named control tools over continuable background subagents: provider-bound `tool-subagent` instances register distinct delegation tools, while this package registers `send_message` and `interrupt_agent` once, plus `list_agents` from its separately loaded `/list-agents` plugin (whose catalog rows use the sessionProjections and live Agent registries). | -| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`, `a live continuable in-process child Agent` | `tool/call`, `tool/result`, `a user-role message in the direct parent session` | - | Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The parent-facing `send_message` tool is installed independently. | +| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`, `ctx.systemPrompt`, `a live continuable in-process child Agent` | `tool/call`, `tool/result`, `a user-role message in the direct parent session` | - | Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The same contribution installs the child-scoped `tool:report` prompt section, which this catalog does not render. The parent-facing `send_message` tool is installed independently. | | `@deepseek-ai/dsh-tool-tasks` | `task_kill`, `task_list`, `task_output` | `ctx.tools`, `ctx.tasks`, `ctx.systemPrompt` | `tool/call`, `tool/result`, `user/message via agent.inject() for background completion notices` | - | The kind-agnostic background-task control surface: background bash commands, PTY sends, and subagents are read, listed, and killed through the same three tools. Loading the plugin attaches the control surface that arms producers' `ctx.tasks.start()`. | | `@deepseek-ai/dsh-tool-todo` | `todo_write` | `ctx.tools`, `owning Agent session` | `tool/call`, `todo/write`, `tool/result` | - | todo_write is session-owned state; UIs render the latest todo/write event as a checklist. `allowParallelInProgress` is required with no default, so the catalog states its choice: `true`, whose description invites several `in_progress` items. A deployment choosing `false` receives the same tool with a description asking for exactly one active task. | | `@deepseek-ai/dsh-tool-workflow` | `workflow` | `ctx.tools`, `ctx.workflows`, `ctx.systemPrompt`, `a calling Agent (exec.agent parents the script children)` | `tool/call`, `tool/result` | - | - | @@ -1189,7 +1189,7 @@ Delegate a self-contained task to a subagent (a separate agent that works in its Source: [`packages/subagent/tool-subagent/src/index.ts`](../packages/subagent/tool-subagent/src/index.ts) -The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped example agents load this package once per subagent backend, so the model additionally sees `subagent_fork` (bound to the fork backend) with an identical schema — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. +The registered tool name is the load-time `toolName` config (default `subagent`); the schema above is that default. The shipped compositions load this package once per subagent backend, so the model additionally sees `subagent_fork` bound to the fork backend. Each instance's description and `run_in_background` parameter follow its own `backgroundMode` and `enableRunInBackground`, so the two shipped schemas are not identical: `subagent` is `continuable`, while `subagent_fork` stays `one-shot` — see `packages/bundle/base/cordis.patch.yml` and `examples/acp-agent/cordis.yml`. ## `@deepseek-ai/dsh-tool-subagent-control` @@ -1268,7 +1268,7 @@ The globally named control tools over continuable background subagents: provider ### `report` -Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it. +Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it. ```json { @@ -1276,7 +1276,7 @@ Report selected content to the agent that started you. Call this zero or more ti "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -1287,7 +1287,7 @@ Report selected content to the agent that started you. Call this zero or more ti Source: [`packages/subagent/tool-subagent-report/src/index.ts`](../packages/subagent/tool-subagent-report/src/index.ts) -Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The parent-facing `send_message` tool is installed independently. +Registered per continuable in-process child rather than globally, so this schema is visible only inside such a child and survives its global `toolFilter`. The same contribution installs the child-scoped `tool:report` prompt section, which this catalog does not render. The parent-facing `send_message` tool is installed independently. ## `@deepseek-ai/dsh-tool-tasks` diff --git a/docs/tool-catalog.zh.md b/docs/tool-catalog.zh.md index cbecdba084..f2e330251d 100644 --- a/docs/tool-catalog.zh.md +++ b/docs/tool-catalog.zh.md @@ -33,9 +33,9 @@ | `@deepseek-ai/dsh-tool-ralph` | `ralph` | `ctx.tools`、`ctx.workflows`、`ctx.subagents`、`ctx.systemPrompt`、`a calling Agent (exec.agent parents every fresh round)` | `tool/call`、`tool/result`、`workflow and child session events during execution` | - | 固定的前台工作流会在每个 Round 启动一个全新的结构化子级；模型只能选择不可变目标和可选的 Round 上限。 | | `@deepseek-ai/dsh-tool-skill` | `skill` | `ctx.tools`、`ctx.agents`、`ctx.skills` | `tool/call`、`tool/result`、`user/message replacement catalogs via agent.inject()` | - | - | | `@deepseek-ai/dsh-tool-session-query` | `session_event_read`、`session_event_search`、`session_event_trace`、`session_search`、`session_trace` | `ctx.tools`、`ctx.systemPrompt`、`ctx.sessionQuery`、`a calling Agent for workspace authority` | `tool/call`、`tool/result` | - | 这 5 个只读工具会隐藏提供方游标，并根据不可变的调用 agent 会话为每个结果授权。该包需要选择启用；需要强制截止时间或限制行内输出的组合还会挂载通用超时或 spill 策略。 | -| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`、`ctx.subagents` | `tool/call`、`tool/result`、`child session events through the chosen provider` | `subagent`、`subagent_fork` | 注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的示例 agent 会为每个 subagent 后端加载一次该包，因此模型还会看到 schema 相同、绑定到 fork 后端的 `subagent_fork`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 | +| `@deepseek-ai/dsh-tool-subagent` | `subagent` | `ctx.tools`、`ctx.subagents` | `tool/call`、`tool/result`、`child session events through the chosen provider` | `subagent`、`subagent_fork` | 注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的组合会为每个 subagent 后端加载一次该包，因此模型还会看到绑定到 fork 后端的 `subagent_fork`。每个实例的描述与 `run_in_background` 参数取决于它自己的 `backgroundMode` 与 `enableRunInBackground`，因此两个随附 schema 并不相同：`subagent` 为 `continuable`，而 `subagent_fork` 保持 `one-shot`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 | | `@deepseek-ai/dsh-tool-subagent-control` | `interrupt_agent`、`list_agents`、`send_message` | `ctx.tools`、`ctx.subagents`、`ctx.agents and ctx.sessionProjections (list_agents only)` | `tool/call`、`tool/result`、`child session events through ctx.subagents` | - | 这些是控制可继续后台 subagent 的全局命名工具：绑定提供方的 `tool-subagent` 实例注册不同的委派工具；本包注册一次 `send_message` 和 `interrupt_agent`，另由 `list_agents` 通过单独加载的 `/list-agents` 插件提供，其目录行使用 sessionProjections 和实时 Agent 注册表。 | -| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`、`a live continuable in-process child Agent` | `tool/call`、`tool/result`、`a user-role message in the direct parent session` | - | 按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。面向父级的 `send_message` 工具单独安装。 | +| `@deepseek-ai/dsh-tool-subagent-report` | `report` | `ctx.subagents`、`ctx.systemPrompt`、`a live continuable in-process child Agent` | `tool/call`、`tool/result`、`a user-role message in the direct parent session` | - | 按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。同一份贡献还会安装子级作用域的 `tool:report` 系统提示词 section，本目录不渲染该 section。面向父级的 `send_message` 工具单独安装。 | | `@deepseek-ai/dsh-tool-tasks` | `task_kill`、`task_list`、`task_output` | `ctx.tools`、`ctx.tasks`、`ctx.systemPrompt` | `tool/call`、`tool/result`、`user/message via agent.inject() for background completion notices` | - | 与任务种类无关的后台任务控制接口：后台 bash 命令、PTY 发送和 subagent 都通过相同的 3 个工具读取、列出和终止。加载该插件会挂接控制接口，从而启用生产方的 `ctx.tasks.start()`。 | | `@deepseek-ai/dsh-tool-todo` | `todo_write` | `ctx.tools`、`owning Agent session` | `tool/call`、`todo/write`、`tool/result` | - | todo_write 是会话所有的状态；UI 将最新的 todo/write 事件渲染为检查清单。`allowParallelInProgress` 是没有默认值的必填项，因此本目录明确选择 `true`，对应描述允许同时存在多个 `in_progress` 项。选择 `false` 的部署会获得同一工具，但描述会要求只能有 1 个活动任务。 | | `@deepseek-ai/dsh-tool-workflow` | `workflow` | `ctx.tools`、`ctx.workflows`、`ctx.systemPrompt`、`a calling Agent (exec.agent parents the script children)` | `tool/call`、`tool/result` | - | - | @@ -1193,7 +1193,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后， 来源：[`packages/subagent/tool-subagent/src/index.ts`](../packages/subagent/tool-subagent/src/index.ts) -注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的示例 agent 会为每个 subagent 后端加载一次该包，因此模型还会看到 schema 相同、绑定到 fork 后端的 `subagent_fork`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 +注册的工具名称取决于加载时 `toolName` 配置（默认为 `subagent`）；上述 schema 对应默认值。随产品发布的组合会为每个 subagent 后端加载一次该包，因此模型还会看到绑定到 fork 后端的 `subagent_fork`。每个实例的描述与 `run_in_background` 参数取决于它自己的 `backgroundMode` 与 `enableRunInBackground`，因此两个随附 schema 并不相同：`subagent` 为 `continuable`，而 `subagent_fork` 保持 `one-shot`；见 `packages/bundle/base/cordis.patch.yml` 和 `examples/acp-agent/cordis.yml`。 ## `@deepseek-ai/dsh-tool-subagent-control` @@ -1272,7 +1272,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后， ### `report` -向启动你的 agent 报告选定内容。你可以调用 0 次或多次，以报告进度、发现或最终答案。报告不会结束你的轮次或完成你的工作，且只有直接父级会收到。失败的调用仍可能已经送达，因此不要盲目重复。 +向启动你的 agent 报告选定内容。在你结束前调用一次，给出自包含的最终结果；当进度或发现会改变该 agent 接下来的行动时，也可以更早调用。该 agent 与你共享工作区，但不会自动收到你的 transcript（文本记录）、工具输出或推理，因此完成你的工作本身并不等于交出结果。报告不会结束你的轮次或完成你的工作，且只有直接父级会收到。失败的调用仍可能已经送达，因此不要盲目重复。 ```json { @@ -1280,7 +1280,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后， "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -1291,7 +1291,7 @@ lsp 工具将提供方选择和语言服务器子进程置于 ctx.lsp 之后， 来源：[`packages/subagent/tool-subagent-report/src/index.ts`](../packages/subagent/tool-subagent-report/src/index.ts) -按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。面向父级的 `send_message` 工具单独安装。 +按可继续的进程内子级注册，而非全局注册，因此该 schema 仅在这种子级内部可见，并且不受其全局 `toolFilter` 影响。同一份贡献还会安装子级作用域的 `tool:report` 系统提示词 section，本目录不渲染该 section。面向父级的 `send_message` 工具单独安装。 ## `@deepseek-ai/dsh-tool-tasks` diff --git a/examples/acp-agent/cordis.yml b/examples/acp-agent/cordis.yml index 59e7b78278..a73d6d36e4 100644 --- a/examples/acp-agent/cordis.yml +++ b/examples/acp-agent/cordis.yml @@ -121,12 +121,16 @@ backgroundMode: continuable maxDepth: 1 +# Fork stays one-shot because a continuable child's `report` tool and prompt +# section precede the inherited history a fork reuses; `run_in_background` is off +# because this example mounts no task service. See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md. - id: tool-subagent-fork name: '@deepseek-ai/dsh-tool-subagent' config: provider: fork toolName: subagent_fork - backgroundMode: continuable + backgroundMode: one-shot + enableRunInBackground: false maxDepth: 1 diff --git a/examples/acp-agent/tests/acp.snapshot.ts b/examples/acp-agent/tests/acp.snapshot.ts index a05bce3269..3b89ae8a35 100644 --- a/examples/acp-agent/tests/acp.snapshot.ts +++ b/examples/acp-agent/tests/acp.snapshot.ts @@ -359,6 +359,7 @@ const SCENARIOS: Scenario[] = [ hasModelTurn: true, recorded: false, pinsChildToolSchemas: [1], + pinsChildSystemPrompts: [1], configPath: SUBAGENT_DURABILITY_FAILURE_CONFIG, }, // Authored policy-inheritance transcript: the root session is switched to @@ -371,6 +372,7 @@ const SCENARIOS: Scenario[] = [ hasModelTurn: true, recorded: false, pinsChildToolSchemas: [1], + pinsChildSystemPrompts: [1], configPath: SUBAGENT_CONTINUABLE_INHERITANCE_CONFIG, }, // The in-process child is published before its first follow-up fails. The @@ -385,13 +387,14 @@ const SCENARIOS: Scenario[] = [ configPath: SUBAGENT_DURABILITY_FAILURE_CONFIG, }, // Authored child-to-parent transcript: the child calls its scope-local - // `report`, quiet delivery reaches the idle parent without waking it, and a - // later parent turn consumes the logged report. + // `report`, the default waking delivery gives the parked parent one ordinary + // turn on that report, and a later prompt still reads it back from the log. { name: 'subagent-report', hasModelTurn: true, recorded: false, pinsChildToolSchemas: [1], + pinsChildSystemPrompts: [1], }, // Authored durable-catalog transcript: the snapshot-only lifecycle marker // fences the second parent turn behind the child's Activation end, so @@ -404,6 +407,7 @@ const SCENARIOS: Scenario[] = [ hasModelTurn: true, recorded: false, pinsChildToolSchemas: [1], + pinsChildSystemPrompts: [1], }, { name: 'subagent-depth-two-rejection', diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl index 4abda00a44..2a084e4a9c 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357538308,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}} {"type":"step/start","seq":5,"time":1786357538310,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730458555,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"ebe0cfa0-a909-47e0-8294-28ad84a8fe77"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"b8a6a626-fd2c-4602-a8d4-8074f229bfb5"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"1b650647-a278-4be1-a76e-bf09540bff75"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357538310,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730458555,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730458555,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl index 0a10247d2e..252bfb326c 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357538469,"data":{"version":2,"mode":"one-shot","provider":"spawn"}} {"type":"step/start","seq":5,"time":1786357538470,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730458703,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"2ac2cc54-9bce-4cfa-a569-a64f51bc30a7"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d327cdff-ad2e-4f9e-9c54-d4448ee11f2a"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c74c4556-1622-4d2e-9745-c5abd087fa83"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357538471,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730458703,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730458703,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md index b613fbac76..dcd7e67bca 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md @@ -141,14 +141,12 @@ interface ToolArgsMap { /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ run_in_background?: boolean; } & Record; - /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */ + /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */ subagent_fork: { /** A short (3-5 word) description of the delegated task, for display. */ description: string; /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */ prompt: string; - /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ - run_in_background?: boolean; } & Record; /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */ task_kill: { diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json index c979e1db87..5b1ee082a7 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json @@ -334,7 +334,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -345,10 +345,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json index d5cbdc2ccc..1f323c3ea9 100644 --- a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json @@ -277,7 +277,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -288,10 +288,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md index cb1b047e93..f02d9ab50a 100644 --- a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md +++ b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md @@ -124,14 +124,12 @@ interface ToolArgsMap { /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ run_in_background?: boolean; } & Record; - /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */ + /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */ subagent_fork: { /** A short (3-5 word) description of the delegated task, for display. */ description: string; /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */ prompt: string; - /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ - run_in_background?: boolean; } & Record; /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */ task_kill: { diff --git a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json index 335be3672b..f00decd284 100644 --- a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json @@ -293,7 +293,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -304,10 +304,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json index 76f60e28d4..0c85c0265d 100644 --- a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json @@ -298,7 +298,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -309,10 +309,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json index 84c4671579..74c8cdf4b0 100644 --- a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json @@ -277,7 +277,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -288,10 +288,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json index 37d7d467be..18a14221d9 100644 --- a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json @@ -256,7 +256,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -267,10 +267,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json index aed34842df..e018ff89a4 100644 --- a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json @@ -460,7 +460,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -471,10 +471,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json index dc07c990aa..d0cc2da730 100644 --- a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json @@ -319,7 +319,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -330,10 +330,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md new file mode 100644 index 0000000000..71f4ed05f8 --- /dev/null +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/system-prompt.1.expected.md @@ -0,0 +1,24 @@ +You are an AI agent powered by the DeepSeek Harness SDK. + +You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug. + +Verify your work by running the code or tests. Keep answers brief and factual. + + +Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files. + +Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes. + +Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session. + +Check the [exit code: N] marker on every bash result; investigate failures before moving on. + +Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering. + +Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked. + +Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls. + +Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out. + +Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn. diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json index 7dd791cf27..6cecb69da2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json @@ -194,13 +194,13 @@ }, { "name": "report", - "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", + "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", "parameters": { "type": "object", "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -272,7 +272,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -283,10 +283,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md new file mode 100644 index 0000000000..71f4ed05f8 --- /dev/null +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/system-prompt.1.expected.md @@ -0,0 +1,24 @@ +You are an AI agent powered by the DeepSeek Harness SDK. + +You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug. + +Verify your work by running the code or tests. Keep answers brief and factual. + + +Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files. + +Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes. + +Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session. + +Check the [exit code: N] marker on every bash result; investigate failures before moving on. + +Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering. + +Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked. + +Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls. + +Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out. + +Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn. diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json index 7dd791cf27..6cecb69da2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json @@ -194,13 +194,13 @@ }, { "name": "report", - "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", + "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", "parameters": { "type": "object", "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -272,7 +272,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -283,10 +283,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl index 44d587c851..cfb7fee5a8 100644 --- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357533600,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth one"}} {"type":"step/start","seq":5,"time":1786357533602,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730456014,"data":{"content":[{"type":"text","text":"Call subagent once. Ask that child to attempt one further subagent call, then report the result."}],"source":{"kind":"user"},"role":"user","id":"a8129357-1bde-4cbd-90b4-6b8ad51d52e1"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"82a11d62-da49-4ad3-a243-789ea3cd7c08"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"722c8476-461f-4a02-904a-f1084b973c95"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357533602,"data":{"title":"Call subagent once. Ask that","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730456014,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730456014,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl index 89de9b1c4f..022584de40 100644 --- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357533628,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth two"}} {"type":"step/start","seq":5,"time":1786357533630,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730456041,"data":{"content":[{"type":"text","text":"Attempt one subagent call beyond the configured cap, then report the rejection."}],"source":{"kind":"user"},"role":"user","id":"d4dc5a16-e542-4dd9-8e82-e6b7829cfc4b"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"d5488efe-eea2-4019-8fcb-7e6e49077d8a"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"fc837f97-bbf9-4c44-9eea-3a8f52c7a77d"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357533630,"data":{"title":"Attempt one subagent call beyond","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730456041,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730456042,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md new file mode 100644 index 0000000000..71f4ed05f8 --- /dev/null +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/system-prompt.1.expected.md @@ -0,0 +1,24 @@ +You are an AI agent powered by the DeepSeek Harness SDK. + +You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug. + +Verify your work by running the code or tests. Keep answers brief and factual. + + +Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files. + +Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes. + +Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session. + +Check the [exit code: N] marker on every bash result; investigate failures before moving on. + +Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering. + +Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked. + +Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls. + +Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out. + +Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn. diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json index 7dd791cf27..6cecb69da2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json @@ -194,13 +194,13 @@ }, { "name": "report", - "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", + "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", "parameters": { "type": "object", "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -272,7 +272,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -283,10 +283,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl index e510929344..442c7f06c8 100644 --- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357524752,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Reply ALPHA only"}} {"type":"step/start","seq":5,"time":1786357524755,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730450187,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"73ce401a-faaf-408a-879e-7485380d537d"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"56ad93fa-0cc0-4ccb-a1b4-258f4801c681"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"36cca213-d6e9-491d-9442-87057994f573"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357524755,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730450187,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730450188,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl index a5f4ab88a4..6e071fa68d 100644 --- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl @@ -27,7 +27,7 @@ {"type":"subagent/descriptor","seq":41,"time":1786357524800,"data":{"version":2,"mode":"one-shot","provider":"fork","label":"Recall project codeword"}} {"type":"step/start","seq":42,"time":1786357524803,"data":{"turn":2,"step":1}} {"type":"user/message","seq":43,"time":1786357524803,"data":{"content":[{"type":"text","text":"What is the project codeword mentioned earlier in this conversation? Reply with exactly that one word and nothing else."}],"source":{"kind":"user"},"role":"user","id":"86e9f144-764f-460d-b72b-262cffe43d77"},"surfaceOp":"append"} -{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"6ea5a774-b0da-47ff-84b7-226a4a207bbf"},"surfaceOp":"append"} +{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"86f2fe9d-3154-41c2-815e-e1c1e18ed895"},"surfaceOp":"append"} {"type":"request/header","seq":45,"time":1786358036900,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"resume"}} {"type":"assistant/chunk","seq":46,"time":1783352148077,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}} {"type":"reasoning-chunks","seq0":47,"time0":1783352148077,"data":{"turn":2,"step":1,"index":0,"dt":[0,0,29,0,0,0,35,0,0,0,0,26,29,31,0,30,0,0,27,1,27,0,1,0,0,31,1,0,0,1790157964],"texts":["The"," user"," is"," asking"," me"," to"," recall"," the"," project"," cod","ew","ord"," that"," was"," mentioned"," earlier"," in"," the"," conversation","."," I"," was"," told"," to"," remember"," it",":"," SA","FF","RON","."]}} diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl index da79d6b23c..e5848d377c 100644 --- a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357521754,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return ALPHA only"}} {"type":"step/start","seq":5,"time":1786357521756,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730447828,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a287f842-f6f2-4a17-ab4c-820e41f498d5"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"f654b5a4-b4c0-4443-8eab-d84624d804f1"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c253c429-6ca2-4ac4-a192-ea1383e90a1e"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357521756,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730447828,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730447828,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl index f7326e37c8..95f756b06c 100644 --- a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357521799,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return BETA only"}} {"type":"step/start","seq":5,"time":1786357521801,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730447881,"data":{"content":[{"type":"text","text":"Reply with exactly the word BETA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"53f6419d-8ddc-4eee-8803-5b68411336f9"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"05843da4-4a5f-46fc-a00f-5257b2bd271d"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"488666ba-7318-4b83-bc3f-d7700024b196"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357521802,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730447881,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730447881,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/input.json b/examples/acp-agent/tests/snapshots/subagent-report/input.json index b15e0e4d71..15ca4b53eb 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/input.json +++ b/examples/acp-agent/tests/snapshots/subagent-report/input.json @@ -14,6 +14,13 @@ { "op": "waitForSubagentTurnEnd" }, + { + "op": "waitForTurnStart", + "minimumTurn": 2 + }, + { + "op": "waitForTurnEnd" + }, { "op": "promptAndWaitForAgentMessage", "text": "Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools.", diff --git a/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl index e32d1bdee0..b83358f93d 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-report/session.1.jsonl @@ -1,14 +1,14 @@ {"type":"session","version":0,"id":"33333333-3333-4333-8333-333333333333","createdAt":1789000001000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1} {"type":"subagent/descriptor","seq":0,"time":1785594881508,"data":{"version":2,"mode":"continuable","provider":"spawn","label":"Report a finding","agentProvider":"deepseek-official","agentModel":"deepseek-v4-flash"}} {"type":"session/end-seed","seq":1,"time":1785594881508,"data":{}} -{"type":"approval/policy","seq":2,"time":1786357530605,"data":{"policy":"never","source":"delegation"}} -{"type":"agent/inbox/spliced","seq":3,"time":1786357530605,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"}]}} -{"type":"turn/start","seq":4,"time":1786357530605,"data":{"turn":1}} -{"type":"agent/inbox/spliced","seq":5,"time":1786357530605,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} -{"type":"step/start","seq":6,"time":1786357530633,"data":{"turn":1,"step":1}} +{"type":"approval/policy","seq":2,"time":1786374158805,"data":{"policy":"never","source":"delegation"}} +{"type":"agent/inbox/spliced","seq":3,"time":1786374158806,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"}]}} +{"type":"turn/start","seq":4,"time":1786374158806,"data":{"turn":1}} +{"type":"agent/inbox/spliced","seq":5,"time":1786374158806,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":6,"time":1786374158840,"data":{"turn":1,"step":1}} {"type":"user/message","seq":7,"time":1785730453639,"data":{"content":[{"type":"text","text":"Call the report tool once with output exactly CHILD_REPORT_OK, then stop."}],"source":{"kind":"user"},"role":"user","id":"9045ac78-393a-4f24-b20d-8999286dd6ce"},"surfaceOp":"append"} -{"type":"user/message","seq":8,"time":1786357530633,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"1677b901-cce7-461a-8b2a-7f119dd9d845"},"surfaceOp":"append"} -{"type":"session/title","seq":9,"time":1786357530633,"data":{"title":"Call the report tool once","messageSeqs":[7],"source":{"kind":"fallback"}}} +{"type":"user/message","seq":8,"time":1786374158840,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"0f107d71-9b56-4ad8-b6f1-d93cb4c82105"},"surfaceOp":"append"} +{"type":"session/title","seq":9,"time":1786374158840,"data":{"title":"Call the report tool once","messageSeqs":[7],"source":{"kind":"fallback"}}} {"type":"request/header","seq":10,"time":1785730453639,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":11,"time":1785730453639,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":12,"time":1785594881546,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -18,7 +18,7 @@ {"type":"assistant/chunk","seq":16,"time":1785730453647,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} {"type":"assistant/message","seq":17,"time":1785730453647,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_report_1","name":"report","arguments":"{\"output\": \"CHILD_REPORT_OK\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c9e50afb-b732-41ab-b0fc-8e98948ad9ec"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[12,13,14,15,16],"surfaceOp":"append"} {"type":"tool/call","seq":18,"time":1785730453647,"data":{"turn":1,"step":1,"callId":"call_report_1","name":"report","arguments":"{\"output\": \"CHILD_REPORT_OK\"}"}} -{"type":"tool/result","seq":19,"time":1785730453654,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_report_1"},"content":[{"type":"tool-result","toolCallId":"call_report_1","content":[{"type":"text","text":"report accepted by the agent that started you as message 1f4b61e2-6c6d-4db8-836b-ac5760c5e484"}],"isError":false}],"role":"user","id":"cee5f084-bfab-423d-b8bc-1b1b7d88d4fa"}},"sourceEventSeqs":[18],"surfaceOp":"append"} +{"type":"tool/result","seq":19,"time":1785730453654,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_report_1"},"content":[{"type":"tool-result","toolCallId":"call_report_1","content":[{"type":"text","text":"report accepted by the agent that started you as message 87627538-d804-4d36-bb10-4768b6fcfb65"}],"isError":false}],"role":"user","id":"22f25be2-d3f9-4558-b3ea-db22fa900aa3"}},"sourceEventSeqs":[18],"surfaceOp":"append"} {"type":"step/end","seq":20,"time":1785730453654,"data":{"turn":1,"step":1}} {"type":"step/start","seq":21,"time":1785730453664,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":22,"time":1785594881567,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl index eb36af2399..edbef518f1 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl @@ -1,10 +1,10 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"5cf78378-e004-4fd5-af4f-cef3b7e190ad"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"e45e4b6d-dae7-42d9-9edb-d24385664025"}]}} {"type":"turn/start","seq":1,"time":1785821411429,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821411429,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1785730453591,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"5cf78378-e004-4fd5-af4f-cef3b7e190ad"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"d1a851a3-604f-4a42-8e5f-4e480857a3b4"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"e45e4b6d-dae7-42d9-9edb-d24385664025"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"9319509b-82df-4848-8f3a-7654804c9c45"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730453592,"data":{"title":"Follow these steps exactly, then","messageSeqs":[4],"source":{"kind":"fallback"}}} {"type":"request/header","seq":7,"time":1785730453592,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730453593,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} @@ -13,9 +13,9 @@ {"type":"assistant/chunk","seq":11,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}}}} {"type":"assistant/chunk","seq":12,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730453601,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"664c39ff-9dac-4bb3-a151-e18a7863d15a"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"dcd1402a-35c7-4748-bfeb-8734e1018732"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730453602,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}} -{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"d7927ad2-29db-4de8-ab4d-59a4ebcddd72"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"b752b617-5282-40b6-bfe0-0535fb069504"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730453613,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730453623,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":1789000000018,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -23,22 +23,32 @@ {"type":"assistant/chunk","seq":21,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}} {"type":"assistant/chunk","seq":22,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730453628,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"571faad7-adbd-480c-922a-1499e1329ead"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"d5bf8946-4dd4-464e-88de-c7d6191c7671"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"step/end","seq":25,"time":1785730453628,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":26,"time":1785730453629,"data":{"turn":1,"reason":{"kind":"completed"}}} -{"type":"agent/inbox/spliced","seq":27,"time":1785730453654,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"824dc60a-f9d7-48ea-a0d4-6d56df83bd4f"}]}} -{"type":"agent/inbox/spliced","seq":28,"time":1785730453673,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"43f17984-22c3-48b9-911e-923a2f68dce0"}]}} -{"type":"turn/start","seq":29,"time":1785821411548,"data":{"turn":2}} -{"type":"agent/inbox/spliced","seq":30,"time":1785730453673,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}} -{"type":"agent/inbox/spliced","seq":31,"time":1785821411548,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} -{"type":"step/start","seq":32,"time":1785730453683,"data":{"turn":2,"step":1}} -{"type":"user/message","seq":33,"time":1785730453683,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"824dc60a-f9d7-48ea-a0d4-6d56df83bd4f"},"surfaceOp":"append"} -{"type":"user/message","seq":34,"time":1785730453683,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"43f17984-22c3-48b9-911e-923a2f68dce0"},"surfaceOp":"append"} -{"type":"assistant/chunk","seq":35,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} -{"type":"assistant/chunk","seq":36,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}} -{"type":"assistant/chunk","seq":37,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}} -{"type":"assistant/chunk","seq":38,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} -{"type":"assistant/chunk","seq":39,"time":1785730453687,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":40,"time":1785730453687,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"77bba235-d2b6-4a32-9ba3-ebb69d9b0654"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[35,36,37,38,39],"surfaceOp":"append"} -{"type":"step/end","seq":41,"time":1785730453687,"data":{"turn":2,"step":1}} -{"type":"turn/end","seq":42,"time":1785730453687,"data":{"turn":2,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":27,"time":1786008565468,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2ba9ef7b-dac7-4f39-8035-35f0bf88d0e7"}]}} +{"type":"turn/start","seq":28,"time":1786008565468,"data":{"turn":2}} +{"type":"agent/inbox/spliced","seq":29,"time":1786008565468,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":30,"time":1786008565474,"data":{"turn":2,"step":1}} +{"type":"user/message","seq":31,"time":1786008565474,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2ba9ef7b-dac7-4f39-8035-35f0bf88d0e7"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":32,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":33,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"REPORT_RECEIVED"}}} +{"type":"assistant/chunk","seq":34,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"REPORT_RECEIVED"}}}} +{"type":"assistant/chunk","seq":35,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":36,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":37,"time":1786008565479,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"REPORT_RECEIVED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0fe8d69e-6640-4bd5-b72b-70f2cf013cde"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} +{"type":"step/end","seq":38,"time":1786008565479,"data":{"turn":2,"step":1}} +{"type":"turn/end","seq":39,"time":1786008565479,"data":{"turn":2,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":40,"time":1786008565505,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"50d55072-bbf2-4e30-8948-1b837d8db9ee"}]}} +{"type":"turn/start","seq":41,"time":1786008565505,"data":{"turn":3}} +{"type":"agent/inbox/spliced","seq":42,"time":1786008565505,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":43,"time":1786008565511,"data":{"turn":3,"step":1}} +{"type":"user/message","seq":44,"time":1786008565511,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"50d55072-bbf2-4e30-8948-1b837d8db9ee"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":45,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":46,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}} +{"type":"assistant/chunk","seq":47,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}} +{"type":"assistant/chunk","seq":48,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":49,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":50,"time":1786008565516,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"de2ca55e-a346-4f25-911d-a33316b9866b"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[45,46,47,48,49],"surfaceOp":"append"} +{"type":"step/end","seq":51,"time":1786008565516,"data":{"turn":3,"step":1}} +{"type":"turn/end","seq":52,"time":1786008565516,"data":{"turn":3,"reason":{"kind":"completed"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl index 09cd3e4a68..cd63d489c7 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl @@ -2,5 +2,6 @@ {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"STARTED"}}}} {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"REPORT_RECEIVED"}}}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"CHILD_REPORT_OK"}}}} {"jsonrpc":"2.0","id":4,"result":{"stopReason":"end_turn"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md b/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md new file mode 100644 index 0000000000..71f4ed05f8 --- /dev/null +++ b/examples/acp-agent/tests/snapshots/subagent-report/system-prompt.1.expected.md @@ -0,0 +1,24 @@ +You are an AI agent powered by the DeepSeek Harness SDK. + +You are a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug. + +Verify your work by running the code or tests. Keep answers brief and factual. + + +Use the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files. + +Use the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes. + +Use the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session. + +Check the [exit code: N] marker on every bash result; investigate failures before moving on. + +Track every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering. + +Use goal tools for one long-running completion objective in the current session. create_goal may infer goal intent from a direct human request in any language; do not create a goal for routine single-turn work. Call get_goal before update_goal and copy its exact goal_id and revision. After session resume or fork, an active goal is disarmed: when a human asks to continue or resume in any wording or language, use update_goal action resume to rearm it. Mark complete only when the objective is actually achieved. Mark blocked only after the same blocking condition persists for at least 3 consecutive goal rounds, and report that concrete condition in blocked_reason; difficulty, uncertainty, or useful remaining work is not blocked. + +Use the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls. + +Use the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out. + +Deliver your result with the report tool before you finish: call it once with a self-contained answer. The agent that started you shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can use. Report earlier as well whenever a partial finding changes what that agent should do next; reporting never ends your turn. diff --git a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json index 7dd791cf27..6cecb69da2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json @@ -194,13 +194,13 @@ }, { "name": "report", - "description": "Report selected content to the agent that started you. Call this zero or more times for progress, findings, or a final answer. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", + "description": "Report selected content to the agent that started you. Call this once before you finish, with a self-contained final result, and earlier for progress or findings that change what that agent does next. That agent shares your workspace but does not automatically receive your transcript, tool output, or reasoning, so finishing your work is not itself a result. Reporting does not end your turn or finish your work, and only your direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.", "parameters": { "type": "object", "properties": { "output": { "type": "string", - "description": "Self-contained content for your parent; it does not see your private work." + "description": "Actionable content for your parent; summarize conclusions and reference relevant shared paths." } }, "required": [ @@ -272,7 +272,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -283,10 +283,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json index e0b7a41bcf..4b7d68131f 100644 --- a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json @@ -256,7 +256,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -267,10 +267,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json index 97c51af14b..9f198a86a5 100644 --- a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json @@ -256,7 +256,7 @@ }, { "name": "subagent_fork", - "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.", "parameters": { "type": "object", "properties": { @@ -267,10 +267,6 @@ "prompt": { "type": "string", "description": "The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new." - }, - "run_in_background": { - "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." } }, "required": [ diff --git a/examples/headless-agent/cordis.yml b/examples/headless-agent/cordis.yml index b6f1b774f2..61a0190142 100644 --- a/examples/headless-agent/cordis.yml +++ b/examples/headless-agent/cordis.yml @@ -118,12 +118,16 @@ backgroundMode: continuable maxDepth: 1 +# Fork stays one-shot because a continuable child's `report` tool and prompt +# section precede the inherited history a fork reuses; `run_in_background` is off +# because this example mounts no task service. See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md. - id: tool-subagent-fork name: '@deepseek-ai/dsh-tool-subagent' config: provider: fork toolName: subagent_fork - backgroundMode: continuable + backgroundMode: one-shot + enableRunInBackground: false maxDepth: 1 # The worker-thread workflow engine fans a model-written JavaScript script's diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl index 8a1c23140b..acbf5f8cc9 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl @@ -1,19 +1,19 @@ {"type":"session","version":0,"id":"22222222-2222-4222-8222-222222222222","createdAt":1783950001000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1} -{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"1c70f3f7-2e85-4808-8376-03d4d3bee6e6"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"48a309a8-4004-43e9-87ea-0bd02e9f5857"}]}} {"type":"turn/start","seq":1,"time":1785821454445,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454445,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"subagent/descriptor","seq":3,"time":1785821454466,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}} {"type":"step/start","seq":4,"time":1785730501506,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"1c70f3f7-2e85-4808-8376-03d4d3bee6e6"},"surfaceOp":"append"} -{"type":"user/message","seq":6,"time":1786358103673,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c9a305d4-add2-453e-8789-4e5c127725f7"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"48a309a8-4004-43e9-87ea-0bd02e9f5857"},"surfaceOp":"append"} +{"type":"user/message","seq":6,"time":1786358103673,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"b8be7a23-617a-4aad-93e6-d12f0593f71d"},"surfaceOp":"append"} {"type":"session/title","seq":7,"time":1786358103673,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":9,"time":1785730501507,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":10,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":11,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"DIRECT_CHILD_OK"}}} {"type":"assistant/chunk","seq":12,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DIRECT_CHILD_OK"}}}} {"type":"assistant/chunk","seq":13,"time":1785498583897,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":14,"time":1785730501507,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c5d4e091-9632-4535-af35-097bc74abdd3"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} +{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"9f582f43-25e4-4f21-a81c-0186f42fadf9"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501507,"data":{"turn":1,"step":1}} {"type":"turn/end","seq":17,"time":1785730501507,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl index 8882bda5af..c5773c21aa 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl @@ -1,19 +1,19 @@ {"type":"session","version":0,"id":"33333333-3333-4333-8333-333333333333","createdAt":1783950002000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1} -{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"12fb8b24-9214-4e43-b3d3-47f2af3531f1"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"11d0804b-a6bb-4dcc-be05-6f35f81f4f13"}]}} {"type":"turn/start","seq":1,"time":1785821454599,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454599,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"subagent/descriptor","seq":3,"time":1785821454618,"data":{"version":2,"mode":"one-shot","provider":"spawn"}} {"type":"step/start","seq":4,"time":1785730501645,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"12fb8b24-9214-4e43-b3d3-47f2af3531f1"},"surfaceOp":"append"} -{"type":"user/message","seq":6,"time":1786358103827,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"14cf8f47-3a7a-4857-a548-02fe407683fb"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"11d0804b-a6bb-4dcc-be05-6f35f81f4f13"},"surfaceOp":"append"} +{"type":"user/message","seq":6,"time":1786358103827,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"0d36194f-6afa-41a4-907e-c9e0583b797a"},"surfaceOp":"append"} {"type":"session/title","seq":7,"time":1786358103827,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":9,"time":1785730501646,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":10,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":11,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"WORKFLOW_CHILD_OK"}}} {"type":"assistant/chunk","seq":12,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"WORKFLOW_CHILD_OK"}}}} {"type":"assistant/chunk","seq":13,"time":1785498584067,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":14,"time":1785730501646,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"bb9c39a9-3239-4ee1-939a-bab0046e3028"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} +{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"eb7781b5-d2ac-4a94-8e21-c9f769524a56"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501646,"data":{"turn":1,"step":1}} {"type":"turn/end","seq":17,"time":1785730501646,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl index 8f0a211969..a3faa19c4b 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl @@ -1,20 +1,20 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1783950000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"2b5f9025-324a-4c76-b883-4af3b5c3060a"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"342d68c5-9a35-46dc-9d80-d18a2375a0cf"}]}} {"type":"turn/start","seq":1,"time":1785821454304,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454304,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1783957884486,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"2b5f9025-324a-4c76-b883-4af3b5c3060a"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"342d68c5-9a35-46dc-9d80-d18a2375a0cf"},"surfaceOp":"append"} {"type":"session/title","seq":5,"time":1785498583779,"data":{"title":"Run this advanced flow exactly","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement a normal local, project, or repository Plugin through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":7,"time":1785730501403,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":8,"time":1783950000007,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":9,"time":1783950000008,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"advanced-mount","name":"cordis_mount","argumentsDelta":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}} {"type":"assistant/chunk","seq":10,"time":1783950000009,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}}} {"type":"assistant/chunk","seq":11,"time":1785498583784,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":12,"time":1785730501404,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b596133f-aafe-4485-9871-ade1dda23373"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} +{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0933e3da-0073-49c2-af4a-76c95ea72f02"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} {"type":"tool/call","seq":14,"time":1785730501404,"data":{"turn":1,"step":1,"callId":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}} -{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"99819f4a-5e53-4a6f-92f6-5be96b765bce"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"716e9e93-6560-4f6d-ad52-0cb8e833da39"}},"sourceEventSeqs":[14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501413,"data":{"turn":1,"step":1}} {"type":"step/start","seq":17,"time":1785730501423,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":18,"time":1783950000017,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -22,11 +22,11 @@ {"type":"assistant/chunk","seq":20,"time":1783950000019,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}}}} {"type":"assistant/chunk","seq":21,"time":1785498583804,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":22,"time":1785730501424,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6697e967-e6bd-46b2-8574-18aeb914e7c6"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"} +{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0a6cdb2d-c83c-4b7c-af56-2bf3fbece4ae"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"} {"type":"tool/call","seq":24,"time":1785730501424,"data":{"turn":1,"step":2,"callId":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}} {"type":"tool/code-dispatch-start","seq":25,"time":1785730501473,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"}}} {"type":"tool/code-dispatch","seq":26,"time":1785730501474,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"},"isError":false,"content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}]}} -{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"831070a8-3dc0-4275-9f11-0476b47b8ef2"}},"sourceEventSeqs":[24],"surfaceOp":"append"} +{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"045ef2ea-7cfb-4a43-a6ef-d4ef7ad979e1"}},"sourceEventSeqs":[24],"surfaceOp":"append"} {"type":"step/end","seq":28,"time":1785730501475,"data":{"turn":1,"step":2}} {"type":"step/start","seq":29,"time":1785730501483,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":30,"time":1783950000029,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -34,9 +34,9 @@ {"type":"assistant/chunk","seq":32,"time":1785037378923,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}}}} {"type":"assistant/chunk","seq":33,"time":1785498583869,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":34,"time":1785730501484,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"cef24cf0-5f9e-4be2-93d8-93a0d89c6e82"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"} +{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"26ff28fa-fab8-4e92-a72f-d3a5b3352573"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"} {"type":"tool/call","seq":36,"time":1785730501484,"data":{"turn":1,"step":3,"callId":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}} -{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"d7ea395e-840a-46c4-a143-b6e15cf74114"}},"sourceEventSeqs":[36],"surfaceOp":"append"} +{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"e33536a0-0801-4f51-a633-baac1167cfee"}},"sourceEventSeqs":[36],"surfaceOp":"append"} {"type":"step/end","seq":38,"time":1785730501508,"data":{"turn":1,"step":3}} {"type":"step/start","seq":39,"time":1785730501521,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":40,"time":1783957884594,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -44,9 +44,9 @@ {"type":"assistant/chunk","seq":42,"time":1785037378946,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}}}} {"type":"assistant/chunk","seq":43,"time":1785498583919,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":44,"time":1785730501522,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"242be4fa-3293-45de-ab55-a017999f2333"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"} +{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"492fd6ab-4937-4083-a8e2-395d5d254c3e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"} {"type":"tool/call","seq":46,"time":1785730501522,"data":{"turn":1,"step":4,"callId":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}} -{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"ee764f11-8827-4984-acef-ec3c5880f5a0"}},"sourceEventSeqs":[46],"surfaceOp":"append"} +{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"31627ac4-1b5d-4213-814c-95599db9a215"}},"sourceEventSeqs":[46],"surfaceOp":"append"} {"type":"step/end","seq":48,"time":1785730501648,"data":{"turn":1,"step":4}} {"type":"step/start","seq":49,"time":1785730501660,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":50,"time":1783957884719,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -54,9 +54,9 @@ {"type":"assistant/chunk","seq":52,"time":1785037379534,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}}}} {"type":"assistant/chunk","seq":53,"time":1785498584085,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":54,"time":1785730501661,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6c3dab82-c2ed-492a-ab0d-f235b340a6c1"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"} +{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0b2fa8a0-7559-4fcf-a34d-bb190f8d9f48"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"} {"type":"tool/call","seq":56,"time":1785730501661,"data":{"turn":1,"step":5,"callId":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}} -{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"74554b42-640e-4119-a413-ee5ac01e546e"}},"sourceEventSeqs":[56],"surfaceOp":"append"} +{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"cfc46278-d37c-4ced-8763-200e37d3290f"}},"sourceEventSeqs":[56],"surfaceOp":"append"} {"type":"step/end","seq":58,"time":1785730501668,"data":{"turn":1,"step":5}} {"type":"step/start","seq":59,"time":1785730501678,"data":{"turn":1,"step":6}} {"type":"assistant/chunk","seq":60,"time":1783957884720,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -64,6 +64,6 @@ {"type":"assistant/chunk","seq":62,"time":1785037379541,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ADVANCED_HEADLESS_OK"}}}} {"type":"assistant/chunk","seq":63,"time":1785498584102,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":64,"time":1785730501679,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1a3bef32-0610-4891-9071-6bdc2e8a8fd2"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"} +{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4c1bdf08-4721-4c52-9f75-56b2c2028c7d"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"} {"type":"step/end","seq":66,"time":1785730501679,"data":{"turn":1,"step":6}} {"type":"turn/end","seq":67,"time":1785730501679,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl index 9e39f76b11..5c68474179 100644 --- a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl +++ b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl @@ -1,32 +1,32 @@ {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"203f911d-905a-43e4-8e1c-fccfe78d445b"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"c5cc1fd3-c9d5-4b49-90b0-34da051cb71e"}]}} {"type":"turn/start","seq":1,"time":1786123401614,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1786123401614,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1786123401667,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"203f911d-905a-43e4-8e1c-fccfe78d445b"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"c5cc1fd3-c9d5-4b49-90b0-34da051cb71e"},"surfaceOp":"append"} {"type":"session/title","seq":5,"time":1786123401667,"data":{"title":"Establish a durable compaction premise","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":7,"time":1786123401669,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash","contextWindow":128000}} {"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_compaction_marker","name":"bash","argumentsDelta":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}} {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}}} {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":24,"outputTokens":6}}}} {"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6e7dfb28-f033-4dbf-9fd9-6b2f71da6906"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} +{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e29c0ce2-2826-4269-a047-a831aec5a4ee"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} {"type":"tool/call","seq":14,"time":1786123401680,"data":{"turn":1,"step":1,"callId":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}} -{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"4fe4e523-3ac9-421c-8a73-1eb033ac8b09"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"8c3b43ac-1a3b-4955-ad40-78a1d6343e2d"}},"sourceEventSeqs":[14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1786123401700,"data":{"turn":1,"step":1}} {"type":"step/start","seq":17,"time":1786123401710,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":18,"time":1786123401715,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"error","failure":{"message":"snapshot request exceeded the model context window","code":"CONTEXT_WINDOW_EXCEEDED"}}}}} {"type":"compact/start","seq":19,"time":1786123401715,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}} {"type":"compact/summary","seq":20,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","summary":[{"type":"text","text":"The request established a durable compaction premise."}],"rawOutput":[{"type":"text","text":"The request established a durable compaction premise."}],"llmStreamCall":true,"shadowedRange":{"start":4,"end":4},"shadowedSeqs":[4],"shadowedTokenCount":264,"provider":"deepseek-official","model":"deepseek-v4-flash","maxTokens":32,"usage":{"inputTokens":20,"outputTokens":4}}} -{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":""}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"b74220bf-4104-4c30-b967-e4570d5eba5d"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}} +{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":""}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"bb3fd14b-4307-4550-9aa9-3326c9ebd92f"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}} {"type":"compact/end","seq":22,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}} {"type":"assistant/chunk","seq":23,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":24,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"text-delta","index":0,"text":"COMPACTION RECOVERED"}}} {"type":"assistant/chunk","seq":25,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"COMPACTION RECOVERED"}}}} {"type":"assistant/chunk","seq":26,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":20,"outputTokens":4}}}} {"type":"assistant/chunk","seq":27,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"103034c2-e156-4808-ba2e-8e17e7107c9e"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"} +{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3712f33d-6371-454f-bf8e-707cb5c74b32"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"} {"type":"step/end","seq":29,"time":1786123401730,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":30,"time":1786123401730,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl index f4f592570a..bfd47888f9 100644 --- a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl +++ b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl @@ -1,21 +1,21 @@ {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"39f39ecc-5772-4814-8feb-46433c71becd"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"d4746327-c682-4874-b8b2-bbc81e98ea03"}]}} {"type":"turn/start","seq":1,"time":1785821457966,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821457966,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"39f39ecc-5772-4814-8feb-46433c71becd"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"a5ae9c04-0652-436f-9b5a-437a3a6ed235"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"d4746327-c682-4874-b8b2-bbc81e98ea03"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"8256fc51-472b-4d50-86ea-b50cef7b1ef6"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730504659,"data":{"title":"Exercise the six PTY tools","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730504660,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"pty-spawn","name":"terminal_open","argumentsDelta":"{\"type\":\"shell\",\"name\":\"main\"}"}}} {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}}}} {"type":"assistant/chunk","seq":12,"time":1785498587439,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730504661,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"8cad9650-de5a-4075-8aa3-1b35e67efc2d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4f71f30a-33e5-4c25-8e66-aa93c993a7e8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730504662,"data":{"turn":1,"step":1,"callId":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}} -{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"1ba1f641-0cf2-496c-895e-3982aa40b0ed"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"e5c2f208-b939-4715-984e-a4537effbbff"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730504671,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730504679,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -23,9 +23,9 @@ {"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}}}} {"type":"assistant/chunk","seq":22,"time":1785498587457,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730504680,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f5850535-906d-4a78-8518-a733ec91bbd8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b1af47f1-f976-41fc-b6f1-8ddc86f2aecd"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"tool/call","seq":25,"time":1785730504680,"data":{"turn":1,"step":2,"callId":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}} -{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"4f19b0d0-b31d-40b7-84e5-2783131cb363"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"} +{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"82baaf37-cfd3-4c59-9697-7a1b36fffa38"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"} {"type":"step/end","seq":27,"time":1785730504688,"data":{"turn":1,"step":2}} {"type":"step/start","seq":28,"time":1785730504696,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -33,9 +33,9 @@ {"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}}}} {"type":"assistant/chunk","seq":32,"time":1785498587473,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":33,"time":1785730504697,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4e0d8eed-877d-4a5b-bd92-e7c4c8f2cf23"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} +{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3de44e0c-f4b8-493e-842b-6aaa029569d0"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} {"type":"tool/call","seq":35,"time":1785730504697,"data":{"turn":1,"step":3,"callId":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}} -{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"72be20be-6022-4363-b9bb-1d9f2cce0e20"}},"sourceEventSeqs":[35],"surfaceOp":"append"} +{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"178561b8-fa84-4dc0-9bc8-d183a722d36a"}},"sourceEventSeqs":[35],"surfaceOp":"append"} {"type":"step/end","seq":37,"time":1785730504704,"data":{"turn":1,"step":3}} {"type":"step/start","seq":38,"time":1785730504712,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":39,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -43,9 +43,9 @@ {"type":"assistant/chunk","seq":41,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}}}} {"type":"assistant/chunk","seq":42,"time":1785498587489,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":43,"time":1785730504713,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"87c2857f-5da3-4050-b081-e044e207be88"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} +{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"fc614320-c742-470b-a9fc-902fd67d5cb4"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} {"type":"tool/call","seq":45,"time":1785730504713,"data":{"turn":1,"step":4,"callId":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}} -{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"c6e3ff43-809c-4bd8-ba88-da7294c3385a"}},"sourceEventSeqs":[45],"surfaceOp":"append"} +{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"3f047431-06fc-4201-b4cd-b611b6741ce3"}},"sourceEventSeqs":[45],"surfaceOp":"append"} {"type":"step/end","seq":47,"time":1785730504721,"data":{"turn":1,"step":4}} {"type":"step/start","seq":48,"time":1785730504730,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":49,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -53,9 +53,9 @@ {"type":"assistant/chunk","seq":51,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}}}} {"type":"assistant/chunk","seq":52,"time":1785498587503,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":53,"time":1785730504731,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"5b53e0d5-1c68-4988-9fab-d885a9122fe8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} +{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1a4e37b3-72fe-4cbe-bbdb-708088894423"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} {"type":"tool/call","seq":55,"time":1785730504731,"data":{"turn":1,"step":5,"callId":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}} -{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"20e3fe4d-3d9d-4771-b75d-1f287fc20048"}},"sourceEventSeqs":[55],"surfaceOp":"append"} +{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"a04c5ae2-3b36-40d0-b570-df158f27d08a"}},"sourceEventSeqs":[55],"surfaceOp":"append"} {"type":"step/end","seq":57,"time":1785730504738,"data":{"turn":1,"step":5}} {"type":"step/start","seq":58,"time":1785730504746,"data":{"turn":1,"step":6}} {"type":"assistant/chunk","seq":59,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -63,9 +63,9 @@ {"type":"assistant/chunk","seq":61,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}}}} {"type":"assistant/chunk","seq":62,"time":1785498587517,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":63,"time":1785730504747,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"178fb9bc-859c-484d-8376-096a705de30a"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"} +{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0656b271-2099-44fe-a770-834bb6c97e32"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"} {"type":"tool/call","seq":65,"time":1785730504747,"data":{"turn":1,"step":6,"callId":"pty-list","name":"terminal_list","arguments":"{}"}} -{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"584673b3-9221-4a42-b9e4-69ce1b9f4d60"}},"sourceEventSeqs":[65],"surfaceOp":"append"} +{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"8b6397b3-9ee8-431e-975a-200b14670d7e"}},"sourceEventSeqs":[65],"surfaceOp":"append"} {"type":"step/end","seq":67,"time":1785730504755,"data":{"turn":1,"step":6}} {"type":"step/start","seq":68,"time":1785730504763,"data":{"turn":1,"step":7}} {"type":"assistant/chunk","seq":69,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -73,6 +73,6 @@ {"type":"assistant/chunk","seq":71,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}} {"type":"assistant/chunk","seq":72,"time":1785498587531,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":3}}}} {"type":"assistant/chunk","seq":73,"time":1785730504764,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"23d457ba-b690-4bb9-b434-86f43c9f4da5"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"} +{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"74b620d5-2737-4b35-b9c9-2829fe59bea5"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"} {"type":"step/end","seq":75,"time":1785730504764,"data":{"turn":1,"step":7}} {"type":"turn/end","seq":76,"time":1785730504764,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/packages/bundle/base/cordis.patch.yml b/packages/bundle/base/cordis.patch.yml index 2fb5ff10d3..b9448ae6f2 100644 --- a/packages/bundle/base/cordis.patch.yml +++ b/packages/bundle/base/cordis.patch.yml @@ -311,12 +311,16 @@ toolName: subagent backgroundMode: continuable + # Fork stays one-shot: a continuable child's `report` tool and prompt + # section precede the inherited history a fork exists to reuse; one-shot + # fork children install neither, keeping the parent's request prefix. + # See .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md. - id: tool-subagent-fork name: '@deepseek-ai/dsh-tool-subagent' config: provider: fork toolName: subagent_fork - backgroundMode: continuable + backgroundMode: one-shot # Optional direct-child return channel; absent from roots and one-shot agents. - id: tool-subagent-report diff --git a/packages/subagent/subagent-fork/README.i18n.yaml b/packages/subagent/subagent-fork/README.i18n.yaml index e40cf9901f..ca23eabb84 100644 --- a/packages/subagent/subagent-fork/README.i18n.yaml +++ b/packages/subagent/subagent-fork/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/subagent/subagent-fork/README.md -README.md: 55475aee7841e91960de79887dfe9bf37afdf9da -README.zh.md: 379218970c406162309659277a4fcb2da885d9ea +README.md: 2bd72058ab7d112f8f317a33842fc4d95b719017 +README.zh.md: 40f9e34c5a8ae8c74de2ae0c9e4676866f0bd083 diff --git a/packages/subagent/subagent-fork/README.md b/packages/subagent/subagent-fork/README.md index 55475aee78..2bd72058ab 100644 --- a/packages/subagent/subagent-fork/README.md +++ b/packages/subagent/subagent-fork/README.md @@ -39,7 +39,7 @@ Forking duplicates retained completed history into separate child requests; the #### KV Cache effect -The child may reuse the inherited byte-identical prefix under the same provider and model. Persona, tool-filter, generated-SDK, or route changes may invalidate reuse before inherited history; later child history is append-only. +The child may reuse the inherited byte-identical prefix under the same provider and model. Persona, tool-filter, generated-SDK, or route changes may invalidate reuse before inherited history; later child history is append-only. Shipped compositions therefore bind this provider to `backgroundMode: one-shot`, because a continuable child additionally carries the child-scoped `report` tool and its prompt section — deltas that precede the inherited history and so invalidate all of it ([the fork-one-shot Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)). ### Parent tool result, indirectly @@ -58,3 +58,4 @@ Append-only; newly visible content follows the reusable request prefix and does ## Known Limitations and Deferred Work - **The seed is a one-time snapshot** — the child sees the parent's completed turns as of the fork and nothing the parent logs afterwards; there is no live context sharing. +- **No shipped composition creates a continuable fork child** — `prepareContinuable` remains implemented and the seam accepts it, but every shipped `cordis.yml` sets `backgroundMode: one-shot` on the fork delegation tool, so the provider's continuable path has no production caller. Reopening it requires the child's system prompt and tool schemas to match the parent's byte for byte, which the [`report` return channel](../tool-subagent-report/README.md) currently prevents. Rationale and the reintroduction condition: [the fork-one-shot Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md). diff --git a/packages/subagent/subagent-fork/README.zh.md b/packages/subagent/subagent-fork/README.zh.md index 379218970c..40f9e34c5a 100644 --- a/packages/subagent/subagent-fork/README.zh.md +++ b/packages/subagent/subagent-fork/README.zh.md @@ -39,7 +39,7 @@ fork 会把保留的已完成历史复制到独立的子 agent 请求中；随 #### KV Cache 影响 -在提供方和模型相同的前提下，子 agent 可以复用继承的逐字节相同前缀。persona、工具过滤、生成 SDK 或路由变化可能在继承历史之前使复用失效；后续子 agent 历史仅追加。 +在提供方和模型相同的前提下，子 agent 可以复用继承的逐字节相同前缀。persona、工具过滤、生成 SDK 或路由变化可能在继承历史之前使复用失效；后续子 agent 历史仅追加。因此随附组合把本提供方绑定为 `backgroundMode: one-shot`：可继续子 agent 还会额外携带作用域局部的 `report` 工具及其提示词 section，而这些增量位于继承历史之前，会使继承历史整体失效（见 [fork 保持 one-shot 的 Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)）。 ### 父 agent 工具结果（间接） @@ -58,3 +58,4 @@ fork 会把保留的已完成历史复制到独立的子 agent 请求中；随 ## 已知限制与暂缓事项 - **初始内容是一次性快照**：子 agent 只能看到 fork 时父 agent 已完成的轮次，看不到父 agent 此后记录的任何内容；不会实时共享上下文。 +- **没有任何随附组合会创建可继续的 fork 子 agent**：`prepareContinuable` 仍然实现完好，seam 也接受它，但每份随附的 `cordis.yml` 都在 fork 委派工具上设置 `backgroundMode: one-shot`，因此该提供方的可继续路径没有生产调用方。重新开放它需要子 agent 的系统提示词与工具 schema 与父 agent 逐字节一致，而这一点目前被 [`report` 返回通道](../tool-subagent-report/README.md)阻止。理由与重新开放条件见 [fork 保持 one-shot 的 Agent Note](../../../.agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md)。 diff --git a/packages/subagent/subagent-fork/src/index.ts b/packages/subagent/subagent-fork/src/index.ts index 0a3ea21aa2..b4d3d51144 100644 --- a/packages/subagent/subagent-fork/src/index.ts +++ b/packages/subagent/subagent-fork/src/index.ts @@ -74,6 +74,12 @@ class ForkProvider implements SubagentProvider { }) } + // TODO(fork-continuable-prefix-reuse): no shipped composition calls this — + // they bind fork to `backgroundMode: one-shot` because a continuable child's + // `report` tool and prompt section precede the inherited history, defeating + // the prefix reuse a fork exists for. Reopening needs a byte-identical child + // system prompt and tool schemas; see issue #2124 and + // .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md. prepareContinuable(request: ContinuableCreateRequest): Promise { // The fork prefix is captured ONCE, at creation: it becomes part of the // child's own durable transcript, so a later cold resume replays that diff --git a/packages/subagent/tool-subagent-report/README.i18n.yaml b/packages/subagent/tool-subagent-report/README.i18n.yaml index 8d64d7e87f..13a892fa8a 100644 --- a/packages/subagent/tool-subagent-report/README.i18n.yaml +++ b/packages/subagent/tool-subagent-report/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/subagent/tool-subagent-report/README.md -README.md: 1772b72d83563e89ed0e3814e515b06a2b36fe0f -README.zh.md: 94101fbc9611a04c56ed3a9171889499c9144c21 +README.md: b38b6b541d9c03174de1515df7a0babe5da055b4 +README.zh.md: 0c5c063eded1e9057ee8886f03dd6494dd9eeed8 diff --git a/packages/subagent/tool-subagent-report/README.md b/packages/subagent/tool-subagent-report/README.md index 1772b72d83..b38b6b541d 100644 --- a/packages/subagent/tool-subagent-report/README.md +++ b/packages/subagent/tool-subagent-report/README.md @@ -2,15 +2,15 @@ English | [中文](README.zh.md) -The optional child-scoped `report` tool is a thin adapter over `ctx.subagents.reportFrom()`. It gives every continuable in-process child a return channel to the Agent that started it. The package registers a continuable-child setup contribution instead of a global tool, so `report` exists only inside those children. Roots, one-shot subagents, remote subagent providers, sibling scopes, and agentless tool execution never present or execute it. Installing this package grants only that child-scoped capability; the parent-to-child direction remains the independent [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md), and continuable mode depends on neither package. +The optional child-scoped `report` tool is a thin adapter over `ctx.subagents.reportFrom()`. It gives every continuable in-process child a return channel to the Agent that started it, and installs the prompt section that instructs the child to use it. The package registers a continuable-child setup contribution instead of a global tool, so the tool and its guidance exist only inside those children. Roots, one-shot subagents, remote subagent providers, sibling scopes, and agentless tool execution never present or execute it. Installing this package grants only that child-scoped capability; the parent-to-child direction remains the independent [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md), and continuable mode depends on neither package. -A child may call `report` zero or many times in one turn. A successful call neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. The tool accepts no recipient: `exec.agent` is the sender's exact live Agent and the authority credential, and the service derives the sole recipient from that child's durable `parentSession`. Success returns the stable `MessageId` of the parent-accepted message, not a read receipt, an inbox-occurrence id, a parent-log acknowledgement, a turn-completion receipt, or a persistence flush. A parent absent from the registry fails the call with `direct parent is not live; report was not delivered` — registry presence governs parent resolution, and a registered parent already in host-owned disposal still accepts while its log admits appends. The service performs no injection, parent cold resume, or offline mailbox write; the durable child transcript remains the recovery source, and a failed tool call does not prove non-delivery (a later `tools/post-execute` veto can fail a call whose report was already accepted). +The child-scoped `tool:report` prompt section instructs the child to call `report` once before finishing, with a self-contained answer, and earlier whenever a partial finding changes what the parent should do next. The instruction is guidance, not enforcement: the mechanism still accepts zero or many calls in one turn, and no runtime path rejects a child that never reports. A successful call neither concludes the turn, settles the Activation, nor prevents later parent follow-ups, and finishing a turn never reports automatically. The tool accepts no recipient: `exec.agent` is the sender's exact live Agent and the authority credential, and the service derives the sole recipient from that child's durable `parentSession`. Success returns the stable `MessageId` of the parent-accepted message, not a read receipt, an inbox-occurrence id, a parent-log acknowledgement, a turn-completion receipt, or a persistence flush. A parent absent from the registry fails the call with `direct parent is not live; report was not delivered` — registry presence governs parent resolution, and a registered parent already in host-owned disposal still accepts while its log admits appends. The service performs no injection, parent cold resume, or offline mailbox write; the durable child transcript remains the recovery source, and a failed tool call does not prove non-delivery (a later `tools/post-execute` veto can fail a call whose report was already accepted). -`reportDelivery` selects parent scheduling for every accepted report. `quiet` (the default) uses `parent.inject()`, adding model-facing context without starting a parent model request: an idle parent's append completes before the call returns, while a report reaching an admitting or running parent stages for the next safe log position. `wakeup` uses `parent.followup()`, creating exactly one ordinary later parent turn and waking a parked parent driver; it never steers an open turn. This is deployment scheduling policy, so the model-facing schema cannot select or override it per call. +`reportDelivery` selects parent scheduling for every accepted report. `wakeup` (the default) uses `parent.followup()`, creating exactly one ordinary later parent turn and waking a parked parent driver; it never steers an open turn. It is the default because a parent that already parked has no other reason to look, so quiet delivery would leave an accepted report unread until something unrelated woke it. `quiet` uses `parent.inject()`, adding model-facing context without starting a parent model request: an idle parent's append completes before the call returns, while a report reaching an admitting or running parent stages for the next safe log position. This is deployment scheduling policy, so the model-facing schema cannot select or override it per call. Scope-local registration deliberately survives the child's global `toolFilter`, so a delegation allow-list cannot remove the only return channel. A deployment that requires a child with no return channel omits this package. -The contribution body is exported as `installReportTool(childCtx, ctx, delivery)` so inspection consumers can install `report` into a minted child scope. The generated tool catalog uses that path because the global registry cannot expose a scope-local schema. Production composition still enters through `apply()`; the subagent seam's contribution registry remains private. +The contribution body is exported as `installReportTool(childCtx, ctx, delivery)` so inspection consumers can install `report` and its guidance into a minted child scope, and returns the one disposer revoking both. The generated tool catalog uses that path because the global registry cannot expose a scope-local schema. Production composition still enters through `apply()`; the subagent seam's contribution registry remains private. ## Model Experience @@ -18,15 +18,15 @@ The contribution body is exported as `installReportTool(childCtx, ctx, delivery) #### What the model sees -The generated [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report): one required `output` string. Its description states that reporting is explicit and repeatable, reaches only the Agent that started the child, and does not end the turn. It carries no recipient or delivery-mode parameter. +The generated [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report): one required `output` string. Its description states that the child must report once before finishing, that reporting reaches only the Agent that started the child, and that it does not end the turn. It carries no recipient or delivery-mode parameter. The separate `tool:report` prompt section repeats the obligation outside the schema, where a child that ignores tool descriptions still reads it. #### Token effect -Fixed schema cost per continuable-child request, and none in any other Agent's requests. +Fixed schema and prompt-section cost per continuable-child request, and none in any other Agent's requests. #### KV Cache effect -Prefix-stable within a child; the schema does not change at runtime. Removing the package revokes the schema from resident children, which changes their next request prefix. +Prefix-stable within a child; neither the schema nor the section changes at runtime. Removing the package revokes both from resident children, which changes their next request prefix. ### Report result @@ -36,7 +36,7 @@ Prefix-stable within a child; the schema does not change at runtime. Removing th #### Token effect -One short acknowledgement per call in the reporting child. The reported content is additionally billed to the parent: quiet delivery adds it to the parent's next request, while waking delivery makes it the sole ordinary message of one new parent turn. +One short acknowledgement per call in the reporting child. The reported content is additionally billed to the parent: waking delivery makes it the sole ordinary message of one new parent turn, while quiet delivery adds it to the parent's next request. #### KV Cache effect @@ -61,6 +61,6 @@ Append-only; the report follows the parent's reusable request prefix. Waking del - **A parent whose host-owned disposal already started can still accept** — `AgentHandle.dispose()` cancels, awaits quiescence, and only then unwinds the scope and leaves the registry; it exposes no signal for "disposal started." A report accepted in that window is appended to the parent's transcript, but that parent will not act on it in this process. A continuation-manager-owned parent rejects forest teardown through the manager's admission boundary. - **Acceptance is weaker than durable delivery** — there is no durable mailbox, idempotency key, delivery receipt, retry protocol, or exactly-once claim. A process failure after one side recorded acceptance leaves the outcome ambiguous, and an external retry may duplicate the report. - **A staged quiet report is not immediately reconstructable** — acceptance returns its stable `MessageId`, but the parent Session reconstructs the framed content only after pending context reaches its ordinary log boundary. -- **Granting waits for the next Activation; revocation is immediate** — installing this package after a child becomes resident grants `report` only on that child's next Activation, while removing the package revokes the schema from resident children immediately. +- **Granting waits for the next Activation; revocation is immediate** — installing this package after a child becomes resident grants `report` and its guidance only on that child's next Activation, while removing the package revokes both from resident children immediately. - **Nested reporting reaches exactly one edge upward** — a grandchild reports to its direct child parent, never to the top-level coordinator, which must explicitly report a derived update later. -- **No rate limiting** — `wakeup` mode can amplify model work when nested children report frequently; the deployment owns that choice by selecting the mode. +- **No rate limiting** — the default `wakeup` mode can amplify model work when nested children report frequently; a deployment that accepts unread reports over that amplification selects `quiet`. diff --git a/packages/subagent/tool-subagent-report/README.zh.md b/packages/subagent/tool-subagent-report/README.zh.md index 94101fbc96..0c5c063ede 100644 --- a/packages/subagent/tool-subagent-report/README.zh.md +++ b/packages/subagent/tool-subagent-report/README.zh.md @@ -2,15 +2,15 @@ [English](README.md) | 中文 -可选的子级作用域 `report` 工具是 `ctx.subagents.reportFrom()` 之上的轻量适配器。它为每个可继续的进程内子级提供一条返回通道，指向启动该子级的 Agent（智能体）。本包注册的是可继续子级设置贡献，而不是全局工具，因此 `report` 只存在于这些子级内部。根 Agent、一次性 subagent、远程 subagent 提供方、同级作用域以及不关联 Agent 的工具执行都不会提供或执行它。安装本包只授予这项子级作用域能力；父到子方向仍由独立的 [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md) 负责，可继续模式不依赖这两个包中的任一个。 +可选的子级作用域 `report` 工具是 `ctx.subagents.reportFrom()` 之上的轻量适配器。它为每个可继续的进程内子级提供一条返回通道，指向启动该子级的 Agent（智能体），并安装指示子级使用该通道的提示词 section。本包注册的是可继续子级设置贡献，而不是全局工具，因此该工具及其指引只存在于这些子级内部。根 Agent、一次性 subagent、远程 subagent 提供方、同级作用域以及不关联 Agent 的工具执行都不会提供或执行它。安装本包只授予这项子级作用域能力；父到子方向仍由独立的 [`@deepseek-ai/dsh-tool-subagent-control`](../tool-subagent-control/README.md) 负责，可继续模式不依赖这两个包中的任一个。 -子级可以在一个轮次中调用 `report` 零次或多次。调用成功既不会结束轮次或结算 Activation，也不会阻止父级后续消息；轮次结束也绝不会自动上报。该工具不接受接收方参数：`exec.agent` 是发送方确切在线的 Agent，也是权限凭据；服务根据该子级持久化的 `parentSession` 推导唯一接收方。成功时返回父级已接受消息的稳定 `MessageId`，不表示已读回执、inbox 中该次出现的 id、父级日志确认、轮次完成回执或持久化刷盘。父级解析由注册表中的存在性决定：父级不在注册表时，调用失败并返回 `direct parent is not live; report was not delivered`；已开始由宿主管理的 dispose（资源释放）但仍在注册表中的父级在其日志仍接受追加时仍会接受。服务不会执行注入、父级冷恢复或离线 mailbox 写入；持久化子级 transcript（文本记录）仍是恢复依据，且工具调用失败不能证明未送达（后续 `tools/post-execute` 否决可能让报告已被接受的调用以失败结束）。 +子级作用域的 `tool:report` 提示词 section 要求子级在结束前调用一次 `report` 并给出自足的答案，并在部分发现会改变父级下一步动作时提前上报。该指令是引导而非强制：机制本身仍接受一个轮次中调用零次或多次，也没有任何运行时路径会拒绝从不上报的子级。调用成功既不会结束轮次或结算 Activation，也不会阻止父级后续消息；轮次结束也绝不会自动上报。该工具不接受接收方参数：`exec.agent` 是发送方确切在线的 Agent，也是权限凭据；服务根据该子级持久化的 `parentSession` 推导唯一接收方。成功时返回父级已接受消息的稳定 `MessageId`，不表示已读回执、inbox 中该次出现的 id、父级日志确认、轮次完成回执或持久化刷盘。父级解析由注册表中的存在性决定：父级不在注册表时，调用失败并返回 `direct parent is not live; report was not delivered`；已开始由宿主管理的 dispose（资源释放）但仍在注册表中的父级在其日志仍接受追加时仍会接受。服务不会执行注入、父级冷恢复或离线 mailbox 写入；持久化子级 transcript（文本记录）仍是恢复依据，且工具调用失败不能证明未送达（后续 `tools/post-execute` 否决可能让报告已被接受的调用以失败结束）。 -`reportDelivery` 为每条已接受的报告选择父级调度方式。`quiet`（默认值）使用 `parent.inject()`，在不启动父级模型请求的情况下添加面向模型的上下文：父级空闲时，追加操作会在调用返回前完成；报告到达正在准入或运行的父级时，则会暂存到下一个安全日志位置。`wakeup` 使用 `parent.followup()`，恰好创建一个普通的后续父级轮次，并唤醒停驻的父级驱动；它绝不会对正在运行的轮次进行 steering（中途引导）。这是部署调度策略，因此面向模型的 schema 不能在单次调用中选择或覆盖该策略。 +`reportDelivery` 为每条已接受的报告选择父级调度方式。`wakeup`（默认值）使用 `parent.followup()`，恰好创建一个普通的后续父级轮次，并唤醒停驻的父级驱动；它绝不会对正在运行的轮次进行 steering（中途引导）。之所以作为默认值：已经停驻的父级没有别的理由再去查看，静默投递会让一条已被接受的报告一直无人阅读，直到别的事件把父级唤醒。`quiet` 使用 `parent.inject()`，在不启动父级模型请求的情况下添加面向模型的上下文：父级空闲时，追加操作会在调用返回前完成；报告到达正在准入或运行的父级时，则会暂存到下一个安全日志位置。这是部署调度策略，因此面向模型的 schema 不能在单次调用中选择或覆盖该策略。 作用域局部注册有意不受子级全局 `toolFilter` 影响，因此委派允许列表无法移除唯一的返回通道。需要子级不具备返回通道的部署应省略本包。 -贡献体以 `installReportTool(childCtx, ctx, delivery)` 导出，以便检查类消费方把 `report` 安装到新创建的子级作用域中。全局注册表无法公开作用域局部 schema，因此生成的工具目录会使用这条路径。生产组合仍通过 `apply()` 进入；subagent seam 的贡献注册表保持私有。 +贡献体以 `installReportTool(childCtx, ctx, delivery)` 导出，以便检查类消费方把 `report` 及其指引安装到新创建的子级作用域中，并返回同时撤销两者的唯一 disposer。全局注册表无法公开作用域局部 schema，因此生成的工具目录会使用这条路径。生产组合仍通过 `apply()` 进入；subagent seam 的贡献注册表保持私有。 ## 模型体验 @@ -18,15 +18,15 @@ #### 模型看到的内容 -已生成的 [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report)：包含一个必填 `output` 字符串。其描述说明上报需要显式调用且可以重复，只会到达启动该子级的 Agent，并且不会结束轮次。它不包含接收方或投递模式参数。 +已生成的 [`report` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent-report)：包含一个必填 `output` 字符串。其描述说明子级必须在结束前上报一次，上报只会到达启动该子级的 Agent，并且不会结束轮次。它不包含接收方或投递模式参数。独立的 `tool:report` 提示词 section 在 schema 之外重申该义务，使忽略工具描述的子级仍能读到。 #### Token 影响 -每个可继续子级请求支付固定的 schema 成本，其他任何 Agent 的请求均无此成本。 +每个可继续子级请求支付固定的 schema 与提示词 section 成本，其他任何 Agent 的请求均无此成本。 #### KV Cache 影响 -子级中的前缀保持稳定；schema 不会在运行时改变。移除本包会从驻留子级中撤销该 schema，从而改变其下一次请求前缀。 +子级中的前缀保持稳定；schema 与该 section 都不会在运行时改变。移除本包会从驻留子级中撤销两者，从而改变其下一次请求前缀。 ### 上报结果 @@ -36,7 +36,7 @@ #### Token 影响 -每次调用都会在执行上报的子级中产生一条简短确认消息。父级还会为上报内容支付 token 成本：静默投递会把内容加入父级的下一次请求，唤醒投递则会使该内容成为一个新父级轮次中唯一的普通消息。 +每次调用都会在执行上报的子级中产生一条简短确认消息。父级还会为上报内容支付 token 成本：唤醒投递会使该内容成为一个新父级轮次中唯一的普通消息，静默投递则把内容加入父级的下一次请求。 #### KV Cache 影响 @@ -61,6 +61,6 @@ - **父级可能在宿主启动 dispose 后继续接受报告**：`AgentHandle.dispose()` 会先取消并等待完全停稳，然后才撤销作用域并离开注册表；它不公开「dispose 已开始」信号。在该窗口内接受的报告会追加到父级 transcript，但该父级不会在本进程中处理它。对于由继续执行管理器拥有的父级，管理器的准入边界会在整片森林拆卸期间拒绝该上报。 - **接受弱于持久投递**：没有持久化 mailbox、幂等键、投递回执、重试协议，也不保证恰好一次。任一侧记录接受后若进程失败，结果都不明确；外部重试可能产生重复上报。 - **暂存的静默报告无法立即重建**：接受时会返回其稳定 `MessageId`，但只有当待处理上下文到达普通日志边界后，父级 Session 才能重建带前缀的内容。 -- **授权须等到下一个 Activation，撤销则立即生效**：子级驻留后再安装本包，只会在该子级的下一个 Activation 中授予 `report`；移除本包则会立即从驻留子级撤销该 schema。 +- **授权须等到下一个 Activation，撤销则立即生效**：子级驻留后再安装本包，只会在该子级的下一个 Activation 中授予 `report` 及其指引；移除本包则会立即从驻留子级撤销两者。 - **嵌套上报只向上到达一条直接边**：孙级只向作为其直接父级的子级上报，不会直接到达顶层协调器；该直接父级必须随后显式发出一条衍生更新。 -- **没有速率限制**：嵌套子级频繁上报时，`wakeup` 模式会放大模型工作量；部署通过选择模式自行承担这一取舍。 +- **没有速率限制**：嵌套子级频繁上报时，默认的 `wakeup` 模式会放大模型工作量；宁可接受报告无人阅读也要避免这种放大的部署应选择 `quiet`。 diff --git a/packages/subagent/tool-subagent-report/package.json b/packages/subagent/tool-subagent-report/package.json index f9dcd90588..8c2eb5d999 100644 --- a/packages/subagent/tool-subagent-report/package.json +++ b/packages/subagent/tool-subagent-report/package.json @@ -35,6 +35,7 @@ "@deepseek-ai/dsh-invariants": "workspace:^", "@deepseek-ai/dsh-llm": "workspace:^", "@deepseek-ai/dsh-subagent": "workspace:^", + "@deepseek-ai/dsh-system-prompt": "workspace:^", "@deepseek-ai/dsh-tools": "workspace:^", "@deepseek-ai/cordis": "workspace:^" }, @@ -52,6 +53,7 @@ "@deepseek-ai/dsh-session-persistence-jsonl": "workspace:^", "@deepseek-ai/dsh-subagent": "workspace:^", "@deepseek-ai/dsh-subagent-spawn": "workspace:^", + "@deepseek-ai/dsh-system-prompt": "workspace:^", "@deepseek-ai/dsh-tool-subagent-control": "workspace:^", "@deepseek-ai/dsh-tools": "workspace:^", "@deepseek-ai/cordis": "workspace:^" diff --git a/packages/subagent/tool-subagent-report/src/index.ts b/packages/subagent/tool-subagent-report/src/index.ts index b33a5a29ab..6f2a6b3c6a 100644 --- a/packages/subagent/tool-subagent-report/src/index.ts +++ b/packages/subagent/tool-subagent-report/src/index.ts @@ -1,7 +1,7 @@ /** - * The child-scoped `report` tool, installed into every continuable in-process - * child's unpublished context. Roots, one-shot children, remote providers, and - * agentless executions never see the registration. + * The child-scoped `report` tool and its usage guidance, installed into every + * continuable in-process child's unpublished context. Roots, one-shot children, + * remote providers, and agentless executions never see the registration. * * @module @deepseek-ai/dsh-tool-subagent-report */ @@ -11,86 +11,131 @@ import z from '@deepseek-ai/schemastery' import type { Agent } from '@deepseek-ai/dsh-agent' import type { ContentBlock } from '@deepseek-ai/dsh-llm' import type { SubagentReportDelivery } from '@deepseek-ai/dsh-subagent' +import type {} from '@deepseek-ai/dsh-system-prompt' import { defineTool } from '@deepseek-ai/dsh-tools' export const name = 'tool-subagent-report' -// The contribution registers only through childCtx.tools, but declaring tools -// makes Loader ordering fail at load instead of the next child materialization. -export const inject = ['subagents', 'tools'] +// The contribution registers only through childCtx.tools and +// childCtx.systemPrompt, but declaring both services makes Loader ordering fail +// at load instead of at the next child materialization. +export const inject = ['subagents', 'tools', 'systemPrompt'] + +/** Guidance order after every per-tool section a continuable child can carry. */ +const REPORT_SECTION_ORDER = 117 /** Config: how accepted reports are scheduled on the parent. */ export interface Config { /** - * Parent scheduling (default `quiet`). `quiet` adds context without waking; - * `wakeup` creates one ordinary later parent turn. + * Parent scheduling (default `wakeup`). `wakeup` creates one ordinary later + * parent turn; `quiet` adds context without waking, so a parked parent learns + * of the report only when something else wakes it. */ reportDelivery?: SubagentReportDelivery } export const Config: z = z.object({ - reportDelivery: z.union(['quiet', 'wakeup'] as const).default('quiet'), + reportDelivery: z.union(['quiet', 'wakeup'] as const).default('wakeup'), }) /** - * Install `report` into one continuable child's scope. - * @param childCtx - child-scoped context receiving the tool. + * Install `report` and its usage guidance into one continuable child's scope. + * Both registrations are owned by that scope and are therefore invisible to the + * child's parent and siblings. + * @param childCtx - child-scoped context receiving the tool and the guidance. * @param ctx - service context used for delivery. * @param delivery - resolved deployment scheduling policy. - * @returns disposer for this one registration. + * @returns disposer that attempts both child registrations before reporting cleanup failures. */ export function installReportTool( childCtx: Context, ctx: Context, delivery: SubagentReportDelivery, ): () => void { - return childCtx.tools.register(defineTool({ - name: 'report', - description: - 'Report selected content to the agent that started you. Call this zero or more times for progress, ' - + 'findings, or a final answer. Reporting does not end your turn or finish your work, and only your ' - + 'direct parent receives it. A failed call may still have arrived, so do not blindly repeat it.', - parameters: { - output: { - type: 'string', - required: true, - description: 'Self-contained content for your parent; it does not see your private work.', - }, - }, - output: { - schema: { - type: 'object', - additionalProperties: false, - properties: { - messageId: { type: 'string', required: true }, + const disposeSection = childCtx.systemPrompt.section({ + name: 'tool:report', + order: REPORT_SECTION_ORDER, + text: 'Deliver your result with the report tool before you finish: call it once with a self-contained ' + + 'answer. The agent that started you shares your workspace but does not automatically receive your ' + + 'transcript, tool output, or reasoning, so a closing remark such as "done" leaves it nothing it can ' + + 'use. Report earlier as well whenever a partial finding changes what that agent should do next; ' + + 'reporting never ends your turn.', + }) + let disposeTool: () => void + try { + disposeTool = childCtx.tools.register(defineTool({ + name: 'report', + description: + 'Report selected content to the agent that started you. Call this once before you finish, with a ' + + 'self-contained final result, and earlier for progress or findings that change what that agent does ' + + 'next. That agent shares your workspace but does not automatically receive your transcript, tool ' + + 'output, or reasoning, so finishing your work is not itself a result. Reporting does not end your ' + + 'turn or finish your work, and only your direct parent receives it. A failed call may still have ' + + 'arrived, so do not blindly repeat it.', + parameters: { + output: { + type: 'string', + required: true, + description: 'Actionable content for your parent; summarize conclusions and reference relevant shared paths.', }, }, - render: (_args, value) => [{ - type: 'text', - text: `report accepted by the agent that started you as message ${value.messageId}`, - }], - }, - async execute(args, exec) { - const content: ContentBlock[] = [{ type: 'text', text: args.output }] - // Scope-local resolution guarantees an Agent. The service still verifies - // its exact live Activation identity at the authority boundary. - const messageId = await ctx.subagents.reportFrom(exec.agent as Agent, content, { - delivery, - signal: exec.signal, - }) - return { messageId } - }, - })) + output: { + schema: { + type: 'object', + additionalProperties: false, + properties: { + messageId: { type: 'string', required: true }, + }, + }, + render: (_args, value) => [{ + type: 'text', + text: `report accepted by the agent that started you as message ${value.messageId}`, + }], + }, + async execute(args, exec) { + const content: ContentBlock[] = [{ type: 'text', text: args.output }] + // Scope-local resolution guarantees an Agent. The service still verifies + // its exact live Activation identity at the authority boundary. + const messageId = await ctx.subagents.reportFrom(exec.agent as Agent, content, { + delivery, + signal: exec.signal, + }) + return { messageId } + }, + })) + } catch (error: unknown) { + try { + disposeSection() + } catch (rollbackError: unknown) { + throw new AggregateError( + [error, rollbackError], + 'failed to register the report tool and roll back its prompt guidance', + ) + } + throw error + } + return () => { + const failures: unknown[] = [] + for (const dispose of [disposeTool, disposeSection]) { + try { + dispose() + } catch (error: unknown) { + failures.push(error) + } + } + if (failures.length > 0) { + throw new AggregateError(failures, 'failed to revoke report tool and prompt registrations') + } + } } /** * Register the continuable-child contribution. - * @param ctx - context carrying tools and the subagent service. + * @param ctx - context carrying tools, the system prompt, and the subagent service. * @param config - deployment scheduling policy. */ export function apply(ctx: Context, config: Config = {}): void { - // Config() applies the schema default ('quiet') at runtime; the schemastery - // return type keeps the input's optional shape, so assert the resolved - // shape here — no runtime fallback exists or is wanted. + // Config() applies the schema default at runtime; the schemastery return + // type keeps the input's optional shape, so assert the resolved one. const { reportDelivery } = Config(config) as { reportDelivery: SubagentReportDelivery } ctx.subagents.registerContinuableSetup(childCtx => installReportTool(childCtx, ctx, reportDelivery)) diff --git a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts index f33901c8c7..53a84294cf 100644 --- a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts +++ b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts @@ -4,6 +4,7 @@ import { tmpdir } from 'node:os' import { join } from 'node:path' import { Context } from '@deepseek-ai/cordis' import type { Agent } from '@deepseek-ai/dsh-agent' +import { assembleContextFor } from '@deepseek-ai/dsh-agent' import AgentLoop from '@deepseek-ai/dsh-agent-loop' import { mountAgentLoopTestDependencies } from '@deepseek-ai/dsh-agent-loop-testkit' import { CallId, LlmAdapter } from '@deepseek-ai/dsh-llm' @@ -96,6 +97,17 @@ function callReport(ctx: Context, child: Agent, output: string, signal = testSig }) } +/** Occupy the child-local report name to force installation rollback. */ +function registerReportConflict(child: Agent): () => void { + return child.ctx.tools.register({ + name: 'report', + description: 'conflicting report fixture', + parameters: { type: 'object', properties: {} }, + output: { schema: { type: 'object', properties: {} }, render: () => [] }, + execute: () => Promise.resolve({}), + }) +} + /** Reports already visible or still pending in one Agent. */ function reports(agent: Agent): { id: string; text: string; sender: string }[] { const visible = agent.session.events.flatMap(event => event.type === 'user/message' ? [event.data] : []) @@ -113,6 +125,12 @@ function renderedText(result: { content: { type: string; text?: string }[] }): s return result.content.flatMap(block => block.type === 'text' ? [block.text ?? ''] : []).join('') } +/** The prompt sections one agent's scope assembles, by name. */ +async function sectionNames(ctx: Context, agent: Agent): Promise { + const assembly = await ctx.systemPrompt.assemble(assembleContextFor(agent)) + return assembly.sections.map(section => section.name) +} + describe('dsh-tool-subagent-report', () => { it('registers report only in continuable child scopes', async () => { const { ctx, parent } = await setup() @@ -309,16 +327,101 @@ describe('dsh-tool-subagent-report', () => { const { ctx, parent, fiber } = await setup() const { child } = await startChild(ctx, parent) expect(ctx.tools.schemas(child).map(schema => schema.name)).toContain('report') + expect(await sectionNames(ctx, child)).toContain('tool:report') await fiber?.dispose() expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report') + expect(await sectionNames(ctx, child)).not.toContain('tool:report') expect((await callReport(ctx, child, 'revoked')).isError).toBe(true) const late = await ctx.plugin(tool, { reportDelivery: 'quiet' }) expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report') + expect(await sectionNames(ctx, child)).not.toContain('tool:report') await late.dispose() }) + it('rolls back prompt guidance when tool registration fails', async () => { + const { ctx, parent } = await setup({ load: false }) + const { child } = await startChild(ctx, parent) + const disposeConflict = registerReportConflict(child) + + expect(() => tool.installReportTool(child.ctx, ctx, 'quiet')).toThrow(/already registered in this scope/) + expect(await sectionNames(ctx, child)).not.toContain('tool:report') + disposeConflict() + }) + + it('aggregates a registration failure with a prompt rollback failure', async () => { + const { ctx, parent } = await setup({ load: false }) + const { child } = await startChild(ctx, parent) + const disposeConflict = registerReportConflict(child) + const rollbackFailure = new Error('prompt rollback listener failed') + let promptChanges = 0 + const off = ctx.on('system-prompt/change', () => { + promptChanges++ + if (promptChanges === 2) throw rollbackFailure + }) + + let failure: unknown + try { + tool.installReportTool(child.ctx, ctx, 'quiet') + } catch (error: unknown) { + failure = error + } + off() + + expect(failure).toBeInstanceOf(AggregateError) + if (!(failure instanceof AggregateError)) throw new Error('expected aggregate installation failure') + expect(failure.errors).toHaveLength(2) + expect(String(failure.errors[0])).toContain('already registered in this scope') + expect(failure.errors[1]).toBe(rollbackFailure) + expect(await sectionNames(ctx, child)).not.toContain('tool:report') + disposeConflict() + }) + + it('attempts both revocations and aggregates change-listener failures', async () => { + const { ctx, parent } = await setup({ load: false }) + const { child } = await startChild(ctx, parent) + const dispose = tool.installReportTool(child.ctx, ctx, 'quiet') + const toolFailure = new Error('tool removal listener failed') + const promptFailure = new Error('prompt removal listener failed') + const offTool = ctx.on('tools/change', () => { throw toolFailure }) + const offPrompt = ctx.on('system-prompt/change', () => { throw promptFailure }) + + let failure: unknown + try { + dispose() + } catch (error: unknown) { + failure = error + } + offPrompt() + offTool() + + expect(failure).toBeInstanceOf(AggregateError) + if (!(failure instanceof AggregateError)) throw new Error('expected aggregate revocation failure') + expect(failure.errors).toEqual([toolFailure, promptFailure]) + expect(ctx.tools.schemas(child).map(schema => schema.name)).not.toContain('report') + expect(await sectionNames(ctx, child)).not.toContain('tool:report') + }) + + it('scopes the report guidance to the child that owns it', async () => { + const { ctx, parent } = await setup() + const { child } = await startChild(ctx, parent, 'first child') + const { child: sibling } = await startChild(ctx, parent, 'second child') + + const assembly = await ctx.systemPrompt.assemble(assembleContextFor(child)) + const guidance = assembly.sections.find(section => section.name === 'tool:report') + // Pins the model-visible instruction that makes the return channel a + // contract rather than an option the child may quietly skip. + expect(guidance?.text).toContain('Deliver your result with the report tool before you finish') + expect(guidance?.text).toContain('reporting never ends your turn') + + expect(await sectionNames(ctx, parent)).not.toContain('tool:report') + // A sibling installs its own copy; neither child can observe the other's. + expect(await sectionNames(ctx, sibling)).toContain('tool:report') + expect((await ctx.systemPrompt.assemble()).sections.map(section => section.name)) + .not.toContain('tool:report') + }) + it('rolls back materialization when a setup contribution revokes itself', async () => { const { ctx, parent } = await setup({ load: false }) const self: { revoke?: () => void } = {} @@ -405,10 +508,29 @@ describe('dsh-tool-subagent-report', () => { it('keeps the namespace plugin shape and validates its default', () => { expect('default' in tool).toBe(false) expect(tool.name).toBe('tool-subagent-report') - expect(tool.inject).toEqual(['subagents', 'tools']) - expect(tool.Config({}).reportDelivery).toBe('quiet') + expect(tool.inject).toEqual(['subagents', 'tools', 'systemPrompt']) + // Waking is the default because a report that never wakes its parent + // cannot deliver a result to an agent that already parked. + expect(tool.Config({}).reportDelivery).toBe('wakeup') expect(() => tool.Config({ reportDelivery: 'shout' } as never)).toThrow() }) + + it('wakes the parent under the default configuration', async () => { + const { ctx, parent, adapter } = await setup({ config: {} }) + const { child } = await startChild(ctx, parent) + const enqueues: string[] = [] + ctx.on('agent/inbox/inserted', ({ agent, message }) => { + if (agent === parent) { + enqueues.push(agent.inbox.nextTurn.some(queued => queued.id === message.id) ? 'queued' : 'steering') + } + }) + + expect((await callReport(ctx, child, 'DEFAULT_WAKES')).isError).toBe(false) + expect(enqueues).toEqual(['queued']) + await vi.waitFor(() => { + expect(adapter.requests.some(request => request.sessionId === parent.id)).toBe(true) + }) + }) }) /** Prove report delivery uses ordinary logged user messages (runtime-context snapshots excluded). */ diff --git a/packages/subagent/tool-subagent-report/tsconfig.json b/packages/subagent/tool-subagent-report/tsconfig.json index 9ccaad69f0..1a2a08671a 100644 --- a/packages/subagent/tool-subagent-report/tsconfig.json +++ b/packages/subagent/tool-subagent-report/tsconfig.json @@ -17,6 +17,9 @@ { "path": "../../llm/llm" }, + { + "path": "../../core/system-prompt" + }, { "path": "../../core/tools" }, diff --git a/packages/support/acp-snapshot/README.i18n.yaml b/packages/support/acp-snapshot/README.i18n.yaml index 5898c5beb1..ca9e9e67f0 100644 --- a/packages/support/acp-snapshot/README.i18n.yaml +++ b/packages/support/acp-snapshot/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/support/acp-snapshot/README.md -README.md: 8a9669b1f555fceb7af6d017d273b62d7dcad1e1 -README.zh.md: efa31df86d3b494ab14fc3742fc84adce570a593 +README.md: 06f1cb67cfcd954254db480ea696d10d81b37438 +README.zh.md: c4a5643f8c5d7e5b62a160cd32e5969187d34046 diff --git a/packages/support/acp-snapshot/README.md b/packages/support/acp-snapshot/README.md index 8a9669b1f5..06f1cb67cf 100644 --- a/packages/support/acp-snapshot/README.md +++ b/packages/support/acp-snapshot/README.md @@ -55,6 +55,8 @@ A scenario booting a differently-composed tree sets its own `configPath` (an ove A pin owns its generated `system-prompt.expected.md` or `tool-schemas.expected.json` by default; `systemPromptSource` and `toolSchemasSource` name another pin when the complete corresponding sequence is identical, so each distinct version is committed once. The pin's `session.jsonl` stores `"system":"{{system}}","tools":"{{tools}}"` while retaining config, reason, and any model-visible prefix. A pin with legitimate mid-run header changes declares `expectedHeaderChanges`; a shared source must declare the same count, and record/refresh rejects claimants that generate different bytes. +A child session whose own scope composes a different request declares it per fixture index: `pinsChildToolSchemas` moves that child's tool sequence into `tool-schemas..expected.json`, and `pinsChildSystemPrompts` moves its prompt into `system-prompt..expected.md`. Each names the `session..jsonl` fixture it describes, leaves every other request-header field to the class pin, and requires its sidecar to exist exactly when declared. A child prompt sidecar must also differ from its class pin, so a redundant copy fails instead of drifting. A continuable child carrying the scope-local `report` tool and its guidance section is the shipped case for both. + Every scenario compares `stdout.expected.jsonl` with cwd-rooted separators canonicalized to `/`. On Windows, `pinsNativeWindowsStdout` additionally compares the complete `stdout.expected.windows.jsonl` after the shared expected output and requires that sidecar exactly when enabled. A scenario requiring a non-Windows host declares `posixOnly`, which skips its run test on Windows while the fixture guards keep covering its committed files everywhere; examples include POSIX process semantics (e.g. cancelling a live bash call kills a detached process group) and generated paths Windows cannot represent. A scenario whose composition needs a usable `pwsh` declares `pwshOnly`; the caller-supplied `hasPwsh` probe (the shipped acp-agent suite follows the executor's own resolution, so Program Files installs count) skips the run test when no usable `pwsh` resolves while the fixture guards keep covering its committed files everywhere. The example also ships a `cordis.snapshot.yml` replay overlay next to its `cordis.yml` (the bin swaps them under `DSH_SNAPSHOT=replay` — [single-source replay config Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)); replay fixtures are served by [`dsh-llm-replay`](../llm-replay/README.md), which this package points at via the `DSH_SNAPSHOT_*` env vars it sets on the child. `pnpm run test:snapshot:record` calls the live LLM and rewrites the recorded scenarios' model fixtures; `pnpm run test:snapshot:refresh` stays keyless, runs the replay overlay, and rewrites stdout, comparable session-log expected outputs, and owned prompt and tool-schema sidecars from the committed model scripts. Fixture roles, record/replay/refresh semantics, and scenario-table fields are documented on `Scenario` and in the [snapshot Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md). diff --git a/packages/support/acp-snapshot/README.zh.md b/packages/support/acp-snapshot/README.zh.md index efa31df86d..c4a5643f8c 100644 --- a/packages/support/acp-snapshot/README.zh.md +++ b/packages/support/acp-snapshot/README.zh.md @@ -55,6 +55,8 @@ defineAcpSnapshotSuite({ 每个 pin 默认拥有其生成的 `system-prompt.expected.md` 或 `tool-schemas.expected.json`；当完整的对应序列相同时，`systemPromptSource` 和 `toolSchemasSource` 指定另一个 pin 作为来源，因此每个不同版本只提交一次。该 pin 的 `session.jsonl` 存储 `"system":"{{system}}","tools":"{{tools}}"`，同时保留配置、原因和任何模型可见前缀。具有合法运行中 header 变更的 pin 声明 `expectedHeaderChanges`；共享来源必须声明相同的 header 变更数量，录制/刷新会拒绝生成不同字节的共享引用方。 +自身作用域组合出不同请求的 child 会话按 fixture 索引单独声明：`pinsChildToolSchemas` 把该 child 的工具序列移入 `tool-schemas..expected.json`，`pinsChildSystemPrompts` 把其提示词移入 `system-prompt..expected.md`。两者都指名自己描述的 `session..jsonl` fixture，其余请求 header 字段仍归类别 pin 所有，并要求 sidecar 恰好在声明时存在。child 提示词 sidecar 还必须与其类别 pin 不同，因此冗余副本会直接失败，而不会悄悄漂移。携带作用域局部 `report` 工具及其指引 section 的可继续 child 是两者的随附用例。 + 每个场景都比较 `stdout.expected.jsonl`，其中以 cwd 为根的分隔符规范化为 `/`。在 Windows 上，`pinsNativeWindowsStdout` 还会在共享预期输出之后比较完整 `stdout.expected.windows.jsonl`，并在启用时精确要求该 sidecar。需要非 Windows 主机的场景声明 `posixOnly`，在 Windows 上跳过运行测试，但 fixture 保护仍在所有平台覆盖其已提交文件；示例包括 POSIX 进程语义（例如取消实时 bash 调用会终止脱离进程组）和 Windows 无法表示的生成路径。组合需要可用 `pwsh` 的场景声明 `pwshOnly`；调用方提供的 `hasPwsh` 探测（随附的 acp-agent 套件遵循执行器自身的解析，因此 Program Files 安装也计入）在解析不到可用 `pwsh` 时跳过运行测试，而 fixture 保护仍处处覆盖其已提交文件。 示例还发布 `cordis.snapshot.yml` 回放 overlay，位于 `cordis.yml` 旁边（bin 在 `DSH_SNAPSHOT=replay` 下交换它们，见[单源回放配置 Agent Note](../../../.agents/notes/archived/testing/2026-07-04-single-source-acp-replay-config.md)）；回放 fixture 由 [`dsh-llm-replay`](../llm-replay/README.md) 提供，该包通过对子级设置的 `DSH_SNAPSHOT_*` env var 指向它。`pnpm run test:snapshot:record` 调用实时 LLM，并重写已记录场景的模型 fixture；`pnpm run test:snapshot:refresh` 保持无密钥，运行回放 overlay，并从已提交模型脚本重写 stdout、可比较会话日志预期输出，以及各 pin 自有的提示词与工具 schema sidecar。Fixture 角色、录制/回放/刷新语义和场景表字段记录在 `Scenario` 以及[快照 Agent Note](../../../.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md) 中。 diff --git a/packages/support/acp-snapshot/src/suite.ts b/packages/support/acp-snapshot/src/suite.ts index 5cdffadf38..1728671c3a 100644 --- a/packages/support/acp-snapshot/src/suite.ts +++ b/packages/support/acp-snapshot/src/suite.ts @@ -46,6 +46,11 @@ function childToolSchemasSnapshot(index: number): string { return `tool-schemas.${index}.expected.json` } +/** Return the dedicated system-prompt sidecar for one child fixture index. */ +function childSystemPromptSnapshot(index: number): string { + return `system-prompt.${index}.expected.md` +} + /** The optional full Windows-native stdout transcript. */ const WINDOWS_STDOUT_SNAPSHOT = 'stdout.expected.windows.jsonl' @@ -116,6 +121,13 @@ export interface Scenario { * request-header field. */ pinsChildToolSchemas?: readonly number[] + /** + * Child fixture indices whose own system prompt is pinned separately, where + * `1` names `session.1.jsonl` and `system-prompt.1.expected.md`. A child + * scope that installs its own prompt section (the continuable `report` + * guidance) composes a prompt the class pin cannot describe. + */ + pinsChildSystemPrompts?: readonly number[] /** * How many changed `request/header` snapshots this PINNING scenario's primary * fixture legitimately carries (default 0). Their full prompt text is kept in @@ -491,6 +503,18 @@ export function formatSystemPromptSnapshot( return snapshot } +/** + * Reject a child prompt sidecar that cannot own distinct, canonical prompt text. + * @param sidecar - committed child prompt snapshot. + * @param classPin - initial prompt snapshot owned by the scenario's header class. + * @param label - repository-relative fixture label for diagnostics. + */ +export function assertChildSystemPromptSnapshot(sidecar: string, classPin: string, label: string): void { + if (sidecar.trim().length === 0) throw new Error(`${label} must pin a non-empty prompt`) + if (!sidecar.endsWith('\n')) throw new Error(`${label} must end in a newline`) + if (sidecar === classPin) throw new Error(`${label} must differ from its class pin`) +} + /** Return the initial-prompt portion of a possibly multi-header snapshot. */ function initialSystemPromptSnapshot(snapshot: string): string { const marker = snapshot.indexOf('\n diff --git a/docs/subsystems/session.zh.md b/docs/subsystems/session.zh.md index 39e8ff1e88..2d9eb45c4a 100644 --- a/docs/subsystems/session.zh.md +++ b/docs/subsystems/session.zh.md @@ -748,7 +748,7 @@ fork(source: SessionForkSource, boundary?: number, childSessionId?: SessionId): Types: [CreateSessionOptions](persistence.md) · [PrepareSessionOptions](persistence.md) · [SessionId](core.md) -Source: [`packages/core/session/src/index.ts:813`](../../packages/core/session/src/index.ts) +Source: [`packages/core/session/src/index.ts:792`](../../packages/core/session/src/index.ts) @@ -777,7 +777,7 @@ Creation announcement during session publication. A synchronous throw vetoes and Types: [Scoped](scope.md) -Source: [`packages/core/session/src/index.ts:75`](../../packages/core/session/src/index.ts) +Source: [`packages/core/session/src/index.ts:54`](../../packages/core/session/src/index.ts) @@ -800,7 +800,7 @@ Emitted once when an announced session leaves the store, including publication r Types: [Scoped](scope.md) -Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts) +Source: [`packages/core/session/src/index.ts:64`](../../packages/core/session/src/index.ts) @@ -825,7 +825,7 @@ Post-commit, fire-and-forget append feed. The listener snapshot resolves before Types: [Scoped](scope.md) -Source: [`packages/core/session/src/index.ts:97`](../../packages/core/session/src/index.ts) +Source: [`packages/core/session/src/index.ts:76`](../../packages/core/session/src/index.ts) @@ -847,5 +847,5 @@ Awaited parallel durability checkpoint: every listener runs and the caller await Types: [Scoped](scope.md) -Source: [`packages/core/session/src/index.ts:106`](../../packages/core/session/src/index.ts) +Source: [`packages/core/session/src/index.ts:85`](../../packages/core/session/src/index.ts) diff --git a/docs/subsystems/subagent.i18n.yaml b/docs/subsystems/subagent.i18n.yaml index 86ecf90b02..be4a4e74ff 100644 --- a/docs/subsystems/subagent.i18n.yaml +++ b/docs/subsystems/subagent.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write docs/subsystems/subagent.md -subagent.md: 42359215b2e21028e13734c2a4cd2488f86a07af -subagent.zh.md: 0d1d4c42fa774867679ff248bfc99b0024b04257 +subagent.md: fbfe83010fc297b6c10baf0631e8cdfd4d05854c +subagent.zh.md: fc2fc85e35eaee1467921dc9f94b3d2a40292bbb diff --git a/docs/subsystems/subagent.md b/docs/subsystems/subagent.md index 42359215b2..fbfe83010f 100644 --- a/docs/subsystems/subagent.md +++ b/docs/subsystems/subagent.md @@ -209,6 +209,27 @@ interface SubagentReportMessageSource { type SubagentReportDelivery = 'quiet' | 'wakeup' ``` +Reporting is the child's own choice, so the manager keeps a separate account of its own: when a resident Activation settles, it delivers one notice to the child's durable direct parent describing how that epoch ended and carrying its final assistant content. That delivery is unconditional for every child whose id a caller received, happens before the ownership release that would let the parent be judged settled, and reaches a resident parent through the same waking-admission accounting as a report. A parent whose own lineage is already tearing down receives it without a wake, because waking a quiescent Agent starts a turn rather than queueing work. Its provenance is a distinct kind so a transcript never presents a runtime account as something the child wrote. + +```ts type-equiv +/** + * Durable attribution for the runtime's own account of a continuable child + * settling. Deliberately a different kind from + * {@link SubagentReportMessageSource}: a report is content the child chose, + * while this message is the manager stating what became of the child, and a + * transcript that merged them would credit the child with words it never wrote. + */ +interface SubagentSettledMessageSource { + readonly kind: 'subagent-settled' + /** A runtime account shown without expanding the row (`notice` context form). */ + readonly form: 'notice' + /** One-line account of how the child ended. */ + readonly summary: string + /** Session id of the child that settled. */ + readonly senderSessionId: SessionId +} +``` + ```ts type-equiv /** Options for one continuable child's report to its direct parent. */ interface SubagentReportOptions { @@ -619,7 +640,7 @@ async start(name: string, request: SubagentStartRequest): Promise Types: [Agent](core.md) · [ContentBlock](llm-streaming.md) · [MessageId](llm-streaming.md) · [SessionId](core.md) -Source: [`packages/subagent/subagent/src/index.ts:170`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:171`](../../packages/subagent/subagent/src/index.ts) @@ -645,7 +666,7 @@ A published child settled. Scope-filtered dispatch uses the same delegating pare Types: [Scoped](scope.md) -Source: [`packages/subagent/subagent/src/index.ts:165`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:166`](../../packages/subagent/subagent/src/index.ts) @@ -662,7 +683,7 @@ A provider became resolvable in the registry. 'subagent/provider-added'(provider: SubagentProvider): void ``` -Source: [`packages/subagent/subagent/src/index.ts:139`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:140`](../../packages/subagent/subagent/src/index.ts) @@ -679,7 +700,7 @@ A provider left the registry. Accepted runs remain holder-owned. 'subagent/provider-removed'(name: string): void ``` -Source: [`packages/subagent/subagent/src/index.ts:145`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:146`](../../packages/subagent/subagent/src/index.ts) @@ -703,5 +724,5 @@ A provider established a published child. For in-process providers, `ctx.agents. Types: [Scoped](scope.md) -Source: [`packages/subagent/subagent/src/index.ts:156`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:157`](../../packages/subagent/subagent/src/index.ts) diff --git a/docs/subsystems/subagent.zh.md b/docs/subsystems/subagent.zh.md index 0d1d4c42fa..fc2fc85e35 100644 --- a/docs/subsystems/subagent.zh.md +++ b/docs/subsystems/subagent.zh.md @@ -209,6 +209,27 @@ interface SubagentReportMessageSource { type SubagentReportDelivery = 'quiet' | 'wakeup' ``` +上报是 child 自己的选择，因此管理器还保有一份属于自己的记账：当驻留 Activation 结算时，它会向该 child 持久化的直接 parent 投递一条通知，说明该 epoch 如何结束，并携带其最终 assistant 内容。对每个调用方拿到过 id 的 child，这条投递都是无条件的；它发生在会让 parent 被判定为已结算的所有权释放之前，并通过与上报相同的唤醒准入记账到达驻留 parent。若 parent 自身所在的谱系已在拆卸中，这条通知会以不唤醒的方式送达，因为唤醒一个静息 Agent 是开启一个轮次，而不是排队等待工作。其来源信息使用一个独立的 kind，因此 transcript（文本记录）绝不会把运行时的记账呈现为 child 自己写下的内容。 + +```ts type-equiv +/** + * Durable attribution for the runtime's own account of a continuable child + * settling. Deliberately a different kind from + * {@link SubagentReportMessageSource}: a report is content the child chose, + * while this message is the manager stating what became of the child, and a + * transcript that merged them would credit the child with words it never wrote. + */ +interface SubagentSettledMessageSource { + readonly kind: 'subagent-settled' + /** A runtime account shown without expanding the row (`notice` context form). */ + readonly form: 'notice' + /** One-line account of how the child ended. */ + readonly summary: string + /** Session id of the child that settled. */ + readonly senderSessionId: SessionId +} +``` + ```ts type-equiv /** Options for one continuable child's report to its direct parent. */ interface SubagentReportOptions { @@ -621,7 +642,7 @@ async start(name: string, request: SubagentStartRequest): Promise Types: [Agent](core.md) · [ContentBlock](llm-streaming.md) · [MessageId](llm-streaming.md) · [SessionId](core.md) -Source: [`packages/subagent/subagent/src/index.ts:170`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:171`](../../packages/subagent/subagent/src/index.ts) @@ -647,7 +668,7 @@ A published child settled. Scope-filtered dispatch uses the same delegating pare Types: [Scoped](scope.md) -Source: [`packages/subagent/subagent/src/index.ts:165`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:166`](../../packages/subagent/subagent/src/index.ts) @@ -664,7 +685,7 @@ A provider became resolvable in the registry. 'subagent/provider-added'(provider: SubagentProvider): void ``` -Source: [`packages/subagent/subagent/src/index.ts:139`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:140`](../../packages/subagent/subagent/src/index.ts) @@ -681,7 +702,7 @@ A provider left the registry. Accepted runs remain holder-owned. 'subagent/provider-removed'(name: string): void ``` -Source: [`packages/subagent/subagent/src/index.ts:145`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:146`](../../packages/subagent/subagent/src/index.ts) @@ -705,5 +726,5 @@ A provider established a published child. For in-process providers, `ctx.agents. Types: [Scoped](scope.md) -Source: [`packages/subagent/subagent/src/index.ts:156`](../../packages/subagent/subagent/src/index.ts) +Source: [`packages/subagent/subagent/src/index.ts:157`](../../packages/subagent/subagent/src/index.ts) diff --git a/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml b/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml new file mode 100644 index 0000000000..4d28b399f0 --- /dev/null +++ b/examples/acp-agent/subagent-report-quiet.cordis.snapshot.yml @@ -0,0 +1,51 @@ +# Keyless counterpart to subagent-report-quiet.cordis.yml: replace the live +# adapter with replay, keep report delivery quiet, and fence the child behind +# the end of its parent's spawn turn so settlement opens the next turn. +- id: base + name: '@deepseek-ai/cordis-plugin-include' + config: + path: ./cordis.yml + patches: + - id: llm-deepseek + name: '@deepseek-ai/dsh-llm-deepseek' + disabled: true + - id: acp-agent + name: '@deepseek-ai/dsh-acp-demo' + config: + provider: deepseek-official + model: deepseek-v4-flash + persistenceRoot: !!js process.env.DSH_SNAPSHOT_SESSIONS_ROOT ?? './.sessions' + persistenceCompression: none + workspaceContext: + maxBytes: 65536 + persona: | + You are a coding assistant powered by the {{model}} model. Your working directory is {{cwd}}. Your bash tool runs under a file sandbox — a `[sandbox: file access denied …]` result is policy, not a command bug. + + Verify your work by running the code or tests. Keep answers brief and factual. + - id: sandbox + name: '@deepseek-ai/dsh-sandbox-local' + config: + runnerCommand: + - bash + - -c + - while [ "$1" != "--" ]; do shift; done; shift; exec "$@" + - passthrough-runner + runnerFailureSignatures: + - 'passthrough-runner: profile rejected' + - id: tool-subagent-report + name: '@deepseek-ai/dsh-tool-subagent-report' + config: + reportDelivery: quiet + - insert: + - id: llm-replay + name: '@deepseek-ai/dsh-llm-replay' + config: + providers: + - id: deepseek-official + name: DeepSeek + models: + - id: deepseek-v4-flash + - id: deepseek-v4-pro + +- id: report-fence + name: './tests/fixtures/subagent-report-fence.ts' diff --git a/examples/acp-agent/subagent-report-quiet.cordis.yml b/examples/acp-agent/subagent-report-quiet.cordis.yml new file mode 100644 index 0000000000..cb1c2b6d34 --- /dev/null +++ b/examples/acp-agent/subagent-report-quiet.cordis.yml @@ -0,0 +1,17 @@ +# Snapshot-only overlay pinning quiet report delivery. The shipped default wakes +# the parent on every accepted report, and the runtime's settlement notice wakes +# it again when the child's Activation ends; two independent wakes have no single +# authored order. Quiet delivery leaves settlement as the only wake, while the +# fixture below holds the child until the parent's spawn turn has closed. +- id: base + name: '@deepseek-ai/cordis-plugin-include' + config: + path: ./cordis.yml + patches: + - id: tool-subagent-report + name: '@deepseek-ai/dsh-tool-subagent-report' + config: + reportDelivery: quiet + +- id: report-fence + name: './tests/fixtures/subagent-report-fence.ts' diff --git a/examples/acp-agent/tests/acp.snapshot.ts b/examples/acp-agent/tests/acp.snapshot.ts index 3b89ae8a35..b1df6b0eef 100644 --- a/examples/acp-agent/tests/acp.snapshot.ts +++ b/examples/acp-agent/tests/acp.snapshot.ts @@ -44,6 +44,9 @@ const CHILD_QUESTION_CONFIG = fileURLToPath(new URL('../child-question.cordis.ym const SESSION_SANDBOX_ROOT_CONFIG = fileURLToPath(new URL('../session-sandbox-root.cordis.yml', import.meta.url)) const RETRY_CONFIG = fileURLToPath(new URL('../retry.cordis.yml', import.meta.url)) const SESSION_TITLE_CONFIG = fileURLToPath(new URL('../session-title.cordis.yml', import.meta.url)) +const SUBAGENT_REPORT_QUIET_CONFIG = fileURLToPath( + new URL('../subagent-report-quiet.cordis.yml', import.meta.url), +) const SUBAGENT_DURABILITY_FAILURE_CONFIG = fileURLToPath( new URL('../subagent-durability-failure.cordis.yml', import.meta.url), ) @@ -353,7 +356,13 @@ const SCENARIOS: Scenario[] = [ // turns on that same child (the parent is never woken with their output), // send_message to an unknown subagent id fails without delivering, and the // child's retained handle is disposed child-first at teardown despite a - // failed final durability confirmation. + // failed final durability confirmation. That failed confirmation is also what + // the settlement notice must report: the child's last turn claimed the third + // message and then died on its durability checkpoint without entering a step, + // so the notice opening the parent's second turn says the child FAILED and the + // parent must not read the earlier answer as final. The scenario's fixture + // fences the child behind the parent's spawn turn so that notice can only + // arrive at an idle parent. { name: 'subagent-continuable', hasModelTurn: true, @@ -367,6 +376,8 @@ const SCENARIOS: Scenario[] = [ // continuable background child's log carries that override as a // `sandbox/mode` `source: 'delegation'` event, so the child's runtime // context states the inherited policy instead of the deployment default. + // The input also waits for the manager-owned settlement turn, keeping that + // delivery from racing transcript harvest. { name: 'subagent-continuable-inheritance', hasModelTurn: true, @@ -387,12 +398,16 @@ const SCENARIOS: Scenario[] = [ configPath: SUBAGENT_DURABILITY_FAILURE_CONFIG, }, // Authored child-to-parent transcript: the child calls its scope-local - // `report`, the default waking delivery gives the parked parent one ordinary - // turn on that report, and a later prompt still reads it back from the log. + // `report`, and the runtime's unconditional settlement notice then wakes the + // parked parent into one ordinary turn that claims both. The overlay pins + // quiet report delivery because two independent wakes have no orderable + // transcript; the shipped waking default is covered by package tests. { name: 'subagent-report', hasModelTurn: true, recorded: false, + overridden: false, + configPath: SUBAGENT_REPORT_QUIET_CONFIG, pinsChildToolSchemas: [1], pinsChildSystemPrompts: [1], }, diff --git a/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts b/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts index 8f5185026e..9ebc0d6ac1 100644 --- a/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts +++ b/examples/acp-agent/tests/fixtures/subagent-durability-failure.ts @@ -31,6 +31,8 @@ const FAILED_CHECKPOINT_TURN = 3 /** Fail the child checkpoint and stabilize the authored follow-up failure ordering. */ export function apply(ctx: Context): void { const followupsAccepted = Promise.withResolvers() + const parentTurnClosed = Promise.withResolvers() + let parentClosed = false const publishedFailure = process.env.DSH_SUBAGENT_PUBLISHED_FAILURE === '1' const persistence = ctx.sessionPersistence const load = persistence.load.bind(persistence) @@ -62,8 +64,22 @@ export function apply(ctx: Context): void { agents.create = create persistence.load = load followupsAccepted.resolve(undefined) + parentTurnClosed.resolve(undefined) }, 'subagent snapshot ordering') + // The manager's settlement notice races whatever the parent is doing when the + // child's Activation ends, and this transcript pins it as the parent's own + // later turn. Hold the child's steps until the parent's spawn turn closes, so + // the notice can only arrive at an idle parent. The parent's turn never awaits + // child model work — its own fences need inbox acceptance only — so the child + // cannot deadlock it. + ctx.on('session/event', (session, event) => { + if (session.header.parentSession !== undefined || event.type !== 'turn/end') return + if (event.data.turn !== 1) return + parentClosed = true + parentTurnClosed.resolve(undefined) + }) + // Remap the placeholder child id in a follow-up to the live child. The child // id the model "knows" is authored into the transcript, while the running // child is minted with a random id, so without this the follow-ups would @@ -91,7 +107,12 @@ export function apply(ctx: Context): void { if (accepted >= 3) followupsAccepted.resolve(undefined) }) ctx.on('agent/pre-step', async ({ agent }, next) => { - if (agent.session.header.parentSession !== undefined) await followupsAccepted.promise + if (agent.session.header.parentSession === undefined) return next() + await followupsAccepted.promise + // The published-failure variant's child never reaches a step (its follow-up + // throws), and its parent turn awaits that child, so only the continuable + // scenario takes the settlement fence. + if (!publishedFailure && !parentClosed) await parentTurnClosed.promise return next() }) diff --git a/examples/acp-agent/tests/fixtures/subagent-report-fence.ts b/examples/acp-agent/tests/fixtures/subagent-report-fence.ts new file mode 100644 index 0000000000..754343e4ec --- /dev/null +++ b/examples/acp-agent/tests/fixtures/subagent-report-fence.ts @@ -0,0 +1,41 @@ +/** + * Loader fixture that holds the report child until its parent's spawn turn ends. + * @module subagent-report-fence + */ + +import type { Context } from '@deepseek-ai/cordis' +import type {} from '@deepseek-ai/dsh-agent-loop' + +/** Fixture plugin name. */ +export const name = 'subagent-report-fence' + +/** + * Keep replay scheduling from folding settlement into the parent's first turn. + * @param ctx - assembled ACP-agent context. + */ +export function apply(ctx: Context): void { + const childReady = Promise.withResolvers() + const parentStopped = Promise.withResolvers() + let hasStopped = false + + ctx.effect(() => { + const disposeSession = ctx.root.on('session/event', (session, event) => { + if (session.header.parentSession !== undefined || event.type !== 'turn/end' || event.data.turn !== 1) return + hasStopped = true + parentStopped.resolve(undefined) + }) + const disposeStep = ctx.root.on('agent/pre-step', async ({ agent, turn, step }, next) => { + if (agent.session.header.parentSession !== undefined) { + childReady.resolve(undefined) + if (!hasStopped) await parentStopped.promise + } else if (turn === 1 && step === 2) { + await childReady.promise + } + return next() + }) + return () => { + disposeStep() + disposeSession() + } + }, 'subagent-report-fence.listeners') +} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl index 2a084e4a9c..41a574fef4 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357538308,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}} {"type":"step/start","seq":5,"time":1786357538310,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730458555,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"ebe0cfa0-a909-47e0-8294-28ad84a8fe77"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"1b650647-a278-4be1-a76e-bf09540bff75"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357538310,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"a74dc391-2558-432c-b665-603b52eba070"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357538310,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730458555,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730458555,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl index 252bfb326c..97bd2468c6 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357538469,"data":{"version":2,"mode":"one-shot","provider":"spawn"}} {"type":"step/start","seq":5,"time":1786357538470,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730458703,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"2ac2cc54-9bce-4cfa-a569-a64f51bc30a7"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c74c4556-1622-4d2e-9745-c5abd087fa83"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357538471,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c2882641-5206-4d05-9b81-03898bc2943e"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357538471,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730458703,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730458703,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl index aea9e3107c..56ae9d71c6 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/session.jsonl @@ -4,7 +4,7 @@ {"type":"agent/inbox/spliced","seq":2,"time":1785821417919,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1783957884486,"data":{"turn":1,"step":1}} {"type":"user/message","seq":4,"time":1785498801761,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_ACP_OK."}],"source":{"kind":"user"},"role":"user","id":"6e45782a-31be-4ba7-8c4a-7411a2027e36"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730458430,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"06416873-c855-452d-8996-ea5cf45223d1"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730458430,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"b6de532e-08ff-42b0-abae-1895dc463500"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730458430,"data":{"title":"Run this advanced flow exactly","messageSeqs":[4],"source":{"kind":"fallback"}}} {"type":"request/header","seq":7,"time":1785498801765,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730458431,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md index dcd7e67bca..e77d9bb220 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/system-prompt.expected.md @@ -132,13 +132,13 @@ interface ToolArgsMap { /** The exact skill name from the available skills list. */ name: string; } & Record; - /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */ + /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */ subagent: { /** A short (3-5 word) description of the delegated task, for display. */ description: string; /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */ prompt: string; - /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ + /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */ run_in_background?: boolean; } & Record; /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */ diff --git a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json index 5b1ee082a7..e5d06a90bf 100644 --- a/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/advanced-toolchain/tool-schemas.expected.json @@ -309,7 +309,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -323,7 +323,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json index 1f323c3ea9..a655e8a4cc 100644 --- a/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/both-mode-turn/tool-schemas.expected.json @@ -252,7 +252,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -266,7 +266,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md index f02d9ab50a..6737c824ea 100644 --- a/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md +++ b/examples/acp-agent/tests/snapshots/code-mode-turn/system-prompt.expected.md @@ -115,13 +115,13 @@ interface ToolArgsMap { /** The exact skill name from the available skills list. */ name: string; } & Record; - /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */ + /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */ subagent: { /** A short (3-5 word) description of the delegated task, for display. */ description: string; /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */ prompt: string; - /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */ + /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */ run_in_background?: boolean; } & Record; /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */ diff --git a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json index f00decd284..5cae77c81b 100644 --- a/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/lsp-definition/tool-schemas.expected.json @@ -268,7 +268,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -282,7 +282,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json index 0c85c0265d..27fdd9aa04 100644 --- a/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/product-subagent-both/tool-schemas.expected.json @@ -231,7 +231,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -245,7 +245,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json index 74c8cdf4b0..e7fadcd731 100644 --- a/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/product-subagent-codex/tool-schemas.expected.json @@ -231,7 +231,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -245,7 +245,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json index 18a14221d9..d7be239de3 100644 --- a/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/pty-tools/tool-schemas.expected.json @@ -231,7 +231,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -245,7 +245,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json index e018ff89a4..96fd6b34de 100644 --- a/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/session-query-spill/tool-schemas.expected.json @@ -435,7 +435,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -449,7 +449,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json index d0cc2da730..698b522fb3 100644 --- a/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-child-question-rejection/tool-schemas.expected.json @@ -294,7 +294,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -308,7 +308,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json index 183e21b557..f07f1d51fc 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/input.json @@ -14,6 +14,13 @@ "op": "waitForSubagentTurnEnd", "child": 1, "minimumTurn": 1 + }, + { + "op": "waitForTurnStart", + "minimumTurn": 2 + }, + { + "op": "waitForTurnEnd" } ] } diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl index 15b1748ea5..293f9c3b83 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/session.jsonl @@ -27,3 +27,16 @@ {"type":"assistant/message","seq":25,"time":1786333735904,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4057a08e-b50e-45e7-beb0-c74485f2b7d6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[20,21,22,23,24],"surfaceOp":"append"} {"type":"step/end","seq":26,"time":1786333735904,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":27,"time":1786333735904,"data":{"turn":1,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":28,"time":1786374357751,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"e0bd4902-daba-4e23-bfcb-9e102fdd203d"}]}} +{"type":"turn/start","seq":29,"time":1786374357751,"data":{"turn":2}} +{"type":"agent/inbox/spliced","seq":30,"time":1786374357751,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":31,"time":1786374357758,"data":{"turn":2,"step":1}} +{"type":"user/message","seq":32,"time":1786374357758,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"e0bd4902-daba-4e23-bfcb-9e102fdd203d"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":33,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":34,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}} +{"type":"assistant/chunk","seq":35,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} +{"type":"assistant/chunk","seq":36,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":37,"time":1786374357764,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":38,"time":1786374357764,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"bbe5ef7b-2a3a-47f4-8475-60945b31a373"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[33,34,35,36,37],"surfaceOp":"append"} +{"type":"step/end","seq":39,"time":1786374357765,"data":{"turn":2,"step":1}} +{"type":"turn/end","seq":40,"time":1786374357765,"data":{"turn":2,"reason":{"kind":"completed"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl index 82ae8907ca..d6a2728b5a 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/stdout.expected.jsonl @@ -2,3 +2,4 @@ {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}} {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json index 6cecb69da2..d0b02de9dd 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable-inheritance/tool-schemas.1.expected.json @@ -247,7 +247,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -261,7 +261,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/input.json b/examples/acp-agent/tests/snapshots/subagent-continuable/input.json index f1354bd76a..380a7825c2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable/input.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/input.json @@ -10,6 +10,17 @@ "op": "prompt", "text": "Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool." }, - { "op": "waitForSubagentTurnEnd", "child": 1, "minimumTurn": 3 } + { + "op": "waitForSubagentTurnEnd", + "child": 1, + "minimumTurn": 3 + }, + { + "op": "waitForTurnStart", + "minimumTurn": 2 + }, + { + "op": "waitForTurnEnd" + } ] } diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl index a759c35a32..60a55d1373 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/session.jsonl @@ -1,10 +1,10 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785730451297,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"125665d3-8c03-4190-b4f9-c27d61d245f4"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785730451297,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"579d3d6d-a57e-4d55-9b48-05832a79d9f8"}]}} {"type":"turn/start","seq":1,"time":1785821408972,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821408972,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1785730451327,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785730451327,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"125665d3-8c03-4190-b4f9-c27d61d245f4"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730451328,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"533e9513-a329-4a36-9a8d-ddaf544b57c3"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785730451327,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. 2. Call send_message twice in a row, both with the subagent id from step 1: first with message 'Now reply with exactly SECOND_OK.', then with message 'Now reply with exactly THIRD_OK.'. 3. Call send_message with subagent_id exactly '22222222-2222-4222-8222-222222222222' (a subagent that does not exist) and message 'Please continue.', and observe that it fails. 4. Reply with the single word DONE. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"579d3d6d-a57e-4d55-9b48-05832a79d9f8"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730451328,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"c4f5f7ed-1c11-4f31-923f-3142c79f0c2c"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730451328,"data":{"title":"Follow these steps exactly, then","messageSeqs":[4],"source":{"kind":"fallback"}}} {"type":"request/header","seq":7,"time":1785730451329,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730451329,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} @@ -13,9 +13,9 @@ {"type":"assistant/chunk","seq":11,"time":1785544945188,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}}} {"type":"assistant/chunk","seq":12,"time":1785544945188,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730451338,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730451338,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4a2c2f8b-66be-4860-9bbf-b83feb56009e"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730451338,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"680da987-6d29-4141-b83d-af57b050c712"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730451338,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}} -{"type":"tool/result","seq":16,"time":1785730451348,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"b54a0d61-4233-40f1-ab3a-3eb58e1b0c61"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730451348,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"7825edb2-080e-49c1-ba74-ad69d16bf566"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730451348,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730451360,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":1789000000019,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -23,9 +23,9 @@ {"type":"assistant/chunk","seq":21,"time":1785544945212,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}}}} {"type":"assistant/chunk","seq":22,"time":1785544945212,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730451364,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":24,"time":1785730451364,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"398dea92-a100-4b2f-a9e1-72629def132d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730451364,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ef6eadc7-165e-4705-b865-3889f0af0f36"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"tool/call","seq":25,"time":1785730451365,"data":{"turn":1,"step":2,"callId":"call_followup_1","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly SECOND_OK.\"}"}} -{"type":"tool/result","seq":26,"time":1785730451377,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"call_followup_1"},"content":[{"type":"tool-result","toolCallId":"call_followup_1","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"45b388fd-6a48-4a02-9f3b-1d642e797c71"}},"sourceEventSeqs":[25],"surfaceOp":"append"} +{"type":"tool/result","seq":26,"time":1785730451377,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"call_followup_1"},"content":[{"type":"tool-result","toolCallId":"call_followup_1","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"ac1214a5-1d91-4fab-8f96-833baca114f8"}},"sourceEventSeqs":[25],"surfaceOp":"append"} {"type":"step/end","seq":27,"time":1785730451377,"data":{"turn":1,"step":2}} {"type":"step/start","seq":28,"time":1785730451390,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":29,"time":1789000000029,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -33,9 +33,9 @@ {"type":"assistant/chunk","seq":31,"time":1785544945241,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}}}} {"type":"assistant/chunk","seq":32,"time":1785544945242,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":33,"time":1785730451394,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":34,"time":1785730451394,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f56cec19-e761-4c77-9237-07d12d334275"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} +{"type":"assistant/message","seq":34,"time":1785730451394,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"33939813-0792-4ac5-8864-ec62a4ddff8e"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} {"type":"tool/call","seq":35,"time":1785730451395,"data":{"turn":1,"step":3,"callId":"call_followup_2","name":"send_message","arguments":"{\"subagent_id\": \"33333333-3333-4333-8333-333333333333\", \"message\": \"Now reply with exactly THIRD_OK.\"}"}} -{"type":"tool/result","seq":36,"time":1785730451406,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"call_followup_2"},"content":[{"type":"tool-result","toolCallId":"call_followup_2","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"193df31b-de9a-4723-b2b1-c7c96a8eaee4"}},"sourceEventSeqs":[35],"surfaceOp":"append"} +{"type":"tool/result","seq":36,"time":1785730451406,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"call_followup_2"},"content":[{"type":"tool-result","toolCallId":"call_followup_2","content":[{"type":"text","text":"message queued as the next turn for subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"a6f64c64-f50c-47cd-a6b3-a3b57d3dc83d"}},"sourceEventSeqs":[35],"surfaceOp":"append"} {"type":"step/end","seq":37,"time":1785730451406,"data":{"turn":1,"step":3}} {"type":"step/start","seq":38,"time":1785730451419,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":39,"time":1789000000039,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -43,9 +43,9 @@ {"type":"assistant/chunk","seq":41,"time":1785544945273,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}}}} {"type":"assistant/chunk","seq":42,"time":1785544945273,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":43,"time":1785730451424,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":44,"time":1785730451424,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"cbf54e23-fb96-45cc-b629-b9cb48fc9876"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} +{"type":"assistant/message","seq":44,"time":1785730451424,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"95eab91d-b103-4033-8e2e-c9c93b1b0211"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} {"type":"tool/call","seq":45,"time":1785730451425,"data":{"turn":1,"step":4,"callId":"call_followup_unknown","name":"send_message","arguments":"{\"subagent_id\": \"22222222-2222-4222-8222-222222222222\", \"message\": \"Please continue.\"}"}} -{"type":"tool/result","seq":46,"time":1785730451437,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"call_followup_unknown"},"content":[{"type":"tool-result","toolCallId":"call_followup_unknown","content":[{"type":"text","text":"Error: subagent \"22222222-2222-4222-8222-222222222222\" is unavailable"}],"isError":true}],"role":"user","id":"d5d961af-c171-40b6-87a1-33d2c14740a7"},"error":{"name":"SubagentError","code":"NOT_RESUMABLE"}},"sourceEventSeqs":[45],"surfaceOp":"append"} +{"type":"tool/result","seq":46,"time":1785730451437,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"call_followup_unknown"},"content":[{"type":"tool-result","toolCallId":"call_followup_unknown","content":[{"type":"text","text":"Error: subagent \"22222222-2222-4222-8222-222222222222\" is unavailable"}],"isError":true}],"role":"user","id":"8a095e4b-3059-420d-856f-1cbd20b6a2e2"},"error":{"name":"SubagentError","code":"NOT_RESUMABLE"}},"sourceEventSeqs":[45],"surfaceOp":"append"} {"type":"step/end","seq":47,"time":1785730451437,"data":{"turn":1,"step":4}} {"type":"step/start","seq":48,"time":1785730451450,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":49,"time":1785394678779,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -53,6 +53,19 @@ {"type":"assistant/chunk","seq":51,"time":1785544945303,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}} {"type":"assistant/chunk","seq":52,"time":1785544945303,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":53,"time":1785730451453,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":54,"time":1785730451453,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"fbea6b64-84cf-4411-abec-48d26b3801da"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} +{"type":"assistant/message","seq":54,"time":1785730451453,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"eb51ecb3-3347-4216-ad4e-c2130c43ecfc"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} {"type":"step/end","seq":55,"time":1785730451454,"data":{"turn":1,"step":5}} {"type":"turn/end","seq":56,"time":1785730451454,"data":{"turn":1,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":57,"time":1786011346660,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"SECOND_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2cf0afd2-ee6e-4a3f-a35a-2fd4d5b665ca"}]}} +{"type":"turn/start","seq":58,"time":1786011346660,"data":{"turn":2}} +{"type":"agent/inbox/spliced","seq":59,"time":1786011346660,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":60,"time":1786012466689,"data":{"turn":2,"step":1}} +{"type":"user/message","seq":61,"time":1786012466689,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"SECOND_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 failed before it finished.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2cf0afd2-ee6e-4a3f-a35a-2fd4d5b665ca"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":62,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":63,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}} +{"type":"assistant/chunk","seq":64,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} +{"type":"assistant/chunk","seq":65,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":66,"time":1786012466693,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":67,"time":1786333283620,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"758adcee-9284-4889-86a2-0181a278a754"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[62,63,64,65,66],"surfaceOp":"append"} +{"type":"step/end","seq":68,"time":1786333283620,"data":{"turn":2,"step":1}} +{"type":"turn/end","seq":69,"time":1786333283620,"data":{"turn":2,"reason":{"kind":"completed"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl index 82ae8907ca..d6a2728b5a 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/stdout.expected.jsonl @@ -2,3 +2,4 @@ {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}} {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json index 6cecb69da2..d0b02de9dd 100644 --- a/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-continuable/tool-schemas.1.expected.json @@ -247,7 +247,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -261,7 +261,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl index cfb7fee5a8..f36586cfb0 100644 --- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357533600,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth one"}} {"type":"step/start","seq":5,"time":1786357533602,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730456014,"data":{"content":[{"type":"text","text":"Call subagent once. Ask that child to attempt one further subagent call, then report the result."}],"source":{"kind":"user"},"role":"user","id":"a8129357-1bde-4cbd-90b4-6b8ad51d52e1"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"722c8476-461f-4a02-904a-f1084b973c95"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357533602,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"78a8aa2f-27d5-45d1-8b06-7b2627fc76ae"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357533602,"data":{"title":"Call subagent once. Ask that","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730456014,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730456014,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl index 022584de40..e267d3ada1 100644 --- a/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-depth-two-rejection/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357533628,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Start depth two"}} {"type":"step/start","seq":5,"time":1786357533630,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730456041,"data":{"content":[{"type":"text","text":"Attempt one subagent call beyond the configured cap, then report the rejection."}],"source":{"kind":"user"},"role":"user","id":"d4dc5a16-e542-4dd9-8e82-e6b7829cfc4b"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"fc837f97-bbf9-4c44-9eea-3a8f52c7a77d"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357533630,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"615d0b40-18da-47be-a235-7422989344f5"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357533630,"data":{"title":"Attempt one subagent call beyond","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730456041,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730456042,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json b/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json index b3b1d8e170..ab0c00f2f2 100644 --- a/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/input.json @@ -14,6 +14,13 @@ "op": "waitForFile", "path": ".dsh-snapshot-subagent-settled" }, + { + "op": "waitForTurnStart", + "minimumTurn": 2 + }, + { + "op": "waitForTurnEnd" + }, { "op": "prompt", "text": "Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool." diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl index 06bd463ba8..2cd0e7fc24 100644 --- a/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/session.jsonl @@ -1,10 +1,10 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785730454756,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"c2febfff-792d-4457-a944-933ff0de0570"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785730454756,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"356b3b62-c8b8-4d2a-84d7-7df1b6e4811e"}]}} {"type":"turn/start","seq":1,"time":1785821412725,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821412725,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1785730454783,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785730454783,"data":{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"c2febfff-792d-4457-a944-933ff0de0570"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730454783,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"c9d1f853-56bc-4082-ae08-00d4bcbb04a6"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785730454783,"data":{"content":[{"type":"text","text":"Call the subagent tool once with run_in_background set to true, description 'Reply with CHILD_OK', and prompt 'Reply with exactly the word CHILD_OK and nothing else.'. Then reply with the single word STARTED. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"356b3b62-c8b8-4d2a-84d7-7df1b6e4811e"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730454783,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"9be42fb0-f0d0-4ab9-a232-fb753f7db482"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730454783,"data":{"title":"Call the subagent tool once","messageSeqs":[4],"source":{"kind":"fallback"}}} {"type":"request/header","seq":7,"time":1785730454784,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730454784,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} @@ -13,9 +13,9 @@ {"type":"assistant/chunk","seq":11,"time":1785531795632,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}}}} {"type":"assistant/chunk","seq":12,"time":1785531795632,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730454793,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730454793,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4dda49d6-5d02-456f-ba95-68699662793d"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730454793,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c8802574-4e43-4ee7-8648-5a132935b5dc"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730454793,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Reply with CHILD_OK\", \"prompt\": \"Reply with exactly the word CHILD_OK and nothing else.\", \"run_in_background\": true}"}} -{"type":"tool/result","seq":16,"time":1785730454804,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"7f1fbe79-98b7-410a-af4d-c40d52a366dc"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730454804,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"c83395ad-93c6-4899-9ae1-8d29f92d4dde"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730454804,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730454814,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":1789000000017,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -23,39 +23,42 @@ {"type":"assistant/chunk","seq":21,"time":1785531795656,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}} {"type":"assistant/chunk","seq":22,"time":1785531795656,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730454820,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":24,"time":1785730454820,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"2ee459fa-21f9-48c6-a42e-3c38eca1e4c9"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730454820,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1fefe87b-4c3c-49b0-860c-8097193f9567"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"step/end","seq":25,"time":1785730454821,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":26,"time":1785730454821,"data":{"turn":1,"reason":{"kind":"completed"}}} -{"type":"agent/inbox/spliced","seq":27,"time":1785730454857,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"3bcac2a3-e0db-465b-94b4-4e2761236475"}]}} -{"type":"turn/start","seq":28,"time":1785821412840,"data":{"turn":2}} -{"type":"agent/inbox/spliced","seq":29,"time":1785821412840,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} -{"type":"step/start","seq":30,"time":1785730454863,"data":{"turn":2,"step":1}} -{"type":"user/message","seq":31,"time":1785730454863,"data":{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"3bcac2a3-e0db-465b-94b4-4e2761236475"},"surfaceOp":"append"} -{"type":"assistant/chunk","seq":32,"time":1785531795686,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":33,"time":1785536135065,"data":{"turn":2,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_list","name":"list_agents","argumentsDelta":"{\"scope\":\"descendants\"}"}}} -{"type":"assistant/chunk","seq":34,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}}}} -{"type":"assistant/chunk","seq":35,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} -{"type":"assistant/chunk","seq":36,"time":1785730454867,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":37,"time":1785730454867,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"199c7794-d782-4957-b7ed-69d094b9c0ef"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} -{"type":"tool/call","seq":38,"time":1785730454867,"data":{"turn":2,"step":1,"callId":"call_list","name":"list_agents","arguments":"{\"scope\":\"descendants\"}"}} -{"type":"tool/result","seq":39,"time":1785730454893,"data":{"turn":2,"step":1,"message":{"source":{"kind":"tool","callId":"call_list"},"content":[{"type":"tool-result","toolCallId":"call_list","content":[{"type":"text","text":"33333333-3333-4333-8333-333333333333 [complete] parent=11111111-1111-4111-8111-111111111111 depth=1 — Reply with CHILD_OK"}],"isError":false}],"role":"user","id":"61ceb650-3e32-413e-9d7e-b6ec8a351858"}},"sourceEventSeqs":[38],"surfaceOp":"append"} -{"type":"step/end","seq":40,"time":1785730454893,"data":{"turn":2,"step":1}} -{"type":"step/start","seq":41,"time":1785730454903,"data":{"turn":2,"step":2}} -{"type":"assistant/chunk","seq":42,"time":1785531795715,"data":{"turn":2,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":43,"time":1785536135106,"data":{"turn":2,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"call_interrupt","name":"interrupt_agent","argumentsDelta":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}}} -{"type":"assistant/chunk","seq":44,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}}}} -{"type":"assistant/chunk","seq":45,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} -{"type":"assistant/chunk","seq":46,"time":1785730454907,"data":{"turn":2,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":47,"time":1785730454907,"data":{"turn":2,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"17c1a7c5-84f5-493e-adb5-6a65219e6ad6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[42,43,44,45,46],"surfaceOp":"append"} -{"type":"tool/call","seq":48,"time":1785730454907,"data":{"turn":2,"step":2,"callId":"call_interrupt","name":"interrupt_agent","arguments":"{\"agent_id\":\"33333333-3333-4333-8333-333333333333\"}"}} -{"type":"tool/result","seq":49,"time":1785730454917,"data":{"turn":2,"step":2,"message":{"source":{"kind":"tool","callId":"call_interrupt"},"content":[{"type":"tool-result","toolCallId":"call_interrupt","content":[{"type":"text","text":"interrupt requested for agent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"bf93cbe0-1946-4aef-a4ce-431790026c6f"}},"sourceEventSeqs":[48],"surfaceOp":"append"} -{"type":"step/end","seq":50,"time":1785730454917,"data":{"turn":2,"step":2}} -{"type":"step/start","seq":51,"time":1785730454927,"data":{"turn":2,"step":3}} -{"type":"assistant/chunk","seq":52,"time":1785531795715,"data":{"turn":2,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} -{"type":"assistant/chunk","seq":53,"time":1785536135106,"data":{"turn":2,"step":3,"chunk":{"type":"text-delta","index":0,"text":"DONE"}}} -{"type":"assistant/chunk","seq":54,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}} -{"type":"assistant/chunk","seq":55,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":2}}}} -{"type":"assistant/chunk","seq":56,"time":1785730454907,"data":{"turn":2,"step":3,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":57,"time":1785730454907,"data":{"turn":2,"step":3,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e4c04cd6-9c23-4c99-a5f7-bac6fc141e95"},"usage":{"inputTokens":10,"outputTokens":2}},"sourceEventSeqs":[52,53,54,55,56],"surfaceOp":"append"} -{"type":"step/end","seq":58,"time":1785730454938,"data":{"turn":2,"step":3}} -{"type":"turn/end","seq":59,"time":1785730454938,"data":{"turn":2,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":27,"time":1786011499211,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"9275a12c-bf9a-48e2-b33b-4fc484e936cb"}]}} +{"type":"turn/start","seq":28,"time":1786011499211,"data":{"turn":2}} +{"type":"agent/inbox/spliced","seq":29,"time":1786011499211,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":30,"time":1786012470343,"data":{"turn":2,"step":1}} +{"type":"user/message","seq":31,"time":1786012470343,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_OK"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"9275a12c-bf9a-48e2-b33b-4fc484e936cb"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":32,"time":1786012470346,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":33,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}} +{"type":"assistant/chunk","seq":34,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} +{"type":"assistant/chunk","seq":35,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":36,"time":1786011499224,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":37,"time":1786012470346,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4620e8c0-dd13-4a2f-87dc-f4b66aa51219"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} +{"type":"step/end","seq":38,"time":1786012470346,"data":{"turn":2,"step":1}} +{"type":"turn/end","seq":39,"time":1786012470346,"data":{"turn":2,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":40,"time":1786012470360,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"7a2a86d0-80a3-4db5-822f-2d3fcbc16e11"}]}} +{"type":"turn/start","seq":41,"time":1786011499224,"data":{"turn":3}} +{"type":"agent/inbox/spliced","seq":42,"time":1786011499224,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":43,"time":1786011499234,"data":{"turn":3,"step":1}} +{"type":"user/message","seq":44,"time":1786011499234,"data":{"content":[{"type":"text","text":"Call list_agents once with scope set to descendants and observe the subagent you started. Then call interrupt_agent once with agent_id set to 33333333-3333-4333-8333-333333333333. Then reply with the single word DONE. Do not call any other tool."}],"source":{"kind":"user"},"role":"user","id":"7a2a86d0-80a3-4db5-822f-2d3fcbc16e11"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":45,"time":1786011499238,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":46,"time":1786011499238,"data":{"turn":3,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_list","name":"list_agents","argumentsDelta":"{}"}}} +{"type":"assistant/chunk","seq":47,"time":1785531795715,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{}"}}}} +{"type":"assistant/chunk","seq":48,"time":1785536135106,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":49,"time":1785730454907,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":50,"time":1786011499238,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_list","name":"list_agents","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"d3402e92-2f7e-4cd5-9537-ae9beedeecab"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[45,46,47,48,49],"surfaceOp":"append"} +{"type":"tool/call","seq":51,"time":1786011499238,"data":{"turn":3,"step":1,"callId":"call_list","name":"list_agents","arguments":"{}"}} +{"type":"tool/result","seq":52,"time":1786011499267,"data":{"turn":3,"step":1,"message":{"source":{"kind":"tool","callId":"call_list"},"content":[{"type":"tool-result","toolCallId":"call_list","content":[{"type":"text","text":"33333333-3333-4333-8333-333333333333 [complete] — Reply with CHILD_OK"}],"isError":false}],"role":"user","id":"8ae233de-8fde-48d7-a9d0-0d9a480a00d0"}},"sourceEventSeqs":[51],"surfaceOp":"append"} +{"type":"step/end","seq":53,"time":1785730454908,"data":{"turn":3,"step":1}} +{"type":"step/start","seq":54,"time":1786011499275,"data":{"turn":3,"step":2}} +{"type":"assistant/chunk","seq":55,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":56,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"text-delta","index":0,"text":"DONE"}}} +{"type":"assistant/chunk","seq":57,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}} +{"type":"assistant/chunk","seq":58,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":2}}}} +{"type":"assistant/chunk","seq":59,"time":1786011499279,"data":{"turn":3,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":60,"time":1786011499279,"data":{"turn":3,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3ac0f29f-72ae-44fb-9414-974470095618"},"usage":{"inputTokens":10,"outputTokens":2}},"sourceEventSeqs":[55,56,57,58,59],"surfaceOp":"append"} +{"type":"step/end","seq":61,"time":1786011499279,"data":{"turn":3,"step":2}} +{"type":"turn/end","seq":62,"time":1786011499279,"data":{"turn":3,"reason":{"kind":"completed"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl index 5a9b409384..4813c19f90 100644 --- a/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/stdout.expected.jsonl @@ -2,5 +2,6 @@ {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"STARTED"}}}} {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"DONE"}}}} {"jsonrpc":"2.0","id":4,"result":{"stopReason":"end_turn"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json index 6cecb69da2..d0b02de9dd 100644 --- a/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-list-agents/tool-schemas.1.expected.json @@ -247,7 +247,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -261,7 +261,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl index 442c7f06c8..c41aad3d0c 100644 --- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357524752,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Reply ALPHA only"}} {"type":"step/start","seq":5,"time":1786357524755,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730450187,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"73ce401a-faaf-408a-879e-7485380d537d"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"36cca213-d6e9-491d-9442-87057994f573"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357524755,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"8b8f6910-a5fc-4839-837f-9b9335252807"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357524755,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730450187,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730450188,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl index 6e071fa68d..9572fee7d9 100644 --- a/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-mixed/session.2.jsonl @@ -27,7 +27,7 @@ {"type":"subagent/descriptor","seq":41,"time":1786357524800,"data":{"version":2,"mode":"one-shot","provider":"fork","label":"Recall project codeword"}} {"type":"step/start","seq":42,"time":1786357524803,"data":{"turn":2,"step":1}} {"type":"user/message","seq":43,"time":1786357524803,"data":{"content":[{"type":"text","text":"What is the project codeword mentioned earlier in this conversation? Reply with exactly that one word and nothing else."}],"source":{"kind":"user"},"role":"user","id":"86e9f144-764f-460d-b72b-262cffe43d77"},"surfaceOp":"append"} -{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"86f2fe9d-3154-41c2-815e-e1c1e18ed895"},"surfaceOp":"append"} +{"type":"user/message","seq":44,"time":1786358036899,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"ccb25ca6-93e6-40e0-8713-ad065065d540"},"surfaceOp":"append"} {"type":"request/header","seq":45,"time":1786358036900,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"resume"}} {"type":"assistant/chunk","seq":46,"time":1783352148077,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"reasoning"}}} {"type":"reasoning-chunks","seq0":47,"time0":1783352148077,"data":{"turn":2,"step":1,"index":0,"dt":[0,0,29,0,0,0,35,0,0,0,0,26,29,31,0,30,0,0,27,1,27,0,1,0,0,31,1,0,0,1790157964],"texts":["The"," user"," is"," asking"," me"," to"," recall"," the"," project"," cod","ew","ord"," that"," was"," mentioned"," earlier"," in"," the"," conversation","."," I"," was"," told"," to"," remember"," it",":"," SA","FF","RON","."]}} diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl index e5848d377c..d6d7283abb 100644 --- a/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.1.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357521754,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return ALPHA only"}} {"type":"step/start","seq":5,"time":1786357521756,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730447828,"data":{"content":[{"type":"text","text":"Reply with exactly the word ALPHA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"a287f842-f6f2-4a17-ab4c-820e41f498d5"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"c253c429-6ca2-4ac4-a192-ea1383e90a1e"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357521756,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"e35564a9-35f0-4a5c-8754-e5502bc6a35b"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357521756,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730447828,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730447828,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl index 95f756b06c..b01b69b45b 100644 --- a/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-multi/session.2.jsonl @@ -6,7 +6,7 @@ {"type":"subagent/descriptor","seq":4,"time":1786357521799,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Return BETA only"}} {"type":"step/start","seq":5,"time":1786357521801,"data":{"turn":1,"step":1}} {"type":"user/message","seq":6,"time":1785730447881,"data":{"content":[{"type":"text","text":"Reply with exactly the word BETA and nothing else."}],"source":{"kind":"user"},"role":"user","id":"53f6419d-8ddc-4eee-8803-5b68411336f9"},"surfaceOp":"append"} -{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"488666ba-7318-4b83-bc3f-d7700024b196"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":1786357521802,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`).\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."},{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"56f96521-b5b9-4e1f-8ca4-38ac7c212d99"},"surfaceOp":"append"} {"type":"session/title","seq":8,"time":1786357521802,"data":{"title":"Reply with exactly the word","messageSeqs":[6],"source":{"kind":"fallback"}}} {"type":"request/header","seq":9,"time":1785730447881,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":10,"time":1785730447881,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl index edbef518f1..f8b5deef29 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-report/session.jsonl @@ -1,10 +1,10 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1789000000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"e45e4b6d-dae7-42d9-9edb-d24385664025"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785730453561,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"b765ae32-73e2-4625-81ba-01095f8c83d0"}]}} {"type":"turn/start","seq":1,"time":1785821411429,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821411429,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1785730453591,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"e45e4b6d-dae7-42d9-9edb-d24385664025"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"9319509b-82df-4848-8f3a-7654804c9c45"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785730453591,"data":{"content":[{"type":"text","text":"Follow these steps exactly, then stop. 1. Call the subagent tool once with run_in_background set to true, description 'Report a finding', and prompt 'Call the report tool once with output exactly CHILD_REPORT_OK, then stop.'. 2. Reply with the single word STARTED. Do not use the bash tool."}],"source":{"kind":"user"},"role":"user","id":"b765ae32-73e2-4625-81ba-01095f8c83d0"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730453592,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations.\n\nApproval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."},{"name":"approval:policy","text":"Approval prompts are disabled in this session: actions that require approval are rejected automatically — do not request sandbox escalation (do not set `sandbox_permissions`)."}]},"role":"user","id":"1f4f5888-2068-4df0-904f-12ffb4aa3321"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730453592,"data":{"title":"Follow these steps exactly, then","messageSeqs":[4],"source":{"kind":"fallback"}}} {"type":"request/header","seq":7,"time":1785730453592,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730453593,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} @@ -13,9 +13,9 @@ {"type":"assistant/chunk","seq":11,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}}}} {"type":"assistant/chunk","seq":12,"time":1785501592851,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730453601,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"dcd1402a-35c7-4748-bfeb-8734e1018732"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730453602,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"97b897d5-0d01-4a6c-ad0c-4776c61c9c68"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730453602,"data":{"turn":1,"step":1,"callId":"call_bg_start","name":"subagent","arguments":"{\"description\": \"Report a finding\", \"prompt\": \"Call the report tool once with output exactly CHILD_REPORT_OK, then stop.\", \"run_in_background\": true}"}} -{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"b752b617-5282-40b6-bfe0-0535fb069504"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730453613,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_bg_start"},"content":[{"type":"tool-result","toolCallId":"call_bg_start","content":[{"type":"text","text":"started subagent 33333333-3333-4333-8333-333333333333"}],"isError":false}],"role":"user","id":"91fb94ce-cf3e-47ed-ab20-8f46cf4aec55"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730453613,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730453623,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":1789000000018,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -23,32 +23,35 @@ {"type":"assistant/chunk","seq":21,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}} {"type":"assistant/chunk","seq":22,"time":1785501592877,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730453628,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"d5bf8946-4dd4-464e-88de-c7d6191c7671"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730453628,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b1b1cf78-11a8-4440-b9f9-2096d15e7884"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"step/end","seq":25,"time":1785730453628,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":26,"time":1785730453629,"data":{"turn":1,"reason":{"kind":"completed"}}} -{"type":"agent/inbox/spliced","seq":27,"time":1786008565468,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2ba9ef7b-dac7-4f39-8035-35f0bf88d0e7"}]}} -{"type":"turn/start","seq":28,"time":1786008565468,"data":{"turn":2}} -{"type":"agent/inbox/spliced","seq":29,"time":1786008565468,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} -{"type":"step/start","seq":30,"time":1786008565474,"data":{"turn":2,"step":1}} -{"type":"user/message","seq":31,"time":1786008565474,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"2ba9ef7b-dac7-4f39-8035-35f0bf88d0e7"},"surfaceOp":"append"} -{"type":"assistant/chunk","seq":32,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} -{"type":"assistant/chunk","seq":33,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"REPORT_RECEIVED"}}} -{"type":"assistant/chunk","seq":34,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"REPORT_RECEIVED"}}}} -{"type":"assistant/chunk","seq":35,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} -{"type":"assistant/chunk","seq":36,"time":1786008565479,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":37,"time":1786008565479,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"REPORT_RECEIVED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0fe8d69e-6640-4bd5-b72b-70f2cf013cde"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} -{"type":"step/end","seq":38,"time":1786008565479,"data":{"turn":2,"step":1}} -{"type":"turn/end","seq":39,"time":1786008565479,"data":{"turn":2,"reason":{"kind":"completed"}}} -{"type":"agent/inbox/spliced","seq":40,"time":1786008565505,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"50d55072-bbf2-4e30-8948-1b837d8db9ee"}]}} -{"type":"turn/start","seq":41,"time":1786008565505,"data":{"turn":3}} -{"type":"agent/inbox/spliced","seq":42,"time":1786008565505,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} -{"type":"step/start","seq":43,"time":1786008565511,"data":{"turn":3,"step":1}} -{"type":"user/message","seq":44,"time":1786008565511,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"50d55072-bbf2-4e30-8948-1b837d8db9ee"},"surfaceOp":"append"} -{"type":"assistant/chunk","seq":45,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} -{"type":"assistant/chunk","seq":46,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}} -{"type":"assistant/chunk","seq":47,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}} -{"type":"assistant/chunk","seq":48,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} -{"type":"assistant/chunk","seq":49,"time":1786008565516,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":50,"time":1786008565516,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"de2ca55e-a346-4f25-911d-a33316b9866b"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[45,46,47,48,49],"surfaceOp":"append"} -{"type":"step/end","seq":51,"time":1786008565516,"data":{"turn":3,"step":1}} -{"type":"turn/end","seq":52,"time":1786008565516,"data":{"turn":3,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":27,"time":1786008565468,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"ec024a7a-5506-4ebf-a9d8-82ce01dc88b4"}]}} +{"type":"agent/inbox/spliced","seq":28,"time":1786012334062,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"Reported."}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"08101cc1-abde-49ca-9745-1d075a3911b5"}]}} +{"type":"turn/start","seq":29,"time":1786012334062,"data":{"turn":2}} +{"type":"agent/inbox/spliced","seq":30,"time":1786012334062,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}} +{"type":"agent/inbox/spliced","seq":31,"time":1786012334062,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":32,"time":1786376144953,"data":{"turn":2,"step":1}} +{"type":"user/message","seq":33,"time":1786376144953,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 reported:"},{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"subagent-report","form":"relay","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"ec024a7a-5506-4ebf-a9d8-82ce01dc88b4"},"surfaceOp":"append"} +{"type":"user/message","seq":34,"time":1786012334068,"data":{"content":[{"type":"text","text":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"Reported."}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent 33333333-3333-4333-8333-333333333333 finished and will do no further work unless you send it more.","senderSessionId":"33333333-3333-4333-8333-333333333333"},"role":"user","id":"08101cc1-abde-49ca-9745-1d075a3911b5"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":35,"time":1786376144959,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":36,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"text-delta","index":0,"text":"SUBAGENT_SETTLED_NOTED"}}} +{"type":"assistant/chunk","seq":37,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} +{"type":"assistant/chunk","seq":38,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":39,"time":1786012334073,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":40,"time":1786376144959,"data":{"turn":2,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c6fecd6e-033f-4be8-98ee-cc0733b18c83"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[35,36,37,38,39],"surfaceOp":"append"} +{"type":"step/end","seq":41,"time":1786376144960,"data":{"turn":2,"step":1}} +{"type":"turn/end","seq":42,"time":1786376144960,"data":{"turn":2,"reason":{"kind":"completed"}}} +{"type":"agent/inbox/spliced","seq":43,"time":1786376145165,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"6c0b0e51-4ad9-4c4b-bbbc-508973862b77"}]}} +{"type":"turn/start","seq":44,"time":1786012334073,"data":{"turn":3}} +{"type":"agent/inbox/spliced","seq":45,"time":1786012334073,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":46,"time":1786012334082,"data":{"turn":3,"step":1}} +{"type":"user/message","seq":47,"time":1786012334083,"data":{"content":[{"type":"text","text":"Repeat back, verbatim, the exact output the background subagent reported to you. Reply with only that text. Do not use any tools."}],"source":{"kind":"user"},"role":"user","id":"6c0b0e51-4ad9-4c4b-bbbc-508973862b77"},"surfaceOp":"append"} +{"type":"assistant/chunk","seq":48,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":49,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_REPORT_OK"}}} +{"type":"assistant/chunk","seq":50,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_REPORT_OK"}}}} +{"type":"assistant/chunk","seq":51,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":52,"time":1786012334087,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":53,"time":1786012334087,"data":{"turn":3,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_REPORT_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"98d44266-6695-482b-910c-0e1e570fe7a6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[48,49,50,51,52],"surfaceOp":"append"} +{"type":"step/end","seq":54,"time":1786012334087,"data":{"turn":3,"step":1}} +{"type":"turn/end","seq":55,"time":1786012334087,"data":{"turn":3,"reason":{"kind":"completed"}}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl index cd63d489c7..785ee28128 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl +++ b/examples/acp-agent/tests/snapshots/subagent-report/stdout.expected.jsonl @@ -2,6 +2,6 @@ {"jsonrpc":"2.0","id":2,"result":{"sessionId":"{{sessionId}}"}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"STARTED"}}}} {"jsonrpc":"2.0","id":3,"result":{"stopReason":"end_turn"}} -{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"REPORT_RECEIVED"}}}} +{"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"SUBAGENT_SETTLED_NOTED"}}}} {"jsonrpc":"2.0","method":"session/update","params":{"sessionId":"{{sessionId}}","update":{"sessionUpdate":"agent_message_chunk","content":{"type":"text","text":"CHILD_REPORT_OK"}}}} {"jsonrpc":"2.0","id":4,"result":{"stopReason":"end_turn"}} diff --git a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json index 6cecb69da2..d0b02de9dd 100644 --- a/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json +++ b/examples/acp-agent/tests/snapshots/subagent-report/tool-schemas.1.expected.json @@ -247,7 +247,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -261,7 +261,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json index 4b7d68131f..adee651183 100644 --- a/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/text-turn/tool-schemas.expected.json @@ -231,7 +231,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -245,7 +245,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json index 9f198a86a5..5050d37d16 100644 --- a/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/web-fetch/tool-schemas.expected.json @@ -231,7 +231,7 @@ }, { "name": "subagent", - "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.", + "description": "Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.", "parameters": { "type": "object", "properties": { @@ -245,7 +245,7 @@ }, "run_in_background": { "type": "boolean", - "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message." + "description": "Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message." } }, "required": [ diff --git a/examples/headless-agent/subagent-settlement.cordis.snapshot.yml b/examples/headless-agent/subagent-settlement.cordis.snapshot.yml new file mode 100644 index 0000000000..4812d676a2 --- /dev/null +++ b/examples/headless-agent/subagent-settlement.cordis.snapshot.yml @@ -0,0 +1,21 @@ +# Keyless assembled-app coverage for continuable child settlement delivery. The +# replay child deliberately never calls report; the parent can reach its final +# answer only if the continuation manager places the child's closing message in +# the parent turn without list_agents, send_message, or a Task collector. + +- id: base + name: '@deepseek-ai/cordis-plugin-include' + config: + path: ./cordis.yml + patches: + - id: llm-deepseek + name: '@deepseek-ai/dsh-llm-deepseek' + disabled: true + - insert: + - id: llm-replay + name: '@deepseek-ai/dsh-llm-replay' + +# Prevent platform scheduling from choosing a streamed-chunk interleave. The +# fence releases only after the real manager notice enters the parent inbox. +- id: settlement-fence + name: './tests/fixtures/subagent-settlement-fence.ts' diff --git a/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts b/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts new file mode 100644 index 0000000000..c2c8a87885 --- /dev/null +++ b/examples/headless-agent/tests/fixtures/subagent-settlement-fence.ts @@ -0,0 +1,43 @@ +/** + * Loader fixture that holds the parent's second step until settlement delivery. + * @module subagent-settlement-fence + */ + +import type { Context } from '@deepseek-ai/cordis' +import type {} from '@deepseek-ai/dsh-agent-loop' +import type {} from '@deepseek-ai/dsh-subagent' + +/** Fixture plugin name. */ +export const name = 'subagent-settlement-fence' + +/** + * Fence the parent's post-spawn request behind admission of the manager notice. + * + * This pins content order, not step placement: the held pre-step runs after its + * own `Inbox.claim()`, so the notice lands after step 2's claim and is claimed at + * step 3 because the child's settlement pipeline is strictly longer than the + * parent's claim path, not because a barrier forces it. + * @param ctx - assembled headless-agent context. + */ +export function apply(ctx: Context): void { + const delivered = Promise.withResolvers() + let hasDelivered = false + + ctx.effect(() => { + const disposeInbox = ctx.root.on('agent/inbox/inserted', ({ agent, message }) => { + if (agent.session.header.parentSession !== undefined || message.source.kind !== 'subagent-settled') return + hasDelivered = true + delivered.resolve(undefined) + }) + const disposeStep = ctx.root.on('agent/pre-step', async ({ agent, turn, step }, next) => { + if (agent.session.header.parentSession === undefined && turn === 1 && step === 2 && !hasDelivered) { + await delivered.promise + } + return next() + }) + return () => { + disposeStep() + disposeInbox() + } + }, 'subagent-settlement-fence.listeners') +} diff --git a/examples/headless-agent/tests/headless.snapshot.ts b/examples/headless-agent/tests/headless.snapshot.ts index 5fc337527e..61c1711566 100644 --- a/examples/headless-agent/tests/headless.snapshot.ts +++ b/examples/headless-agent/tests/headless.snapshot.ts @@ -45,6 +45,8 @@ const credentialsConfigPath = fileURLToPath(new URL('../credentials.cordis.snaps const invalidCredentialScenarioDir = join(snapshotsDir, 'invalid-credential') const ralphScenarioDir = join(snapshotsDir, 'ralph-loop') const ralphConfigPath = fileURLToPath(new URL('../ralph.cordis.snapshot.yml', import.meta.url)) +const settlementScenarioDir = join(snapshotsDir, 'subagent-settlement') +const settlementConfigPath = fileURLToPath(new URL('../subagent-settlement.cordis.snapshot.yml', import.meta.url)) const startupFailureConfigPath = fileURLToPath(new URL('./fixtures/startup-activation-error/cordis.yml', import.meta.url)) const startupFailureExpected = join(snapshotsDir, 'startup-activation-error', 'stderr.expected.txt') const binScript = fileURLToPath(new URL('./fixtures/headless-driver.ts', import.meta.url)) @@ -777,6 +779,77 @@ describe('headless stream-json snapshots', () => { expect(normalized).toBe(await readFile(streamExpected, 'utf8')) }, LOADER_SMOKE_TEST_TIMEOUT_MS) + it('delivers a continuable child result without parent polling', async () => { + const parentReplay = join(settlementScenarioDir, 'parent.replay.jsonl') + const parentOverride = join(settlementScenarioDir, 'parent.override.json') + const childReplay = join(settlementScenarioDir, 'child.replay.jsonl') + const childExpected = join(settlementScenarioDir, 'child.expected.jsonl') + const streamExpected = join(settlementScenarioDir, 'stream-json.expected.jsonl') + const task = 'Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list.' + let runCwd = '' + const result = await runLoaderSmoke({ + label: 'continuable settlement headless stream-json snapshot', + tempDirPrefix: 'headless-snapshot-subagent-settlement-', + binScript, + libBinScript: binScript, + configPath: settlementConfigPath, + binArgs: [settlementConfigPath, task], + tsconfigPath, + env: { + // The override fully supplies the parent script; the child fixture + // remains separate so replay binds it to the fresh child Session. + DSH_SNAPSHOT_FILE: parentReplay, + DSH_SNAPSHOT_OVERRIDE: parentOverride, + DSH_SNAPSHOT_CHILD_FILES: childReplay, + NODE_OPTIONS: [process.env.NODE_OPTIONS, '--disable-warning=ExperimentalWarning'].filter(Boolean).join(' '), + }, + prepare: (cwd) => { runCwd = cwd }, + inspect: async (cwd) => { + const logs = await persistedLogs(cwd) + expect(logs).toHaveLength(2) + const parent = logs.find(log => typeof log.header.parentSession !== 'string') + const child = logs.find(log => typeof log.header.parentSession === 'string') + if (parent === undefined || child === undefined) throw new Error('missing persisted parent or child log') + + const parentRecords = parseJsonl(parent.content) + const calls = parentRecords.filter(record => record.type === 'tool/call') + expect(calls.map(record => (record.data as JsonObject | undefined)?.name)).toEqual(['subagent']) + const callArguments = (calls[0]?.data as JsonObject | undefined)?.arguments + if (typeof callArguments !== 'string') throw new Error('subagent call did not persist its arguments') + expect(JSON.parse(callArguments)).toMatchObject({ run_in_background: true }) + + const notices = parentRecords.flatMap((record) => { + if (record.type !== 'agent/inbox/spliced') return [] + const inserted = (record.data as JsonObject | undefined)?.inserted + if (!Array.isArray(inserted)) return [] + return (inserted as JsonObject[]).filter((message) => { + const source = message.source as JsonObject | undefined + return source?.kind === 'subagent-settled' + }) + }) + expect(notices).toHaveLength(1) + expect(JSON.stringify(notices[0])).toContain('CHILD_RESULT') + + const context = contextFromLogs([parent.content, child.content]) + const normalizedChild = scrubRequestHeaders(normalizeSessionLog(child.content, context)) + if (refreshing) await writeFile(childExpected, normalizedChild) + expect(normalizedChild).toBe(await readFile(childExpected, 'utf8')) + expect(normalizedChild).toContain('CHILD_RESULT') + expect(normalizedChild).not.toContain('"name":"report"') + }, + }) + + expect(result.stderr).toBe('') + const records = parseJsonl(result.stdout) + expect(records.at(-1)).toMatchObject({ + type: 'result', + output: 'PARENT_RECEIVED_CHILD_RESULT', + }) + const normalized = normalizeHeadlessStream(result.stdout, runCwd) + if (refreshing) await writeFile(streamExpected, normalized) + expect(normalized).toBe(await readFile(streamExpected, 'utf8')) + }, LOADER_SMOKE_TEST_TIMEOUT_MS) + it('replays persistent PTY tools through the one-shot app', async () => { const input = JSON.parse(await readFile(join(ptyScenarioDir, 'input.json'), 'utf8')) as { steps?: { op?: unknown; text?: unknown }[] diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl index acbf5f8cc9..ed147a24e6 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.1.jsonl @@ -1,19 +1,19 @@ {"type":"session","version":0,"id":"22222222-2222-4222-8222-222222222222","createdAt":1783950001000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1} -{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"48a309a8-4004-43e9-87ea-0bd02e9f5857"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498583877,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"debf537f-cdc4-4b03-aa58-bed881573aab"}]}} {"type":"turn/start","seq":1,"time":1785821454445,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454445,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"subagent/descriptor","seq":3,"time":1785821454466,"data":{"version":2,"mode":"one-shot","provider":"spawn","label":"Check direct child"}} {"type":"step/start","seq":4,"time":1785730501506,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"48a309a8-4004-43e9-87ea-0bd02e9f5857"},"surfaceOp":"append"} -{"type":"user/message","seq":6,"time":1786358103673,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"b8be7a23-617a-4aad-93e6-d12f0593f71d"},"surfaceOp":"append"} -{"type":"session/title","seq":7,"time":1786358103673,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"user/message","seq":5,"time":1785730501506,"data":{"content":[{"type":"text","text":"Reply with exactly DIRECT_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"debf537f-cdc4-4b03-aa58-bed881573aab"},"surfaceOp":"append"} +{"type":"user/message","seq":6,"time":1786373992181,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"78fcf31b-a20e-49aa-a558-e468c776e1f4"},"surfaceOp":"append"} +{"type":"session/title","seq":7,"time":1786373992181,"data":{"title":"Reply with exactly DIRECT_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} +{"type":"request/header","seq":8,"time":1785498583897,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":9,"time":1785730501507,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":10,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":11,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"DIRECT_CHILD_OK"}}} {"type":"assistant/chunk","seq":12,"time":1783957884564,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DIRECT_CHILD_OK"}}}} {"type":"assistant/chunk","seq":13,"time":1785498583897,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":14,"time":1785730501507,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"9f582f43-25e4-4f21-a81c-0186f42fadf9"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} +{"type":"assistant/message","seq":15,"time":1785730501507,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e8fa01a4-826b-4b6d-b342-39d76e8643cf"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501507,"data":{"turn":1,"step":1}} {"type":"turn/end","seq":17,"time":1785730501507,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl index c5773c21aa..dc4340c8a6 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.2.jsonl @@ -1,19 +1,19 @@ {"type":"session","version":0,"id":"33333333-3333-4333-8333-333333333333","createdAt":1783950002000,"cwd":"{{cwd}}","parentSession":"11111111-1111-4111-8111-111111111111","origin":"subagent","delegationDepth":1} -{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"11d0804b-a6bb-4dcc-be05-6f35f81f4f13"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498584048,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"0e1bad92-814b-454d-93d9-843b938e7486"}]}} {"type":"turn/start","seq":1,"time":1785821454599,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454599,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"subagent/descriptor","seq":3,"time":1785821454618,"data":{"version":2,"mode":"one-shot","provider":"spawn"}} {"type":"step/start","seq":4,"time":1785730501645,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"11d0804b-a6bb-4dcc-be05-6f35f81f4f13"},"surfaceOp":"append"} -{"type":"user/message","seq":6,"time":1786358103827,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"0d36194f-6afa-41a4-907e-c9e0583b797a"},"surfaceOp":"append"} -{"type":"session/title","seq":7,"time":1786358103827,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"user/message","seq":5,"time":1785730501645,"data":{"content":[{"type":"text","text":"Reply with exactly WORKFLOW_CHILD_OK and nothing else."}],"source":{"kind":"user"},"role":"user","id":"0e1bad92-814b-454d-93d9-843b938e7486"},"surfaceOp":"append"} +{"type":"user/message","seq":6,"time":1786373992411,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"56005931-a16c-412c-b184-8a5a24b2a799"},"surfaceOp":"append"} +{"type":"session/title","seq":7,"time":1786373992411,"data":{"title":"Reply with exactly WORKFLOW_CHILD_OK and","messageSeqs":[5],"source":{"kind":"fallback"}}} +{"type":"request/header","seq":8,"time":1785498584067,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":9,"time":1785730501646,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":10,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":11,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"WORKFLOW_CHILD_OK"}}} {"type":"assistant/chunk","seq":12,"time":1783957884701,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"WORKFLOW_CHILD_OK"}}}} {"type":"assistant/chunk","seq":13,"time":1785498584067,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":14,"time":1785730501646,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"eb7781b5-d2ac-4a94-8e21-c9f769524a56"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} +{"type":"assistant/message","seq":15,"time":1785730501646,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"WORKFLOW_CHILD_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"090217d2-8555-4c5d-bad2-7a6f24e37f3f"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[10,11,12,13,14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501646,"data":{"turn":1,"step":1}} {"type":"turn/end","seq":17,"time":1785730501646,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl index a3faa19c4b..3c2c4aaf35 100644 --- a/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl +++ b/examples/headless-agent/tests/snapshots/advanced-toolchain/session.jsonl @@ -1,20 +1,20 @@ {"type":"session","version":0,"id":"11111111-1111-4111-8111-111111111111","createdAt":1783950000000,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"342d68c5-9a35-46dc-9d80-d18a2375a0cf"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498583746,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"a690e220-0f80-4410-9622-80a43791c4c2"}]}} {"type":"turn/start","seq":1,"time":1785821454304,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821454304,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1783957884486,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"342d68c5-9a35-46dc-9d80-d18a2375a0cf"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: try a no-op temporary Cordis Plugin named snapshot-marker; use run_code to inspect the live temporary Plugins through tools.cordis_inspect; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; stop dyn-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"a690e220-0f80-4410-9622-80a43791c4c2"},"surfaceOp":"append"} {"type":"session/title","seq":5,"time":1785498583779,"data":{"title":"Run this advanced flow exactly","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":6,"time":1785498583782,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model. Your working directory is {{cwd}}.\n\nVerify your work by running the code or tests. Keep answers brief and factual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.\n\n## Writing code for run_code\n\nPass `run_code` the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped). Inside the program:\n\n- Call tools as `await tools.name(args)` — quoted access for exotic names: `tools[\"my-tool\"](args)`. Every call resolves to the tool's typed canonical JSON value. Tool arguments must be lossless JSON.\n- A FAILED tool call rejects with `ToolCallError`, whose `toolName` identifies the failed tool and whose `message` is human-readable — `try/catch` it to handle and continue.\n- Independent read-only calls MAY overlap under `Promise.all` (safe calls run concurrently; mutating calls run alone, in submission order). Sequence dependent work with `await`.\n- Emit results with `return` and/or `console.log(...)`. ONLY what you print or return comes back to you — intermediate tool results never enter the conversation, so extract just what you need.\n\nThe available tools:\n\n```ts\ntype JsonValue = null | boolean | number | string | JsonValue[] | { [key: string]: JsonValue }\n\ninterface ToolArgsMap {\n /** Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`. */\n bash: {\n /** The bash command to execute. */\n command: string;\n /** Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\". */\n description: string;\n /** Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry. */\n timeoutMs?: number;\n /** Working directory for this command. Defaults to the session workspace; a relative path is resolved against it. */\n workdir?: string;\n /** Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc. */\n cordis_inspect: {\n /** Limit the report to one section. Omit for all sections. */\n what?: \"services\" | \"plugins\" | \"tools\" | \"temporary\" | \"api\" | \"events\";\n /** Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\". */\n name?: string;\n } & Record;\n /** Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime. */\n cordis_mount: {\n /** JavaScript body returning a temporary Plugin; evaluated now and saved nowhere. */\n code: string;\n } & Record;\n /** Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins. */\n cordis_unmount: {\n /** The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart. */\n id: string;\n } & Record;\n /** Edit an existing UTF-8 text file by replacing literal text. */\n edit: {\n /** Path to edit, resolved by the filesystem backend. */\n file_path: string;\n /** Literal text to replace. Must match exactly. */\n old_string: string;\n /** Literal replacement text. Use an empty string to delete the match. */\n new_string: string;\n /** Replace all matches. Defaults to false; when false, old_string must appear exactly once. */\n replace_all?: boolean;\n } & Record;\n /** Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op. */\n interrupt_agent: {\n /** The agent id of the running agent to interrupt. */\n agent_id: string;\n } & Record;\n /** Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools. */\n ralph: {\n /** The immutable completion objective for every fresh Ralph round. */\n objective: string;\n /** Optional positive safe-integer round cap, bounded by the deployment ceiling. */\n maxRounds?: number;\n } & Record;\n /** Read a UTF-8 text file and return line-numbered content. */\n read: {\n /** Path to read, resolved by the filesystem backend. */\n file_path: string;\n /** 1-based first line to return. Defaults to 1. */\n offset?: number;\n /** Maximum number of lines to return. Defaults to 2000. */\n limit?: number;\n } & Record;\n /** Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered. */\n send_message: {\n /** The subagent id returned when the background subagent was started. */\n subagent_id: string;\n /** The message to deliver to the subagent. */\n message: string;\n } & Record;\n /** Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill. */\n skill: {\n /** The exact skill name from the available skills list. */\n name: string;\n } & Record;\n /** Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work. */\n subagent: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs. */\n prompt: string;\n /** Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message. */\n run_in_background?: boolean;\n } & Record;\n /** Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps. */\n subagent_fork: {\n /** A short (3-5 word) description of the delegated task, for display. */\n description: string;\n /** The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new. */\n prompt: string;\n } & Record;\n /** Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops. */\n task_kill: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Optional short reason, recorded in the log and forwarded to the task. */\n reason?: string;\n } & Record;\n /** List your background tasks (running and finished) with their ids, kinds, and statuses. */\n task_list: Record;\n /** Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap. */\n task_output: {\n /** Task id returned by the tool that started the background work. */\n task_id: string;\n /** Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive. */\n wait?: boolean;\n /** Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum. */\n timeout_ms?: number;\n } & Record;\n /** Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished). */\n todo_write: {\n /** The COMPLETE task list, replacing any previous list. */\n todos: ({\n /** What the task is — a short imperative line. */\n content: string;\n /** pending (not started) | in_progress (now) | completed (done). */\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n } & Record;\n /** Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn. The workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result. Script-body hooks: - `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly. - `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages. - `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`. - `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim. Misused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`. Constraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes. */\n workflow: {\n /** The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `). */\n script: string;\n /** The workflow identity block (plain JSON — never code). */\n meta: {\n /** Short kebab-case workflow name. */\n name: string;\n /** One-line description of what the workflow does. */\n description: string;\n /** Optional guidance on when this workflow applies. */\n whenToUse?: string;\n /** Optional phase declarations matched by phase() calls. */\n phases?: ({\n /** The phase title phase() calls match by exact string. */\n title: string;\n /** Optional one-line description of the phase. */\n detail?: string;\n /** Optional provider override this phase is expected to use. */\n provider?: string;\n /** Optional model override this phase is expected to use. */\n model?: string;\n } & Record)[];\n } & Record;\n /** Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}). */\n args?: Record;\n } & Record;\n /** Create or fully replace a UTF-8 text file. */\n write: {\n /** Path to write, resolved by the filesystem backend. */\n file_path: string;\n /** Full UTF-8 text content to write. */\n content: string;\n } & Record;\n}\n\ninterface ToolOutputMap {\n bash: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"foreground\";\n exitCode: number | null;\n signal: string | null;\n timedOut: boolean;\n aborted: boolean;\n timeoutMs: number;\n stdout: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n stderr: {\n text: string;\n truncated: boolean;\n spillPath?: string;\n };\n sandbox?: {\n mode: string;\n denied: boolean;\n enforcement?: string;\n runnerFailed?: boolean;\n };\n };\n cordis_inspect: string;\n cordis_mount: {\n id: string;\n pluginName: string;\n state: \"pending\" | \"loading\" | \"active\" | \"failed\" | \"disposed\" | \"unloading\";\n provides: string[];\n waitingFor: string[];\n };\n cordis_unmount: {\n id: string;\n pluginName: string;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n ralph: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n read: {\n path: string;\n offset: number;\n lines: {\n number: number;\n text: string;\n }[];\n totalLines: number;\n };\n send_message: {\n messageId: string;\n };\n skill: {\n name: string;\n provider: string;\n resourceBase?: {\n kind: \"directory\";\n path: string;\n } | {\n kind: \"url\";\n url: string;\n } | {\n kind: \"opaque\";\n description: string;\n };\n content: string;\n };\n subagent: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n subagent_fork: {\n kind: \"background\";\n taskId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\n };\n task_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n task_list: ({\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n })[];\n task_output: {\n text: string;\n task: {\n id: string;\n kind: string;\n label: string;\n status: \"running\" | \"stopping\" | \"completed\" | \"killed\" | \"failed\";\n detail?: string;\n startedAt: number;\n finishedAt?: number;\n };\n };\n todo_write: {\n todos: ({\n content: string;\n status: \"pending\" | \"in_progress\" | \"completed\";\n })[];\n counts: {\n pending: number;\n inProgress: number;\n completed: number;\n };\n };\n workflow: {\n runId: string;\n agentsStarted: number;\n result: JsonValue;\n };\n write: {\n path: string;\n operation: \"create\" | \"update\";\n before: string | null;\n after: string;\n };\n}\n\ntype ToolName = keyof ToolOutputMap\n\ndeclare class ToolCallError extends Error {\n readonly name: \"ToolCallError\";\n readonly toolName: ToolName;\n}\n\ndeclare const tools: {\n [K in ToolName]: (args: ToolArgsMap[K]) => Promise;\n}\n```","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_inspect","description":"Inspect the live Cordis runtime in the current DSH process. Read-only. Sections: `services` (every provided ctx service and the plugin fiber that owns it), `plugins` (all live plugin fibers with their lifecycle states), `tools` (the model-facing tools currently registered, i.e. what you can call), `temporary` (only temporary Plugins created by cordis_mount: id, name, state, provided services, awaited services, and lifetime), `api` (method signatures AND argument/return type shapes for every LIVE service — read this before writing plugin code that calls a service), `events` (every harness event with its dispatch mode and exact signature — pick listener targets here). Temporary Plugins exist only in memory, remain active across later turns, and disappear after cordis_unmount, toolset unload, or DSH restart; they are not restored automatically. The `temporary` section is a subset of `plugins`. Omit `what` to get all six sections. With `what:\"api\"` or `what:\"events\"`, pass an exact `name` to narrow to one service/event and include its original source JSDoc.","parameters":{"type":"object","properties":{"what":{"type":"string","description":"Limit the report to one section. Omit for all sections.","enum":["services","plugins","tools","temporary","api","events"]},"name":{"type":"string","description":"Exact service key or event name whose original JSDoc to include; valid only with what:\"api\" or what:\"events\"."}}}},{"name":"cordis_mount","description":"Mount a temporary Cordis Plugin in the current DSH process. This creates an in-memory runtime Plugin, not an installed or configured Plugin. It remains active across later turns until cordis_unmount, toolset unload, or DSH restart. It does not create files, install a package, change cordis.yml or personal/project config, survive restart, or automatically become permanent. To keep it, ask the Agent to implement an SDK Plugin or installable profile bundle through the regular development workflow. It may affect other sessions in the same process; the sandbox is not a security boundary, and injected services reach the real runtime. `code` runs now as the body of an async JavaScript function in an isolated sandbox and MUST `return` a plugin. Two forms: FUNCTION form `return (ctx) => { … }` — declares no inject, so it can register tools, listen to events, and provide services, but reaching ANY service (e.g. ctx.bash) throws; use it only when you need no services. OBJECT form `return { name?, inject: ['bash', 'llm', …], apply(ctx) { … } }` — declares dependencies, and cordis activates the plugin only after the services exist; PREFER this form. You may reach ONLY the services you list in inject: an undeclared service throws even if it exists, because an undeclared dependency would not be cleaned up if its provider is unmounted. BEFORE calling a service from your code, read cordis_inspect what:\"api\" — it lists method signatures AND the type shapes of their arguments/returns (do not guess a field's type; e.g. a bash run's stdout is an object, not a string). Inside `apply`, use the standard cordis API: `ctx.on(event, listener)` to observe events (see cordis_inspect what:\"events\"), or call `harness.registerTool(ctx, harness.defineTool({ name, description, parameters: { text: { type: 'string', required: true } }, output: { schema: { type: 'string' }, render(_args, value) { return [{ type: 'text', text: value }] } }, async execute(args) { return args.text } }))` to give yourself a new tool — it becomes callable on your NEXT step. Tool parameters: each key IS a property — { type: 'string'|'number'|'integer'|'boolean'|'null'|'object'|'array'|'json', required?: true, description?, enum?, const?, items?, properties? }; every direct DSL object declares additionalProperties: true|false, and oneOf: [schema, schema, ...] replaces type for an exact-one union. A raw JSON-Schema { type: 'object', properties, required?: […] } wrapper is also accepted with open-by-default objects. A tool's `execute` MUST return the lossless JSON value declared by `output.schema`; `output.render(args, value)` separately returns Native/model content blocks. Temporary Plugins can COMPOSE: one Plugin may `ctx.provide('name', value)` a service and another may declare `inject: ['name']` to consume it — the consumer stays pending until the provider exists and returns to pending when the provider is unmounted. Everything registered inside `apply` is cleaned up automatically by cordis_unmount. Sandbox globals: `console` (tagged `[cordis:]`, writes through to the harness terminal), `harness.defineTool`, `harness.registerTool`, `btoa`, `atob`, `TextEncoder`, `TextDecoder`. Node APIs are DISABLED — do filesystem/network/timer work through the cordis services, never Node built-ins: `require`, `setTimeout`/`setInterval`, and `fetch` throw redirect errors; `process` and `Buffer` are undefined. Instead use inject: ['fs'] + ctx.fs for files, inject: ['web'] + ctx.web for HTTP, inject: ['bash'] + ctx.bash for processes, and inject: ['timer'] + ctx.setTimeout/ctx.setInterval for timing (fiber effects, auto-cleaned when unmounted) — cordis_inspect what:\"api\" shows what THIS runtime provides. Write PLAIN JavaScript, not TypeScript (no `as`, no type annotations). Cautions: (1) waterfall events (e.g. tools/pre-execute) hand the listener a trailing `next` callback which MUST be called — returning without `next()` SHORT-CIRCUITS the call; prefer plain notification events unless you intend to intercept. (2) Never await something that only resolves after the current turn (your code runs INSIDE a tool call of that turn — it would deadlock). (3) Your `ctx` is a restricted façade: you can register tools, observe events, provide/consume services, and use timers, but framework internals (ctx.root, ctx.fiber, ctx.extend, ctx.plugin, …) are withheld. It is not a security boundary though — the services you inject (e.g. ctx.bash) reach the real runtime.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"JavaScript body returning a temporary Plugin; evaluated now and saved nowhere."}},"required":["code"]}},{"name":"cordis_unmount","description":"Unmount a current-process temporary Plugin created by cordis_mount. Waits for its tools, listeners, services, timers, and other owned effects to clean up completely. Only dyn-N temporary ids are accepted; this cannot remove Loader, configured, or installed Plugins.","parameters":{"type":"object","properties":{"id":{"type":"string","description":"The temporary Plugin id returned by cordis_mount (for example \"dyn-1\"); valid only in this process and invalid after unmount or restart."}},"required":["id"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"run_code","description":"Execute a TypeScript program against the available tools. Write the BODY of an async function (erasable syntax only; top-level `await` and `return` work) and call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it.","parameters":{"type":"object","properties":{"code":{"type":"string","description":"The program: the body of an async TypeScript function."},"description":{"type":"string","description":"Clear, concise description of what this program does in active voice, 5-10 words (shown in the UI). Examples: \"Count TODO markers across packages\"; \"Read failing test and its fixture\"; \"Rename config key in every cordis.yml\"."}},"required":["code","description"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":7,"time":1785730501403,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":8,"time":1783950000007,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":9,"time":1783950000008,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"advanced-mount","name":"cordis_mount","argumentsDelta":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}} {"type":"assistant/chunk","seq":10,"time":1783950000009,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}}}} {"type":"assistant/chunk","seq":11,"time":1785498583784,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":12,"time":1785730501404,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0933e3da-0073-49c2-af4a-76c95ea72f02"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} +{"type":"assistant/message","seq":13,"time":1785730501404,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f0df95b4-63ae-4fdd-83b1-6d10b7fb831e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} {"type":"tool/call","seq":14,"time":1785730501404,"data":{"turn":1,"step":1,"callId":"advanced-mount","name":"cordis_mount","arguments":"{\"code\":\"return { name: 'snapshot-marker', apply() {} }\"}"}} -{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"716e9e93-6560-4f6d-ad52-0cb8e833da39"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-mount"},"content":[{"type":"tool-result","toolCallId":"advanced-mount","content":[{"type":"text","text":"Temporary Plugin dyn-1 is running (plugin \"snapshot-marker\"; available until unmounted or DSH restarts)."}],"isError":false}],"role":"user","id":"a9e79dc0-3f12-4bf2-a287-e0a3a73f9c28"}},"sourceEventSeqs":[14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1785730501413,"data":{"turn":1,"step":1}} {"type":"step/start","seq":17,"time":1785730501423,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":18,"time":1783950000017,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -22,11 +22,11 @@ {"type":"assistant/chunk","seq":20,"time":1783950000019,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}}}} {"type":"assistant/chunk","seq":21,"time":1785498583804,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":22,"time":1785730501424,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0a6cdb2d-c83c-4b7c-af56-2bf3fbece4ae"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"} +{"type":"assistant/message","seq":23,"time":1785730501424,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4aff202c-1298-49be-a8b0-3d53fc105e71"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"} {"type":"tool/call","seq":24,"time":1785730501424,"data":{"turn":1,"step":2,"callId":"advanced-code","name":"run_code","arguments":"{\"code\": \"return await tools.cordis_inspect({ what: 'temporary' })\", \"description\": \"Run the scripted inspection program\"}"}} {"type":"tool/code-dispatch-start","seq":25,"time":1785730501473,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"}}} {"type":"tool/code-dispatch","seq":26,"time":1785730501474,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_inspect","arguments":{"what":"temporary"},"isError":false,"content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}]}} -{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"045ef2ea-7cfb-4a43-a6ef-d4ef7ad979e1"}},"sourceEventSeqs":[24],"surfaceOp":"append"} +{"type":"tool/result","seq":27,"time":1785730501475,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"## Temporary Plugins\n- Temporary Plugin dyn-1: snapshot-marker [running] — provides: none; waiting for: none; lifetime: until unmounted or DSH restarts"}],"isError":false}],"role":"user","id":"bbbb0d25-2353-4bdb-a329-5846bd82fa9d"}},"sourceEventSeqs":[24],"surfaceOp":"append"} {"type":"step/end","seq":28,"time":1785730501475,"data":{"turn":1,"step":2}} {"type":"step/start","seq":29,"time":1785730501483,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":30,"time":1783950000029,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -34,9 +34,9 @@ {"type":"assistant/chunk","seq":32,"time":1785037378923,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}}}} {"type":"assistant/chunk","seq":33,"time":1785498583869,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":34,"time":1785730501484,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"26ff28fa-fab8-4e92-a72f-d3a5b3352573"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"} +{"type":"assistant/message","seq":35,"time":1785730501484,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"f481573d-7672-42e5-aca0-621eb19f0740"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[30,31,32,33,34],"surfaceOp":"append"} {"type":"tool/call","seq":36,"time":1785730501484,"data":{"turn":1,"step":3,"callId":"advanced-direct-child","name":"subagent","arguments":"{\"description\":\"Check direct child\",\"prompt\":\"Reply with exactly DIRECT_CHILD_OK and nothing else.\"}"}} -{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"e33536a0-0801-4f51-a633-baac1167cfee"}},"sourceEventSeqs":[36],"surfaceOp":"append"} +{"type":"tool/result","seq":37,"time":1785730501508,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"advanced-direct-child"},"content":[{"type":"tool-result","toolCallId":"advanced-direct-child","content":[{"type":"text","text":"DIRECT_CHILD_OK"}],"isError":false}],"role":"user","id":"f83932ec-a8ed-4f40-957c-98c2fe467e93"}},"sourceEventSeqs":[36],"surfaceOp":"append"} {"type":"step/end","seq":38,"time":1785730501508,"data":{"turn":1,"step":3}} {"type":"step/start","seq":39,"time":1785730501521,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":40,"time":1783957884594,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -44,9 +44,9 @@ {"type":"assistant/chunk","seq":42,"time":1785037378946,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}}}} {"type":"assistant/chunk","seq":43,"time":1785498583919,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":44,"time":1785730501522,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"492fd6ab-4937-4083-a8e2-395d5d254c3e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"} +{"type":"assistant/message","seq":45,"time":1785730501522,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"452b40e7-0d9d-4611-aa86-6c03c05f4420"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[40,41,42,43,44],"surfaceOp":"append"} {"type":"tool/call","seq":46,"time":1785730501522,"data":{"turn":1,"step":4,"callId":"advanced-workflow","name":"workflow","arguments":"{\"script\":\"phase('Delegate')\\nconst reply = await agent('Reply with exactly WORKFLOW_CHILD_OK and nothing else.', { label: 'workflow-child' })\\nreturn { reply }\",\"meta\":{\"name\":\"advanced-headless-snapshot\",\"description\":\"exercise one workflow child through the headless agent\"}}"}} -{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"31627ac4-1b5d-4213-814c-95599db9a215"}},"sourceEventSeqs":[46],"surfaceOp":"append"} +{"type":"tool/result","seq":47,"time":1785730501647,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"advanced-workflow"},"content":[{"type":"tool-result","toolCallId":"advanced-workflow","content":[{"type":"text","text":"workflow \"advanced-headless-snapshot\" completed (1 agent).\nReturn value:\n{\n \"reply\": \"WORKFLOW_CHILD_OK\"\n}"}],"isError":false}],"role":"user","id":"b7785180-8d83-47a3-937c-307b8725662d"}},"sourceEventSeqs":[46],"surfaceOp":"append"} {"type":"step/end","seq":48,"time":1785730501648,"data":{"turn":1,"step":4}} {"type":"step/start","seq":49,"time":1785730501660,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":50,"time":1783957884719,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -54,9 +54,9 @@ {"type":"assistant/chunk","seq":52,"time":1785037379534,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}}}} {"type":"assistant/chunk","seq":53,"time":1785498584085,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":54,"time":1785730501661,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0b2fa8a0-7559-4fcf-a34d-bb190f8d9f48"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"} +{"type":"assistant/message","seq":55,"time":1785730501661,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"9d62978e-ba44-465f-8fb0-f08e55de8437"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[50,51,52,53,54],"surfaceOp":"append"} {"type":"tool/call","seq":56,"time":1785730501661,"data":{"turn":1,"step":5,"callId":"advanced-unmount","name":"cordis_unmount","arguments":"{\"id\":\"dyn-1\"}"}} -{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"cfc46278-d37c-4ced-8763-200e37d3290f"}},"sourceEventSeqs":[56],"surfaceOp":"append"} +{"type":"tool/result","seq":57,"time":1785730501668,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-unmount"},"content":[{"type":"tool-result","toolCallId":"advanced-unmount","content":[{"type":"text","text":"Temporary Plugin dyn-1 was unmounted and removed."}],"isError":false}],"role":"user","id":"a771d283-634b-4c0f-98a0-83b6e98b9966"}},"sourceEventSeqs":[56],"surfaceOp":"append"} {"type":"step/end","seq":58,"time":1785730501668,"data":{"turn":1,"step":5}} {"type":"step/start","seq":59,"time":1785730501678,"data":{"turn":1,"step":6}} {"type":"assistant/chunk","seq":60,"time":1783957884720,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -64,6 +64,6 @@ {"type":"assistant/chunk","seq":62,"time":1785037379541,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ADVANCED_HEADLESS_OK"}}}} {"type":"assistant/chunk","seq":63,"time":1785498584102,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":64,"time":1785730501679,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4c1bdf08-4721-4c52-9f75-56b2c2028c7d"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"} +{"type":"assistant/message","seq":65,"time":1785730501679,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"text","text":"ADVANCED_HEADLESS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3b822282-edac-4abb-9a4c-e91e5724764f"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[60,61,62,63,64],"surfaceOp":"append"} {"type":"step/end","seq":66,"time":1785730501679,"data":{"turn":1,"step":6}} {"type":"turn/end","seq":67,"time":1785730501679,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl index 5c68474179..0ab6d3f13e 100644 --- a/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl +++ b/examples/headless-agent/tests/snapshots/compaction-recovery/session.jsonl @@ -1,32 +1,32 @@ {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"c5cc1fd3-c9d5-4b49-90b0-34da051cb71e"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1786123401613,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"ab1f37e8-5e1b-4b15-a980-17ee9449b35d"}]}} {"type":"turn/start","seq":1,"time":1786123401614,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1786123401614,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1786123401667,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"c5cc1fd3-c9d5-4b49-90b0-34da051cb71e"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1786123401667,"data":{"content":[{"type":"text","text":"Establish a durable compaction premise before continuing. Record every part of this historical evidence: the snapshot uses keyless replay; persistence uses JSONL; the assembled headless application loads its real Cordis composition; model-visible inputs remain logged; tool calls and results remain paired and ordered; context overflow retains the original failure while recovery is attempted; compaction opens with compact/start and closes with compact/end; a successful auxiliary summary records compact/summary provenance; the replacement surface shadows only an older balanced range; the checkpoint remains smaller than the history it replaces; the newest tool result remains verbatim; the retried request sees that checkpoint; the final response proves the same turn continued; deterministic snapshot evidence stays separate from the live-provider smoke; and no external API key is needed. Emit one alpha marker through bash, then finish the task after any required recovery with the exact words COMPACTION RECOVERED."}],"source":{"kind":"user"},"role":"user","id":"ab1f37e8-5e1b-4b15-a980-17ee9449b35d"},"surfaceOp":"append"} {"type":"session/title","seq":5,"time":1786123401667,"data":{"title":"Establish a durable compaction premise","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":6,"time":1786123401668,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":7,"time":1786123401669,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash","contextWindow":128000}} {"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"call_compaction_marker","name":"bash","argumentsDelta":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}} {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}}}} {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":24,"outputTokens":6}}}} {"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e29c0ce2-2826-4269-a047-a831aec5a4ee"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} +{"type":"assistant/message","seq":13,"time":1786123401680,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"2cf46d07-dbd9-4aad-ba5e-5954eb1a7ea4"},"usage":{"inputTokens":24,"outputTokens":6}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} {"type":"tool/call","seq":14,"time":1786123401680,"data":{"turn":1,"step":1,"callId":"call_compaction_marker","name":"bash","arguments":"{\"command\":\"printf 'alpha\\n'\",\"description\":\"Emit compaction premise marker\"}"}} -{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"8c3b43ac-1a3b-4955-ad40-78a1d6343e2d"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"tool/result","seq":15,"time":1786123401700,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"call_compaction_marker"},"content":[{"type":"tool-result","toolCallId":"call_compaction_marker","content":[{"type":"text","text":"alpha\n"}],"isError":false}],"role":"user","id":"a1c85dd0-920f-40f5-9c61-b33bc255517d"}},"sourceEventSeqs":[14],"surfaceOp":"append"} {"type":"step/end","seq":16,"time":1786123401700,"data":{"turn":1,"step":1}} {"type":"step/start","seq":17,"time":1786123401710,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":18,"time":1786123401715,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"error","failure":{"message":"snapshot request exceeded the model context window","code":"CONTEXT_WINDOW_EXCEEDED"}}}}} {"type":"compact/start","seq":19,"time":1786123401715,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}} {"type":"compact/summary","seq":20,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","summary":[{"type":"text","text":"The request established a durable compaction premise."}],"rawOutput":[{"type":"text","text":"The request established a durable compaction premise."}],"llmStreamCall":true,"shadowedRange":{"start":4,"end":4},"shadowedSeqs":[4],"shadowedTokenCount":264,"provider":"deepseek-official","model":"deepseek-v4-flash","maxTokens":32,"usage":{"inputTokens":20,"outputTokens":4}}} -{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":""}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"bb3fd14b-4307-4550-9aa9-3326c9ebd92f"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}} +{"type":"user/message","seq":21,"time":1786123401725,"data":{"content":[{"type":"text","text":"This is an automatically generated checkpoint condensing an earlier span of the conversation to free up context. Treat the captured context as established background and build on it without restating it. Continue the task directly from the messages that follow, without acknowledging this checkpoint.\n\n"},{"type":"text","text":"The request established a durable compaction premise."},{"type":"text","text":""}],"source":{"kind":"plugin","plugin":"compact","compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1"},"role":"user","id":"1541cd3a-8b31-45d2-a44e-8034d844cd8e"},"sourceEventSeqs":[19,20,4],"surfaceOp":{"op":"replace","start":4,"end":4}} {"type":"compact/end","seq":22,"time":1786123401725,"data":{"compactionId":"338e88fa-e78b-4d49-bd38-8f919e85f5e1","turn":1}} {"type":"assistant/chunk","seq":23,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} {"type":"assistant/chunk","seq":24,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"text-delta","index":0,"text":"COMPACTION RECOVERED"}}} {"type":"assistant/chunk","seq":25,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"COMPACTION RECOVERED"}}}} {"type":"assistant/chunk","seq":26,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":20,"outputTokens":4}}}} {"type":"assistant/chunk","seq":27,"time":1786123401730,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3712f33d-6371-454f-bf8e-707cb5c74b32"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"} +{"type":"assistant/message","seq":28,"time":1786123401730,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"COMPACTION RECOVERED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3ac58347-7e9a-442e-84aa-734c071142c2"},"usage":{"inputTokens":20,"outputTokens":4}},"sourceEventSeqs":[23,24,25,26,27],"surfaceOp":"append"} {"type":"step/end","seq":29,"time":1786123401730,"data":{"turn":1,"step":2}} {"type":"turn/end","seq":30,"time":1786123401730,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl index bfd47888f9..2fd78968ad 100644 --- a/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl +++ b/examples/headless-agent/tests/snapshots/pty-tools/session.jsonl @@ -1,21 +1,21 @@ {"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","delegationDepth":0} -{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"d4746327-c682-4874-b8b2-bbc81e98ea03"}]}} +{"type":"agent/inbox/spliced","seq":0,"time":1785498587408,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"cb953265-fb22-488e-91cb-e7f250313654"}]}} {"type":"turn/start","seq":1,"time":1785821457966,"data":{"turn":1}} {"type":"agent/inbox/spliced","seq":2,"time":1785821457966,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}} -{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"d4746327-c682-4874-b8b2-bbc81e98ea03"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"8256fc51-472b-4d50-86ea-b50cef7b1ef6"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785498587436,"data":{"content":[{"type":"text","text":"Exercise the six PTY tools in order, including one missing-session signal error, then reply DONE."}],"source":{"kind":"user"},"role":"user","id":"cb953265-fb22-488e-91cb-e7f250313654"},"surfaceOp":"append"} +{"type":"user/message","seq":5,"time":1785730504659,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nCurrent DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"sandbox:policy","text":"Current DSH file policy: danger-full-access. The DSH file sandbox does not restrict file modifications by available operations."}]},"role":"user","id":"cebb0669-2ea4-46dc-988d-540ff52c6ae6"},"surfaceOp":"append"} {"type":"session/title","seq":6,"time":1785730504659,"data":{"title":"Exercise the six PTY tools","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: you receive only its subagent id, never its result, and it works on its own. Use this for work whose result you do not need returned by this call; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call never returns its result; send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} +{"type":"request/header","seq":7,"time":1785498587438,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"You are an AI agent powered by the DeepSeek Harness SDK.\n\nYou are headless-agent, a coding assistant powered by the deepseek-v4-flash model.\n\nVerify your work by running the code or tests. Keep answers brief and\nfactual.\n\n\nUse the read tool — not shell commands like cat — to inspect text files. Results include line numbers. Use offset and limit to continue reading large files.\n\nUse the write tool to create files or completely replace file contents. Existing files are overwritten, so read an existing file first (the default fs-policy requires it) and prefer edit for targeted changes.\n\nUse the edit tool for targeted changes to existing UTF-8 text files. It replaces literal old_string with new_string; by default old_string must appear exactly once. If old_string appears multiple times, provide a more specific old_string or set replace_all to true. Read the file first (the default fs-policy requires it), unless you just created or edited it in this session.\n\nCheck the [exit code: N] marker on every bash result; investigate failures before moving on.\n\nUse a terminal session only when work needs persistent terminal state or interactive stdin; prefer bash/read/write/edit for bounded one-shot operations. Track every terminal session id and close sessions that no longer matter. An inferred_idle or timeout result does not prove the foreground command exited.\n\nTrack every background task id you start. You are notified in-session when a task finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running task's work. Before giving a final answer, collect every still-relevant task with task_output (set wait: true only when you are genuinely blocked on it), and task_kill tasks that stopped mattering.\n\nUse the workflow tool ONLY when the user explicitly asks for a workflow or for large multi-agent orchestration: you write a JavaScript script (the tool description documents the exact format) that fans work out across many subagents with phases and structured results. For one or two delegations, prefer plain subagent calls.\n\nUse the ralph tool ONLY when the direct human explicitly asks for a Ralph loop or fresh-agent iterative execution. Each Ralph round starts a fresh child with no conversation seed and uses the shared workspace as durable memory. Completion and blockers are worker reports, not independent evaluation. Use same-session goal tools for ordinary long-running objectives, and plain subagents or workflows for bounded delegation and fan-out.","tools":[{"name":"bash","description":"Execute a bash command (`bash -c`) and return its stdout/stderr. Each call runs in a fresh shell: no state (cwd, variables, functions) persists between calls — pass `workdir` instead of using `cd`. Non-zero exits are reported as `[exit code: N]`. Current harness environment facts are exposed through managed `$DSH_*` variables; inspect them when needed. Commands may run under a file sandbox; a blocked file operation is reported as `[sandbox: file access denied under mode]` — a policy denial, not a bug in the command; do not retry another way. Long output is truncated to its tail; the full output is saved to a file whose path is reported when available. Set `run_in_background: true` for long-running commands: the call returns a task id immediately; read its output with `task_output` and stop it with `task_kill`.","parameters":{"type":"object","properties":{"command":{"type":"string","description":"The bash command to execute."},"description":{"type":"string","description":"Clear, concise description of what this command does in active voice, 5-10 words (shown in the UI). Examples: \"ls\" → \"List files in current directory\"; \"git status\" → \"Show working tree status\"; \"npm install\" → \"Install package dependencies\"."},"timeoutMs":{"type":"number","description":"Timeout in milliseconds. The executor applies its configured default and cap, and kills the command on expiry."},"workdir":{"type":"string","description":"Working directory for this command. Defaults to the session workspace; a relative path is resolved against it."},"run_in_background":{"type":"boolean","description":"Run in the background and return a task id immediately (collect with task_output, stop with task_kill). No timeout applies."}},"required":["command","description"]}},{"name":"edit","description":"Edit an existing UTF-8 text file by replacing literal text.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to edit, resolved by the filesystem backend."},"old_string":{"type":"string","description":"Literal text to replace. Must match exactly."},"new_string":{"type":"string","description":"Literal replacement text. Use an empty string to delete the match."},"replace_all":{"type":"boolean","description":"Replace all matches. Defaults to false; when false, old_string must appear exactly once."}},"required":["file_path","old_string","new_string"]}},{"name":"interrupt_agent","description":"Request cancellation of a background agent's current turn by its agent id. The target may be your direct child or a deeper agent created under you. Only the current turn stops: messages already queued for the agent stay parked until a later send_message, agents it started keep running, and the agent itself stays available for follow-ups. This call returns as soon as the stop request is accepted, so the target may keep running briefly; interrupting an agent that already finished is an accepted no-op.","parameters":{"type":"object","properties":{"agent_id":{"type":"string","description":"The agent id of the running agent to interrupt."}},"required":["agent_id"]}},{"name":"ralph","description":"Run a foreground fresh-agent Ralph loop toward one immutable objective. Use only when the direct human explicitly asks for Ralph or fresh-agent iteration. Each round opens a new child with no parent conversation or prior child session; the shared workspace is long-term memory, and only a bounded structured report crosses rounds. The call returns when a worker reports completion or a concrete blocker, or at the round limit. Ordinary long-running same-session work belongs to goal tools.","parameters":{"type":"object","properties":{"objective":{"type":"string","description":"The immutable completion objective for every fresh Ralph round."},"maxRounds":{"type":"number","description":"Optional positive safe-integer round cap, bounded by the deployment ceiling."}},"required":["objective"]}},{"name":"read","description":"Read a UTF-8 text file and return line-numbered content.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to read, resolved by the filesystem backend."},"offset":{"type":"number","description":"1-based first line to return. Defaults to 1."},"limit":{"type":"number","description":"Maximum number of lines to return. Defaults to 2000."}},"required":["file_path"]}},{"name":"send_message","description":"Send a message to a background subagent by its subagent id, continuing the same conversation. It becomes the subagent's next turn: if it is still working, the message waits until its current turn finishes, so it cannot redirect work already underway. This call returns no answer from the subagent — only confirmation that the message was delivered — so use it to give it more work. A failure means the message was NOT delivered.","parameters":{"type":"object","properties":{"subagent_id":{"type":"string","description":"The subagent id returned when the background subagent was started."},"message":{"type":"string","description":"The message to deliver to the subagent."}},"required":["subagent_id","message"]}},{"name":"skill","description":"Load the full instructions for an available skill. Call this with the exact skill name from the session skill catalog before acting on a task that names or clearly matches that skill.","parameters":{"type":"object","properties":{"name":{"type":"string","description":"The exact skill name from the available skills list."}},"required":["name"]}},{"name":"subagent","description":"Delegate a self-contained task to a subagent (a separate agent that works in its own context) and return its final result. Use this to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent runs to completion and you receive only its final answer, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. Set `run_in_background: true` to start a background subagent that keeps its conversation: this call returns only its subagent id, and the subagent works on its own from there. You are told when it finishes, so never poll or wait on it; `send_message` sends it more work.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The complete, self-contained task for the subagent. It does not share this conversation's context, so include everything it needs."},"run_in_background":{"type":"boolean","description":"Run as a background subagent that keeps its conversation and return only its subagent id. This call does not wait for it; you are told when it finishes. Send it more work with send_message."}},"required":["description","prompt"]}},{"name":"subagent_fork","description":"Delegate a task to a subagent that inherits this conversation: a child agent seeded with all completed turns so far (it does not see the current in-flight turn), returning only its final result. Use this when the subtask builds on this conversation's context — a follow-up analysis, a review, a continuation — without consuming this conversation's context for the work itself. You receive only its final answer, not its intermediate steps.","parameters":{"type":"object","properties":{"description":{"type":"string","description":"A short (3-5 word) description of the delegated task, for display."},"prompt":{"type":"string","description":"The task for the subagent. It already sees this conversation's completed turns, so build on them freely and state only what is new."}},"required":["description","prompt"]}},{"name":"task_kill","description":"Request cancellation of a running background task by task id. Returns immediately; the task settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"reason":{"type":"string","description":"Optional short reason, recorded in the log and forwarded to the task."}},"required":["task_id"]}},{"name":"task_list","description":"List your background tasks (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"task_output","description":"Read a background task. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Every response ends with `[status: ...]`. Reads are non-blocking unless `wait: true`, which waits up to the configured cap.","parameters":{"type":"object","properties":{"task_id":{"type":"string","description":"Task id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the task reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the task alive."},"timeout_ms":{"type":"number","description":"Max wait in milliseconds (only meaningful with wait: true). Defaults to the configured wait timeout; capped by the configured maximum."}},"required":["task_id"]}},{"name":"terminal_close","description":"Close one persistent terminal and wait until its captured owned process tree is gone.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."}},"required":["sessionId"]}},{"name":"terminal_list","description":"List persistent terminal sessions owned by the current agent.","parameters":{"type":"object","properties":{}}},{"name":"terminal_open","description":"Create a persistent, owner-isolated terminal session from a registered backend type. Use this for shell or REPL state that must survive across tool calls.","parameters":{"type":"object","properties":{"type":{"type":"string","description":"Registered terminal backend type, usually \"shell\"."},"name":{"type":"string","description":"Optional owner-local display name such as \"main\" or \"gdb\"."},"cwd":{"type":"string","description":"Initial working directory. Defaults to the deployment workspace root."}},"required":["type"]}},{"name":"terminal_read","description":"Read a bounded page of retained output from a persistent terminal without sending input.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"offset":{"type":"number","description":"Newest-relative line offset (default 0)."},"count":{"type":"number","description":"Requested line count (default 500; backend caps apply)."}},"required":["sessionId"]}},{"name":"terminal_send","description":"Send text to a persistent terminal. By default Enter is submitted and the call waits for a prompt, stdin wait, output silence, timeout, or session exit. Background mode returns a task id for task_output/task_kill.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id returned by terminal_open or terminal_list."},"text":{"type":"string","description":"UTF-8 text to write to the terminal."},"submit":{"type":"boolean","description":"Submit Enter after text (default true). Set false for control characters or incomplete REPL input."},"run_in_background":{"type":"boolean","description":"Return a task id immediately; collect with task_output or stop with task_kill."}},"required":["sessionId","text"]}},{"name":"terminal_signal","description":"Send an allowed signal to the current foreground process group of a persistent terminal.","parameters":{"type":"object","properties":{"sessionId":{"type":"string","description":"Terminal session id."},"signal":{"type":"string","description":"Signal to deliver. Shell-targeted SIGKILL is rejected; use terminal_close.","enum":["SIGINT","SIGTERM","SIGKILL","SIGTSTP","SIGHUP"]}},"required":["sessionId","signal"]}},{"name":"todo_write","description":"Record and update a structured task list for the current work. Send the ENTIRE list every call — it REPLACES the previous list (there are no partial updates, no per-item edits). Use it to plan multi-step work and show progress: add one todo per concrete step before you start. Mark every todo being actively worked on `in_progress` — several at once when work genuinely runs in parallel (e.g. concurrent subagents or background commands), one for sequential work; while work remains, at least one task should be `in_progress`. Mark a todo `completed` the moment it is done (do not batch completions), and allow no `in_progress` item only once all work is complete. Skip the list for trivial single-step tasks. Statuses: `pending` (not started), `in_progress` (being worked on now), `completed` (finished).","parameters":{"type":"object","properties":{"todos":{"type":"array","description":"The COMPLETE task list, replacing any previous list.","items":{"type":"object","additionalProperties":false,"properties":{"content":{"type":"string","description":"What the task is — a short imperative line."},"status":{"type":"string","description":"pending (not started) | in_progress (now) | completed (done).","enum":["pending","in_progress","completed"]}},"required":["content","status"]}}},"required":["todos"]}},{"name":"workflow","description":"Run a JavaScript workflow script that orchestrates subagents at scale. Use this for work that fans out across many independent pieces — an audit over many files, a migration, multi-angle research, adversarial verification of findings — where you write the orchestration as a script instead of delegating turn by turn.\n\nThe workflow's identity rides the `meta` parameter as JSON: required `name` (short kebab-case) and `description` strings, optional `whenToUse` string and `phases` array (`{title, detail?, provider?, model?}`). The `script` parameter is the plain JavaScript body ONLY (NOT TypeScript, and NO `export const meta` statement — meta is a parameter, not code), running with top-level await; end with `return ` — the value must be JSON-serializable and is this tool's result.\n\nScript-body hooks:\n- `agent(prompt, opts?): Promise` — run one subagent to completion. Without `opts.schema` it resolves to the child's final text; with `opts.schema` (an object-rooted JSON Schema using ONLY type/properties/required/additionalProperties/items/enum/const/oneOf — no pattern/format/numeric bounds) it resolves to the validated object. Resolves `null` when the child fails (filter with `.filter(Boolean)`). Other opts: `label` (display), `phase` (progress group), and independent `provider`/`model` LLM target overrides (either may be provided alone). Anything else (`effort`/`isolation`/`agentType`) is rejected loudly.\n- `pipeline(items, ...stages): Promise` — run each item through the stages independently with NO barrier between stages (prefer this for multi-stage work). Each stage receives `(prev, item, index)`. An ordinary stage throw drops that ITEM to `null` and skips its remaining stages.\n- `parallel(thunks): Promise` — run zero-argument functions concurrently and await ALL of them (a barrier; use only when a stage genuinely needs every prior result together). A throwing thunk resolves to `null`.\n- `phase(title)` — start a progress phase; `log(message)` — narrate progress; `args` — the tool call's `args` input, verbatim.\n\nMisused hooks (bad arguments, unknown options, unsupported schemas, tripped caps) throw errors that ALWAYS kill the script — they never dissolve into a per-item `null`.\n\nConstraints: concurrency and total-agent caps apply; no filesystem, network, timers, or Node.js APIs are provided — the agents do the work, the script only coordinates them. The run executes in the foreground: this call returns when the whole script finishes.","parameters":{"type":"object","properties":{"script":{"type":"string","description":"The plain-JS workflow script body (top-level await allowed; NO `export const meta` statement; end with `return `)."},"meta":{"type":"object","description":"The workflow identity block (plain JSON — never code).","additionalProperties":true,"properties":{"name":{"type":"string","description":"Short kebab-case workflow name."},"description":{"type":"string","description":"One-line description of what the workflow does."},"whenToUse":{"type":"string","description":"Optional guidance on when this workflow applies."},"phases":{"type":"array","description":"Optional phase declarations matched by phase() calls.","items":{"type":"object","additionalProperties":true,"properties":{"title":{"type":"string","description":"The phase title phase() calls match by exact string."},"detail":{"type":"string","description":"Optional one-line description of the phase."},"provider":{"type":"string","description":"Optional provider override this phase is expected to use."},"model":{"type":"string","description":"Optional model override this phase is expected to use."}},"required":["title"]}}},"required":["name","description"]},"args":{"type":"object","description":"Optional JSON input exposed to the script as the `args` global (wrap a bare list as a field, e.g. {\"files\": [...]}).","additionalProperties":true}},"required":["script","meta"]}},{"name":"write","description":"Create or fully replace a UTF-8 text file.","parameters":{"type":"object","properties":{"file_path":{"type":"string","description":"Path to write, resolved by the filesystem backend."},"content":{"type":"string","description":"Full UTF-8 text content to write."}},"required":["file_path","content"]}}]},"reason":"initial"}} {"type":"request/context","seq":8,"time":1785730504660,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} {"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} {"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"pty-spawn","name":"terminal_open","argumentsDelta":"{\"type\":\"shell\",\"name\":\"main\"}"}}} {"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}}}} {"type":"assistant/chunk","seq":12,"time":1785498587439,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":13,"time":1785730504661,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"4f71f30a-33e5-4c25-8e66-aa93c993a7e8"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} +{"type":"assistant/message","seq":14,"time":1785730504661,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0ba2d7ce-8feb-410c-a223-c45374a7e203"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} {"type":"tool/call","seq":15,"time":1785730504662,"data":{"turn":1,"step":1,"callId":"pty-spawn","name":"terminal_open","arguments":"{\"type\":\"shell\",\"name\":\"main\"}"}} -{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"e5c2f208-b939-4715-984e-a4537effbbff"}},"sourceEventSeqs":[15],"surfaceOp":"append"} +{"type":"tool/result","seq":16,"time":1785730504671,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"pty-spawn"},"content":[{"type":"tool-result","toolCallId":"pty-spawn","content":[{"type":"text","text":"started terminal session pty-1 (main) [type: shell]\ndsh> "}],"isError":false}],"role":"user","id":"d5d25a8d-2980-4b77-99d4-72883cc36f6e"}},"sourceEventSeqs":[15],"surfaceOp":"append"} {"type":"step/end","seq":17,"time":1785730504671,"data":{"turn":1,"step":1}} {"type":"step/start","seq":18,"time":1785730504679,"data":{"turn":1,"step":2}} {"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -23,9 +23,9 @@ {"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}}}} {"type":"assistant/chunk","seq":22,"time":1785498587457,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":23,"time":1785730504680,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b1af47f1-f976-41fc-b6f1-8ddc86f2aecd"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} +{"type":"assistant/message","seq":24,"time":1785730504680,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"022de086-1aa9-4c05-89f5-a56c63724fd6"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} {"type":"tool/call","seq":25,"time":1785730504680,"data":{"turn":1,"step":2,"callId":"pty-send","name":"terminal_send","arguments":"{\"sessionId\":\"pty-1\",\"text\":\"printf 'PTY_OK\\\\n'\"}"}} -{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"82baaf37-cfd3-4c59-9697-7a1b36fffa38"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"} +{"type":"tool/result","seq":26,"time":1785730504688,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"pty-send"},"content":[{"type":"tool-result","toolCallId":"pty-send","content":[{"type":"text","text":"K\ndsh> \n[wait: stdin_read]\n[session: running]\n[output truncated]"}],"isError":false}],"role":"user","id":"b1260dd8-eff9-433c-96cb-9604b1b6a64e"},"meta":{"viewport":"printf 'PTY_OK\\n'\nPTY_OK\ndsh> ","waitReason":"stdin_read","sessionStatus":{"kind":"running"},"truncated":false}},"sourceEventSeqs":[25],"surfaceOp":"append"} {"type":"step/end","seq":27,"time":1785730504688,"data":{"turn":1,"step":2}} {"type":"step/start","seq":28,"time":1785730504696,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -33,9 +33,9 @@ {"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}}}} {"type":"assistant/chunk","seq":32,"time":1785498587473,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":33,"time":1785730504697,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"3de44e0c-f4b8-493e-842b-6aaa029569d0"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} +{"type":"assistant/message","seq":34,"time":1785730504697,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ce06d9e5-07cf-4d32-a085-5587038d7a92"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} {"type":"tool/call","seq":35,"time":1785730504697,"data":{"turn":1,"step":3,"callId":"pty-read","name":"terminal_read","arguments":"{\"sessionId\":\"pty-1\",\"offset\":0,\"count\":20}"}} -{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"178561b8-fa84-4dc0-9bc8-d183a722d36a"}},"sourceEventSeqs":[35],"surfaceOp":"append"} +{"type":"tool/result","seq":36,"time":1785730504704,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"pty-read"},"content":[{"type":"tool-result","toolCallId":"pty-read","content":[{"type":"text","text":"dsh> printf 'PTY_OK\\n'\nPTY_OK\ndsh> \n[lines: 0-3 of 3]"}],"isError":false}],"role":"user","id":"8249c9b0-d783-4a5a-b60c-a4fe35a66526"}},"sourceEventSeqs":[35],"surfaceOp":"append"} {"type":"step/end","seq":37,"time":1785730504704,"data":{"turn":1,"step":3}} {"type":"step/start","seq":38,"time":1785730504712,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":39,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -43,9 +43,9 @@ {"type":"assistant/chunk","seq":41,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}}}} {"type":"assistant/chunk","seq":42,"time":1785498587489,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":43,"time":1785730504713,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"fc614320-c742-470b-a9fc-902fd67d5cb4"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} +{"type":"assistant/message","seq":44,"time":1785730504713,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6bc7d9c0-20d3-4556-aa55-8d2f43f004a4"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} {"type":"tool/call","seq":45,"time":1785730504713,"data":{"turn":1,"step":4,"callId":"pty-signal","name":"terminal_signal","arguments":"{\"sessionId\":\"pty-missing\",\"signal\":\"SIGINT\"}"}} -{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"3f047431-06fc-4201-b4cd-b611b6741ce3"}},"sourceEventSeqs":[45],"surfaceOp":"append"} +{"type":"tool/result","seq":46,"time":1785730504721,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"pty-signal"},"content":[{"type":"tool-result","toolCallId":"pty-signal","content":[{"type":"text","text":"Error: unknown PTY session pty-missing"}],"isError":true}],"role":"user","id":"cd54f0b1-11a2-4b1c-b530-ba0d0909f4d8"}},"sourceEventSeqs":[45],"surfaceOp":"append"} {"type":"step/end","seq":47,"time":1785730504721,"data":{"turn":1,"step":4}} {"type":"step/start","seq":48,"time":1785730504730,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":49,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -53,9 +53,9 @@ {"type":"assistant/chunk","seq":51,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}}}} {"type":"assistant/chunk","seq":52,"time":1785498587503,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":53,"time":1785730504731,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"1a4e37b3-72fe-4cbe-bbdb-708088894423"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} +{"type":"assistant/message","seq":54,"time":1785730504731,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"90745bdd-9fc0-4e28-bc1a-187340f1ed40"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} {"type":"tool/call","seq":55,"time":1785730504731,"data":{"turn":1,"step":5,"callId":"pty-kill","name":"terminal_close","arguments":"{\"sessionId\":\"pty-1\"}"}} -{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"a04c5ae2-3b36-40d0-b570-df158f27d08a"}},"sourceEventSeqs":[55],"surfaceOp":"append"} +{"type":"tool/result","seq":56,"time":1785730504738,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"pty-kill"},"content":[{"type":"tool-result","toolCallId":"pty-kill","content":[{"type":"text","text":"closed terminal session pty-1"}],"isError":false}],"role":"user","id":"15f8d343-4410-4fdf-a9ef-c2b71db3cc6c"}},"sourceEventSeqs":[55],"surfaceOp":"append"} {"type":"step/end","seq":57,"time":1785730504738,"data":{"turn":1,"step":5}} {"type":"step/start","seq":58,"time":1785730504746,"data":{"turn":1,"step":6}} {"type":"assistant/chunk","seq":59,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -63,9 +63,9 @@ {"type":"assistant/chunk","seq":61,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}}}} {"type":"assistant/chunk","seq":62,"time":1785498587517,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} {"type":"assistant/chunk","seq":63,"time":1785730504747,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0656b271-2099-44fe-a770-834bb6c97e32"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"} +{"type":"assistant/message","seq":64,"time":1785730504747,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"pty-list","name":"terminal_list","arguments":"{}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"c1050cd4-0f07-43c5-bd50-4edf9341a9ff"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"} {"type":"tool/call","seq":65,"time":1785730504747,"data":{"turn":1,"step":6,"callId":"pty-list","name":"terminal_list","arguments":"{}"}} -{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"8b6397b3-9ee8-431e-975a-200b14670d7e"}},"sourceEventSeqs":[65],"surfaceOp":"append"} +{"type":"tool/result","seq":66,"time":1785730504755,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"pty-list"},"content":[{"type":"tool-result","toolCallId":"pty-list","content":[{"type":"text","text":"(no terminal sessions)"}],"isError":false}],"role":"user","id":"6afd008d-3796-4358-a6bb-4dac091768e1"}},"sourceEventSeqs":[65],"surfaceOp":"append"} {"type":"step/end","seq":67,"time":1785730504755,"data":{"turn":1,"step":6}} {"type":"step/start","seq":68,"time":1785730504763,"data":{"turn":1,"step":7}} {"type":"assistant/chunk","seq":69,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -73,6 +73,6 @@ {"type":"assistant/chunk","seq":71,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"DONE"}}}} {"type":"assistant/chunk","seq":72,"time":1785498587531,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":3}}}} {"type":"assistant/chunk","seq":73,"time":1785730504764,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"74b620d5-2737-4b35-b9c9-2829fe59bea5"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"} +{"type":"assistant/message","seq":74,"time":1785730504764,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"DONE"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"be896da6-8db2-4277-bdf8-063cbc0150aa"},"usage":{"inputTokens":10,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"} {"type":"step/end","seq":75,"time":1785730504764,"data":{"turn":1,"step":7}} {"type":"turn/end","seq":76,"time":1785730504764,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl new file mode 100644 index 0000000000..9d63ab152c --- /dev/null +++ b/examples/headless-agent/tests/snapshots/subagent-settlement/child.expected.jsonl @@ -0,0 +1,20 @@ +{"type":"session","version":0,"id":"{{sessionId}}","createdAt":0,"cwd":"{{cwd}}","parentSession":"{{sessionId}}","origin":"subagent","delegationDepth":1} +{"type":"subagent/descriptor","seq":0,"time":0,"data":{"version":2,"mode":"continuable","provider":"spawn","label":"Return child result","agentProvider":"deepseek-official","agentModel":"deepseek-v4-flash"}} +{"type":"session/end-seed","seq":1,"time":0,"data":{}} +{"type":"agent/inbox/spliced","seq":2,"time":0,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Reply with exactly CHILD_RESULT and nothing else. Do not call report."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"}]}} +{"type":"turn/start","seq":3,"time":0,"data":{"turn":1}} +{"type":"agent/inbox/spliced","seq":4,"time":0,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} +{"type":"step/start","seq":5,"time":0,"data":{"turn":1,"step":1}} +{"type":"user/message","seq":6,"time":0,"data":{"content":[{"type":"text","text":"Reply with exactly CHILD_RESULT and nothing else. Do not call report."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"} +{"type":"user/message","seq":7,"time":0,"data":{"content":[{"type":"text","text":"Current runtime context. This snapshot supersedes earlier runtime-context snapshots.\n\nYou are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}],"source":{"kind":"plugin","plugin":"@deepseek-ai/dsh-system-prompt","form":"snapshot","sections":[{"name":"subagent:delegation","text":"You are a delegated subagent: your permission scope was fixed when you were started and cannot be widened from inside this session — operations that require approval are rejected automatically. When the task needs access beyond that scope, do not retry the denied operation; state the limitation in your reply so the delegating agent can handle it."}]},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"} +{"type":"session/title","seq":8,"time":0,"data":{"title":"Reply with exactly CHILD_RESULT and","messageSeqs":[6],"source":{"kind":"fallback"}}} +{"type":"request/header","seq":9,"time":0,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} +{"type":"request/context","seq":10,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} +{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_RESULT"}}} +{"type":"assistant/chunk","seq":13,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_RESULT"}}}} +{"type":"assistant/chunk","seq":14,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":15,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":16,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[11,12,13,14,15],"surfaceOp":"append"} +{"type":"step/end","seq":17,"time":0,"data":{"turn":1,"step":1}} +{"type":"turn/end","seq":18,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl new file mode 100644 index 0000000000..0f7362448a --- /dev/null +++ b/examples/headless-agent/tests/snapshots/subagent-settlement/child.replay.jsonl @@ -0,0 +1,6 @@ +{"type":"session","version":0,"id":"subagent-settlement-child","createdAt":2,"delegationDepth":1} +{"type":"assistant/chunk","seq":0,"time":1,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":1,"time":2,"data":{"turn":1,"step":1,"chunk":{"type":"text-delta","index":0,"text":"CHILD_RESULT"}}} +{"type":"assistant/chunk","seq":2,"time":3,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CHILD_RESULT"}}}} +{"type":"assistant/chunk","seq":3,"time":4,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}} +{"type":"assistant/chunk","seq":4,"time":5,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json new file mode 100644 index 0000000000..bdb77f505e --- /dev/null +++ b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.override.json @@ -0,0 +1,32 @@ +[ + { + "kind": "chunks", + "chunks": [ + { "type": "block-start", "index": 0, "blockType": "tool-call" }, + { "type": "tool-call-delta", "index": 0, "id": "start-child", "name": "subagent", "argumentsDelta": "{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}" }, + { "type": "block-end", "index": 0, "block": { "type": "tool-call", "id": "start-child", "name": "subagent", "arguments": "{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}" } }, + { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } }, + { "type": "finish", "reason": { "kind": "tool-calls" } } + ] + }, + { + "kind": "chunks", + "chunks": [ + { "type": "block-start", "index": 0, "blockType": "text" }, + { "type": "text-delta", "index": 0, "text": "STARTED" }, + { "type": "block-end", "index": 0, "block": { "type": "text", "text": "STARTED" } }, + { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } }, + { "type": "finish", "reason": { "kind": "stop" } } + ] + }, + { + "kind": "chunks", + "chunks": [ + { "type": "block-start", "index": 0, "blockType": "text" }, + { "type": "text-delta", "index": 0, "text": "PARENT_RECEIVED_CHILD_RESULT" }, + { "type": "block-end", "index": 0, "block": { "type": "text", "text": "PARENT_RECEIVED_CHILD_RESULT" } }, + { "type": "usage", "usage": { "inputTokens": 10, "outputTokens": 5 } }, + { "type": "finish", "reason": { "kind": "stop" } } + ] + } +] diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl new file mode 100644 index 0000000000..b4216aa0ed --- /dev/null +++ b/examples/headless-agent/tests/snapshots/subagent-settlement/parent.replay.jsonl @@ -0,0 +1 @@ +{"type":"session","version":0,"id":"subagent-settlement-parent","createdAt":1,"delegationDepth":0} diff --git a/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl b/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl new file mode 100644 index 0000000000..bc181b78ad --- /dev/null +++ b/examples/headless-agent/tests/snapshots/subagent-settlement/stream-json.expected.jsonl @@ -0,0 +1,38 @@ +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":0,"time":0,"data":{"target":"next-turn","start":0,"inserted":[{"content":[{"type":"text","text":"Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"}]}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"turn/start","seq":1,"time":0,"data":{"turn":1}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":2,"time":0,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"user/message","seq":4,"time":0,"data":{"content":[{"type":"text","text":"Start one continuable background subagent and answer from its completion notice. Do not call list_agents, send_message, task_output, or task_list."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"session/title","seq":5,"time":0,"data":{"title":"Start one continuable background subagen","messageSeqs":[4],"source":{"kind":"fallback"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"request/header","seq":6,"time":0,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"request/context","seq":7,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"start-child","name":"subagent","argumentsDelta":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":13,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":14,"time":0,"data":{"turn":1,"step":1,"callId":"start-child","name":"subagent","arguments":"{\"description\":\"Return child result\",\"prompt\":\"Reply with exactly CHILD_RESULT and nothing else. Do not call report.\",\"run_in_background\":true}"}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":15,"time":0,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"start-child"},"content":[{"type":"tool-result","toolCallId":"start-child","content":[{"type":"text","text":"started subagent {{sessionId}}"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[14],"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":16,"time":0,"data":{"turn":1,"step":1}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":17,"time":0,"data":{"target":"next-step","start":0,"inserted":[{"content":[{"type":"text","text":"Background subagent {{sessionId}} finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent {{sessionId}} finished and will do no further work unless you send it more.","senderSessionId":"{{sessionId}}"},"role":"user","id":"{{sessionId}}"}]}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":18,"time":0,"data":{"turn":1,"step":2}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":20,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"text-delta","index":0,"text":"STARTED"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"STARTED"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":22,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":23,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":24,"time":0,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"text","text":"STARTED"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":25,"time":0,"data":{"turn":1,"step":2}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":26,"time":0,"data":{"target":"next-step","start":0,"removedCount":1,"inserted":[]}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/start","seq":27,"time":0,"data":{"turn":1,"step":3}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"user/message","seq":28,"time":0,"data":{"content":[{"type":"text","text":"Background subagent {{sessionId}} finished and will do no further work unless you send it more."},{"type":"text","text":"Its closing message:"},{"type":"text","text":"CHILD_RESULT"}],"source":{"kind":"subagent-settled","form":"notice","summary":"Background subagent {{sessionId}} finished and will do no further work unless you send it more.","senderSessionId":"{{sessionId}}"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":30,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"text-delta","index":0,"text":"PARENT_RECEIVED_CHILD_RESULT"}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PARENT_RECEIVED_CHILD_RESULT"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":32,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":10,"outputTokens":5}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":33,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":34,"time":0,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"text","text":"PARENT_RECEIVED_CHILD_RESULT"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":10,"outputTokens":5}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"step/end","seq":35,"time":0,"data":{"turn":1,"step":3}}} +{"type":"session_event","sessionId":"{{sessionId}}","event":{"type":"turn/end","seq":36,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}}} +{"type":"result","sessionId":"{{sessionId}}","output":"PARENT_RECEIVED_CHILD_RESULT","usage":{"inputTokens":30,"outputTokens":15}} diff --git a/knip.json b/knip.json index 2f755dbb5f..a721608a76 100644 --- a/knip.json +++ b/knip.json @@ -38,6 +38,7 @@ "headless-agent/tests/fixtures/semantic-checkpoint-agent.ts", "headless-agent/tests/fixtures/subagent-diagnostic-agent.ts", "headless-agent/tests/fixtures/subagent-inheritance-agent.ts", + "headless-agent/tests/fixtures/subagent-settlement-fence.ts", "headless-agent/tests/fixtures/workspace-context-resume-agent.ts", "headless-agent/tests/fixtures/goal-domain/seed-goal.ts", "headless-agent/tests/fixtures/time-context-driver.ts", @@ -50,6 +51,7 @@ "acp-agent/tests/fixtures/parent-sandbox-override.ts", "acp-agent/tests/fixtures/partial-landlock-sandbox.ts", "acp-agent/tests/fixtures/subagent-durability-failure.ts", + "acp-agent/tests/fixtures/subagent-report-fence.ts", "acp-agent/tests/fixtures/subagent-settlement-marker.ts", "acp-agent/tests/fixtures/workspace-context-compaction.ts", "acp-agent/tests/fixtures/subagent/subagent-acp/mock-delegating-llm.ts", diff --git a/packages/core/agent/README.i18n.yaml b/packages/core/agent/README.i18n.yaml index 477c3a5cb6..8022432bc1 100644 --- a/packages/core/agent/README.i18n.yaml +++ b/packages/core/agent/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/core/agent/README.md -README.md: 0fb65b94a3b311aa9f0df09d39dd937cba4cc7b4 -README.zh.md: 44f67483343a98c280317793ece544bd0b984596 +README.md: 61294df5cca1b03f9f158678f2992a6e4fbaaffd +README.zh.md: 5576a07330926cfd8fd5bd42c8a0351a375f035c diff --git a/packages/core/agent/README.md b/packages/core/agent/README.md index 0fb65b94a3..61294df5cc 100644 --- a/packages/core/agent/README.md +++ b/packages/core/agent/README.md @@ -58,6 +58,8 @@ Inbox live notifications are deliberately per-message and minimal: `agent/inbox/ Turn and step boundaries and the model token stream are durable `session/event` facts rather than mirrored `agent/*` notifications. Consumers read `turn/*`, `step/*`, and `assistant/chunk` from the session feed; tool policy and outcome observation belong to the complete pipeline documented by [`dsh-tools`](../tools/README.md). +`foldConsumedWork(events)` reads that feed back for the one question the turn sequence cannot answer alone: what became of the work a log consumed. It returns the latest `turn/end` that accounts for consumed work — a turn that entered a model step, or one that claimed inbox input and then failed, was stopped, or was rejected before reaching one — plus whether accepted work was later cancelled out of the inbox unrun. Both facts come from the log, so a cancellation reads the same whichever owner issued it. A no-step turn that took nothing, or emptied its claim and completed, describes no work and is skipped; a `blocked` end over claimed input is an account, because rejection discarded that input. + ### Agent interface (`types.ts`) The handle every plugin programs against: diff --git a/packages/core/agent/README.zh.md b/packages/core/agent/README.zh.md index 44f6748334..5576a07330 100644 --- a/packages/core/agent/README.zh.md +++ b/packages/core/agent/README.zh.md @@ -58,6 +58,8 @@ inbox 的实时通知刻意采用逐消息的最小载荷：`agent/inbox/inserte 轮次和步骤边界以及模型 token 流是持久 `session/event` 事实，而不是镜像的 `agent/*` 通知。消费方从会话事件流读取 `turn/*`、`step/*` 和 `assistant/chunk`；工具策略与结果观测属于 [`dsh-tools`](../tools/README.md) 记录的完整流水线。 +`foldConsumedWork(events)` 把这条事件流读回来，回答仅凭轮次序列无法回答的那个问题：一份日志消费掉的工作最终怎样了。它返回能够为已消费工作作出交代的最新 `turn/end`——即进入过模型 step 的轮次，或者认领了 inbox 输入、但在进入 step 之前失败、被停下或被拒绝的轮次——并额外给出「已接受的工作此后是否被从 inbox 中取消且从未运行」。两项事实都来自日志，因此无论由哪个所有者发起取消，读出来都一样。没有取走任何输入、或认领批次被改写清空后正常结束的无 step 轮次不描述工作，会被跳过；认领过输入、以 `blocked` 结束的轮次则是一份交代，因为拒绝把这些输入一并丢弃了。 + ### Agent 接口（`types.ts`） 每个插件面向的 handle： diff --git a/packages/core/agent/src/consumed-work.ts b/packages/core/agent/src/consumed-work.ts new file mode 100644 index 0000000000..b86ac836ef --- /dev/null +++ b/packages/core/agent/src/consumed-work.ts @@ -0,0 +1,108 @@ +/** + * How one agent log accounts for the work it consumed. + * + * The turn and step vocabulary alone cannot answer this. A turn that stops + * before its first step leaves a `turn/end` shaped exactly like the balanced + * no-op turns a rejection or an empty claim produces, so reading turns in + * isolation either credits cut-short work as finished or convicts every no-op. + * The missing fact is the inbox's own record: {@link Inbox} logs each mutation + * with `removedCount` and marks a cancellation `outcome: 'canceled'`, which + * separates a turn claiming its input from work being dropped unrun. + * + * @module @deepseek-ai/dsh-agent/consumed-work + */ + +import type { SessionEvent, TurnEndReason } from '@deepseek-ai/dsh-session' + +/** How one agent log accounts for the work it consumed. */ +export interface ConsumedWork { + /** + * The latest closed turn that accounts for consumed work: one that entered a + * model step, or one that claimed inbox input and then failed, was stopped, + * or was rejected. Absent when no turn closed over any work. + */ + readonly end?: SessionEvent<'turn/end'> + /** + * Whether accepted work was cancelled out of the inbox, unrun, after that + * turn. This is the only account of input a cancellation took before any turn + * could open over it — no `turn/end` describes it. + */ + readonly droppedUnrun: boolean +} + +/** + * Whether a turn that consumed input but never reached a step ends in a way + * that accounts for that input. Only a `completed` end does not: it had + * nothing left to run once its claim was rewritten away. A `blocked` end is + * that input's ending too — the pre-step rejection that produced it discarded + * the claimed messages, so the work it took will never run. + * @param reason - the turn's recorded ending. + * @returns whether the ending accounts for the input the turn took. + */ +function accountsForClaim(reason: TurnEndReason): boolean { + switch (reason.kind) { + case 'completed': + return false + case 'blocked': + case 'aborted': + case 'interrupted': + case 'error': + return true + /* v8 ignore next 4 -- unreachable: the one unnamed built-in, `max-tokens`, requires a step, + * so its turn short-circuits as stepped before this call, and `TurnEndReasonMap` is + * merge-extensible, so a backend-added variant cannot be listed; an unnameable ending over + * consumed input must not read as success. */ + default: + return true + } +} + +/** + * Fold one agent log, or an owned suffix of one, into its account of consumed + * work. Single pass, and every input is the log itself: no caller has to sample + * live state before cancelling, so a cancellation issued by anyone — the owner's + * teardown, an ancestor's interrupt, an unloading plugin — reads the same. + * @param events - the log, or an owned suffix, to fold. + * @returns the accounting turn when one closed, and whether work was dropped unrun after it. + */ +export function foldConsumedWork(events: readonly SessionEvent[]): ConsumedWork { + const stepped = new Set() + const claimed = new Set() + let open: number | undefined + let end: SessionEvent<'turn/end'> | undefined + let droppedUnrun = false + for (const event of events) { + switch (event.type) { + case 'turn/start': + open = event.data.turn + break + case 'step/start': + stepped.add(event.data.turn) + break + case 'agent/inbox/spliced': { + const { removedCount, outcome, inserted } = event.data + if (removedCount === undefined) break + // A replacement keeps the work pending under a new identity, so only a + // cancellation that leaves nothing behind drops it. + if (outcome === 'canceled') droppedUnrun ||= inserted.length === 0 + // Claims are the loop's own step-boundary reads, always inside a turn. + else if (open !== undefined) claimed.add(open) + break + } + case 'turn/end': { + const { turn, reason } = event.data + open = undefined + if (stepped.delete(turn) || (claimed.delete(turn) && accountsForClaim(reason))) { + end = event + // Anything dropped before this turn closed is what its own ending + // reports; only a later drop is still unaccounted for. + droppedUnrun = false + } + break + } + default: + break + } + } + return { ...end === undefined ? {} : { end }, droppedUnrun } +} diff --git a/packages/core/agent/src/index.ts b/packages/core/agent/src/index.ts index dffc1f01b3..5211942f11 100644 --- a/packages/core/agent/src/index.ts +++ b/packages/core/agent/src/index.ts @@ -18,6 +18,7 @@ import type { Agent, AgentOptions } from './runtime-types.ts' export * from './runtime-types.ts' export * from './types.ts' export * from './inbox.ts' +export * from './consumed-work.ts' export * from './model-selection.ts' export { agentCarrier, agentEvents, assembleContextFor, emitAgentEvent } from './dispatch.ts' export type { AgentEventDispatch, AgentSubjectEvent } from './dispatch.ts' diff --git a/packages/core/agent/tests/consumed-work.spec.ts b/packages/core/agent/tests/consumed-work.spec.ts new file mode 100644 index 0000000000..3b79d67620 --- /dev/null +++ b/packages/core/agent/tests/consumed-work.spec.ts @@ -0,0 +1,160 @@ +import { describe, expect, it } from 'vitest' +import { createUserMessage } from '@deepseek-ai/dsh-llm' +import { Session, SessionId } from '@deepseek-ai/dsh-session' +import type { TurnEndReason } from '@deepseek-ai/dsh-session' +import { foldConsumedWork } from '@deepseek-ai/dsh-agent' + +/** One pending message, as the inbox records it. */ +function message(text: string) { + return createUserMessage({ content: [{ type: 'text', text }], source: { kind: 'user' } }) +} + +/** Log an accepted message the way `Inbox.append()` does. */ +function accept(session: Session, text: string): void { + session.append('agent/inbox/spliced', { target: 'next-turn', start: 0, inserted: [message(text)] }) +} + +/** Log the step-boundary read of one pending message, as `Inbox.claim()` does. */ +function claim(session: Session): void { + session.append('agent/inbox/spliced', { target: 'next-turn', start: 0, removedCount: 1, inserted: [] }) +} + +/** Log a cancellation of one pending message, as `Inbox.clear()` does. */ +function cancelPending(session: Session): void { + session.append('agent/inbox/spliced', { + target: 'next-turn', start: 0, removedCount: 1, inserted: [], outcome: 'canceled', + }) +} + +/** Run one whole turn that reached a model step. */ +function steppedTurn(session: Session, turn: number, reason: TurnEndReason): void { + session.append('turn/start', { turn }) + claim(session) + session.append('step/start', { turn, step: 1 }) + session.append('step/end', { turn, step: 1 }) + session.append('turn/end', { turn, reason }) +} + +describe('foldConsumedWork', () => { + it('reports nothing for a log that consumed no work', () => { + const session = Session.create(SessionId('empty')) + accept(session, 'queued') + + expect(foldConsumedWork(session.events)).toEqual({ droppedUnrun: false }) + }) + + it('reports the latest turn that entered a model step', () => { + const session = Session.create(SessionId('stepped')) + steppedTurn(session, 1, { kind: 'completed' }) + steppedTurn(session, 2, { kind: 'max-tokens' }) + + expect(foldConsumedWork(session.events).end?.data) + .toEqual({ turn: 2, reason: { kind: 'max-tokens' } }) + }) + + it('reports a turn that claimed its input and then failed before any step', () => { + const session = Session.create(SessionId('failed-claim')) + steppedTurn(session, 1, { kind: 'completed' }) + // The step boundary runs the durability checkpoint and prompt assembly, so a + // turn can take its input and then fail without entering a step. + session.append('turn/start', { turn: 2 }) + claim(session) + session.append('turn/end', { turn: 2, reason: { kind: 'error', error: { message: 'ENOSPC', code: 'UNKNOWN' } } }) + + expect(foldConsumedWork(session.events).end?.data.turn).toBe(2) + }) + + it('reports a turn that claimed its input and was then stopped before any step', () => { + const session = Session.create(SessionId('stopped-claim')) + steppedTurn(session, 1, { kind: 'completed' }) + session.append('turn/start', { turn: 2 }) + claim(session) + session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'user' } } }) + + expect(foldConsumedWork(session.events).end?.data.turn).toBe(2) + }) + + it('ignores a turn stopped, failed, or rejected without taking any input', () => { + const session = Session.create(SessionId('no-claim')) + steppedTurn(session, 1, { kind: 'completed' }) + session.append('turn/start', { turn: 2 }) + session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'parent' } } }) + session.append('turn/start', { turn: 3 }) + session.append('turn/end', { turn: 3, reason: { kind: 'error', error: { message: 'x', code: 'UNKNOWN' } } }) + session.append('turn/start', { turn: 4 }) + session.append('turn/end', { turn: 4, reason: { kind: 'blocked' } }) + + // None of these turns describes work: they opened, found nothing of their own, and closed. + expect(foldConsumedWork(session.events).end?.data.turn).toBe(1) + }) + + it('reports a turn whose claimed input a pre-step rejection discarded', () => { + const session = Session.create(SessionId('rejected-claim')) + steppedTurn(session, 1, { kind: 'completed' }) + session.append('turn/start', { turn: 2 }) + claim(session) + session.append('turn/end', { turn: 2, reason: { kind: 'blocked' } }) + + // Rejection does not retain the claimed messages, so the `blocked` end is + // the only account of input that will never run. + expect(foldConsumedWork(session.events).end?.data.turn).toBe(2) + }) + + it('ignores a claim its own turn emptied', () => { + const session = Session.create(SessionId('emptied-claim')) + steppedTurn(session, 1, { kind: 'completed' }) + session.append('turn/start', { turn: 2 }) + claim(session) + session.append('turn/end', { turn: 2, reason: { kind: 'completed' } }) + + // An emptied claim ran nothing and dropped nothing: a listener rewrote the + // batch away, which is not this log's account of the work. + expect(foldConsumedWork(session.events).end?.data.turn).toBe(1) + }) + + it('credits a claim with no open turn to no turn at all', () => { + const session = Session.create(SessionId('mid-turn-suffix')) + steppedTurn(session, 1, { kind: 'completed' }) + // An owned suffix can begin inside a turn whose start it does not contain, + // so a claim may appear with no turn to attribute it to. + claim(session) + session.append('turn/end', { turn: 2, reason: { kind: 'aborted', reason: { kind: 'user' } } }) + + expect(foldConsumedWork(session.events).end?.data.turn).toBe(1) + }) + + it('reports work cancelled out of the inbox after the last accounting turn', () => { + const session = Session.create(SessionId('dropped')) + steppedTurn(session, 1, { kind: 'completed' }) + accept(session, 'never runs') + cancelPending(session) + + // No turn opened over it, so only the cancellation says the work was cut short. + expect(foldConsumedWork(session.events)).toEqual({ + end: session.events.find(event => event.type === 'turn/end'), + droppedUnrun: true, + }) + }) + + it('keeps a replacement pending rather than counting it as dropped', () => { + const session = Session.create(SessionId('replaced')) + steppedTurn(session, 1, { kind: 'completed' }) + session.append('agent/inbox/spliced', { + target: 'next-turn', start: 0, removedCount: 1, inserted: [message('rewritten')], outcome: 'canceled', + }) + + expect(foldConsumedWork(session.events).droppedUnrun).toBe(false) + }) + + it('lets a later accounting turn absorb an earlier drop', () => { + const session = Session.create(SessionId('absorbed')) + steppedTurn(session, 1, { kind: 'completed' }) + cancelPending(session) + steppedTurn(session, 2, { kind: 'completed' }) + + expect(foldConsumedWork(session.events)).toEqual({ + end: session.events.findLast(event => event.type === 'turn/end'), + droppedUnrun: false, + }) + }) +}) diff --git a/packages/core/session/src/index.ts b/packages/core/session/src/index.ts index 5251ca9408..477d08ae19 100644 --- a/packages/core/session/src/index.ts +++ b/packages/core/session/src/index.ts @@ -34,27 +34,6 @@ export { deriveEventMessage, foldSurface, isAppendSurfaceEvent, isReplacementSur export { canonicalHeader, foldRequestHeader, headerEquals } from './request-header.ts' export { KNOWN_SESSION_EVENT_TYPES } from './known-event-types.ts' -/** - * Find the latest closed turn that entered at least one model step, ignoring - * balanced no-step turns produced by rejection, empty input, or cancellation. - * @param events - session events, or an owned suffix, to inspect. - * @returns the latest matching turn end, or `undefined`. - */ -export function findLastMessageTurnEnd( - events: readonly SessionEvent[], -): SessionEvent<'turn/end'> | undefined { - const steppedTurns = new Set() - let latest: SessionEvent<'turn/end'> | undefined - for (const event of events) { - if (event.type === 'step/start') { - steppedTurns.add(event.data.turn) - continue - } - if (event.type === 'turn/end' && steppedTurns.delete(event.data.turn)) latest = event - } - return latest -} - declare module '@deepseek-ai/cordis' { interface Context { sessions: SessionStore diff --git a/packages/core/session/tests/session.spec.ts b/packages/core/session/tests/session.spec.ts index 7fe7ee01c7..d29624c030 100644 --- a/packages/core/session/tests/session.spec.ts +++ b/packages/core/session/tests/session.spec.ts @@ -7,28 +7,11 @@ import SessionStore, { Session, SessionEvent, SessionId, - findLastMessageTurnEnd, snapshotSessionEvent, } from '@deepseek-ai/dsh-session' import type { CreateSessionOptions, SessionEventType, SessionHeader, SessionSurface, TodoItem } from '@deepseek-ai/dsh-session' describe('Session', () => { - it('finds the latest closed turn that entered a model step', () => { - const session = Session.create(SessionId('last-message-turn')) - session.append('turn/start', { turn: 1 }) - session.append('turn/end', { turn: 1, reason: { kind: 'blocked' } }) - - expect(findLastMessageTurnEnd(session.events)).toBeUndefined() - - session.append('turn/start', { turn: 2 }) - session.append('step/start', { turn: 2, step: 1 }) - session.append('step/end', { turn: 2, step: 1 }) - session.append('turn/end', { turn: 2, reason: { kind: 'max-tokens' } }) - - expect(findLastMessageTurnEnd(session.events)?.data) - .toEqual({ turn: 2, reason: { kind: 'max-tokens' } }) - }) - it('exposes one stable readonly surface view', () => { const session = Session.create(SessionId('surface-view')) const surface = session.surface diff --git a/packages/subagent/subagent-inprocess/src/index.ts b/packages/subagent/subagent-inprocess/src/index.ts index b0ab7425a5..0e7942fa55 100644 --- a/packages/subagent/subagent-inprocess/src/index.ts +++ b/packages/subagent/subagent-inprocess/src/index.ts @@ -13,8 +13,9 @@ import { randomUUID } from 'node:crypto' import type { Context } from '@deepseek-ai/cordis' +import { foldConsumedWork } from '@deepseek-ai/dsh-agent' import type { Agent, AgentHandle } from '@deepseek-ai/dsh-agent' -import { findLastMessageTurnEnd, SessionId, type SessionEvent, type TurnEndReason } from '@deepseek-ai/dsh-session' +import { SessionId, type SessionEvent, type TurnEndReason } from '@deepseek-ai/dsh-session' import { createUserMessage, type ContentBlock } from '@deepseek-ai/dsh-llm' import { appendDelegatedPolicyOverrides, @@ -52,6 +53,10 @@ function toStopReason(reason: TurnEndReason | undefined): SubagentStopReason { return 'max-tokens' case 'aborted': return 'aborted' + // A pre-step rejection discarded the claimed prompt: the task was + // declined, and the caller must not read the run as done. + case 'blocked': + return 'refusal' case 'error': case 'interrupted': default: @@ -207,7 +212,11 @@ function readResult( structured?: { captured?: { value: unknown } | undefined }, ): SubagentResult { const own = child.session.events.slice(boundary) - const lastEnd = findLastMessageTurnEnd(own) + // `droppedUnrun` is deliberately unread: a one-shot prompt is claimed by its + // awaited first turn almost immediately, and the owner's own teardown is the + // `cancelled` flag below. A cancellation with no accounting turn resolves + // `error` through `toStopReason(undefined)`, which never overstates success. + const lastEnd = foldConsumedWork(own).end // The seam's canonical selection rule; a partial answer survives cancel and truncation. const output: ContentBlock[] = finalAssistantOutput(own) ?? [] const recorded = toStopReason(lastEnd?.data.reason) diff --git a/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts b/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts index 4b9041ebf9..bc25a79c23 100644 --- a/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts +++ b/packages/subagent/subagent-inprocess/tests/subagent-inprocess.spec.ts @@ -82,6 +82,20 @@ describe('startInProcessRun', () => { await run.dispose() }) + it('reports a prompt a pre-step rejection discarded as refusal, not completion', async () => { + const { ctx, parent } = await setup([]) + // A UserPromptSubmit deny or a policy plugin: the child claims its prompt, + // the rejection discards it, and the turn closes `blocked` with no step. + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject === parent) return next() + return { kind: 'reject' as const } + }) + + const run = await startInProcessRun(request(parent), {}) + await expect(run.result).resolves.toMatchObject({ stopReason: 'refusal' }) + await run.dispose() + }) + it('does not add a final durability checkpoint to a foreground run', async () => { const { ctx, parent } = await setup([textResponse('driver answer')]) let flushes = 0 diff --git a/packages/subagent/subagent/README.i18n.yaml b/packages/subagent/subagent/README.i18n.yaml index c70bcdf58b..d6284bd1d8 100644 --- a/packages/subagent/subagent/README.i18n.yaml +++ b/packages/subagent/subagent/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/subagent/subagent/README.md -README.md: 28f649ef54bbf88feda24a9ce197c2c366f8349b -README.zh.md: 595c5e5e7fffc367f2e3fd8142b779dc22fb3b79 +README.md: 6373f09e44a5bb87679d2afe67321f49c7174fef +README.zh.md: a657e6371677caa97f0884d3f741a95b3cc6c437 diff --git a/packages/subagent/subagent/README.md b/packages/subagent/subagent/README.md index 28f649ef54..6373f09e44 100644 --- a/packages/subagent/subagent/README.md +++ b/packages/subagent/subagent/README.md @@ -76,6 +76,14 @@ The manager derives three internal residency conditions from Agent quiescence an The manager reserves the child identity, resolves the durable descriptor, calls `ctx.agents.create()` (or `ctx.agents.resume()` for cold resume) through a private activation-owner scope, installs the returned `AgentHandle` in the Activation, establishes any continuable-parent ownership, and then submits the prompt. Cold resume never dispatches through a provider because the persisted Session already holds the initial prefix and the folded descriptor is the whole reconstruction input. +### Settlement delivery + +When a resident Activation settles, the manager tells the child's durable direct parent, in the parent's own turn stream, that the child produced everything it is going to. Delivery is unconditional for every child whose id a caller actually received: it does not consider whether the child called `report`, because the endings that most need an account — a token ceiling, a model failure, cancellation, teardown — are exactly the ones where the child never got to choose. A materialization rolled back before its first accepted message stays silent, since that caller was told the child was not established. The message carries the epoch's stop reason, its final assistant content when it produced any, and durable provenance `{ kind: 'subagent-settled', form: 'notice', senderSessionId: }` — a different source kind from a child-authored `subagent-report`, so a transcript never credits the child with words the runtime wrote. + +Two ordering rules make the delivery reliable rather than lucky, and both are why this belongs to the manager instead of an external `subagent/end` listener. First, the send happens **before** the child's ownership release, while the parent still counts the child and is therefore structurally unable to be judged settled. Second, a parent that is itself a resident Activation receives the message through the same waking-admission accounting as a report, so the window between the synchronous send and the microtask that admits it is not mistaken for quiescence — `Agent.status` folds context maintenance into `idle`, and a waking send behind maintenance only arms a deferred wake. Without either rule the parent can be disposed with the notice still in an inbox that `cancel()` clears, which loses it silently. + +An idle parent receives the notice as one ordinary later turn. A busy parent is steered into its nearest step boundary instead, so several children settling together cost one step rather than one turn each; steering rather than injecting also means a driver that retires between the status read and the send still claims the message. A parent whose own lineage is already draining receives the notice by injection, with no wake at all: `Agent.followup()` on a quiescent parent starts a turn and `cancel()` does not arm against a later one, so waking during teardown would spend a model request on an Agent its host is about to dispose — once per tree layer, since each layer's notice then wakes the layer above it. The injected message reaches a parent that is still reading its inbox, and the log records the account either way, but it does not outlive that parent's own disposal: `AgentHandle.dispose()` is a `keepInbox: false` cancel, which durably cancels an unclaimed notice. A resumed parent therefore has no pending notice to read: `list_agents` tells it which children exist and whether each is live or stored, while the outcome itself stays in the child's own Session, which a `send_message` reaches by resuming that child. A parent that has left the registry is not an error: the notice is dropped and the child's own Session remains the durable record. Delivery never blocks or fails teardown — a rejected send is logged, because retaining a child to retry a notice would pin its whole ancestry in `waiting` forever. + A continuation-managed parent Activation records each child Session id in an `ownedChildren` set before the child can run and disposes only after every owned child Activation completes `AgentHandle` disposal (child-first). Teardown propagates Agent cancellation top-down before awaiting slow descendants, while handle release remains child-first. Top-level and other non-continuation Agents have no Activation and stay outside this waiting graph. Final settlement awaits a best-effort `ctx.sessions.flush(child.session)` before handle disposal. A listener rejection is logged without failing the Activation because listener participation does not identify a persistence backend; the persisted state may therefore be missing or stale on resume. ## Lifecycle events @@ -100,11 +108,25 @@ Continuable Activations await a best-effort final session flush without treating ## Model Experience +### Settlement notice + +#### What the model sees + +One user-role parent message opening with the outcome — `Background subagent finished and will do no further work unless you send it more.`, or the matching line for a child that was stopped, ran out of room, declined, or failed — followed by `Its closing message:` and the child's final assistant content, or `It left no closing message.` when it produced none. This is the service's only direct parent-side contribution; delegation schemas, parent continuation and discovery, and the child-scoped `report` belong to `dsh-tool-subagent`, `dsh-tool-subagent-control`, and `dsh-tool-subagent-report`. + +#### Token effect + +One notice per settled Activation in the parent's request, sized by the child's final message. A child that both reports and settles costs the parent both. + +#### KV Cache effect + +Append-only in the parent: the notice follows its reusable request prefix. Reaching an idle parent starts one independent model request; reaching a busy one does not. + ### Child delegation-scope statement #### What the model sees -Every in-process child's runtime-context snapshot carries the `subagent:delegation` statement below, after the sandbox-policy and approval-policy sentences; parent-side rendering stays with `dsh-tool-subagent` (delegation schemas), `dsh-tool-subagent-control` (continuation and discovery), and `dsh-tool-subagent-report` (the child-scoped `report`). +Every in-process child's runtime-context snapshot carries the `subagent:delegation` statement below, after the sandbox-policy and approval-policy sentences. ##### The delegation-scope statement diff --git a/packages/subagent/subagent/README.zh.md b/packages/subagent/subagent/README.zh.md index 595c5e5e7f..a657e63716 100644 --- a/packages/subagent/subagent/README.zh.md +++ b/packages/subagent/subagent/README.zh.md @@ -76,6 +76,14 @@ subagent seam 允许一个 agent（智能体）通过具名提供方把工作委 管理器预留子 agent 身份、解析持久化描述符，通过私有的 activation-owner 作用域调用 `ctx.agents.create()`（冷恢复时为 `ctx.agents.resume()`），把返回的 `AgentHandle` 安装到 Activation 中，建立任何可继续父级所有权，然后提交提示词。冷恢复绝不通过提供方分发，因为持久化 Session 已持有初始前缀，折叠后的描述符即是全部重建输入。 +### 结算投递 + +当一个驻留 Activation 结算时，管理器会在父级自身的轮次流中告知该子级持久化的直接父级：这个子级已经产出它将产出的全部内容。对每个调用方真正拿到过 id 的子级，这条投递都是无条件的：它不考虑该子级是否调用过 `report`，因为最需要这条投递的结束方式——token 上限、模型失败、取消、拆卸——恰恰是子级根本没有机会选择的那些。在第一条消息被接受之前就回滚的物化保持静默，因为那位调用方已被告知该子级未建立。消息会携带该 epoch 的终止原因、它产出过的最终 assistant 内容，以及持久化来源 `{ kind: 'subagent-settled', form: 'notice', senderSessionId: }`——与子级自撰的 `subagent-report` 是不同的来源 kind，因此 transcript（文本记录）绝不会把运行时写下的话算到子级头上。 + +有两条顺序规则让这条投递可靠而非侥幸，它们也正是这件事属于管理器而非外部 `subagent/end` listener 的原因。第一，发送发生在子级所有权释放**之前**，此时父级仍然计入该子级，因此在结构上不可能被判定为已结算。第二，本身就是驻留 Activation 的父级会通过与 report 相同的唤醒准入记账接收该消息，因此同步发送与承认它的那个 microtask 之间的窗口不会被误判为静止——`Agent.status` 会把上下文维护折叠成 `idle`，而维护期间的唤醒发送只会预置一次延后唤醒。缺少其中任一条规则，父级都可能在通知仍留在 inbox 时被 dispose，而 `cancel()` 会清空该 inbox，于是通知被静默丢失。 + +空闲父级会以一个普通的后续轮次收到该通知。繁忙父级则被 steer 到其最近的 step 边界，因此同时结算的多个子级只消耗一个 step，而不是各自一个轮次；采用 steer 而非 inject 还意味着：即便驱动在状态读取与发送之间退出，该消息仍会被认领。若父级自身所在的谱系已在 draining，则该通知改为 inject 送达，完全不唤醒：对静息父级调用 `Agent.followup()` 会开启一个轮次，而 `cancel()` 不会对之后的轮次设防，因此在拆卸期间唤醒，会在宿主即将 dispose 的 Agent 上白花一次模型请求——而且每层树各一次，因为每层自己的通知又会唤醒它上面那层。被 inject 的消息会送达仍在读取自身 inbox 的父级，而无论如何日志都会记录这份记账；但它不会比该父级自身的 dispose 活得更久：`AgentHandle.dispose()` 是一次 `keepInbox: false` 的 cancel，会持久地取消尚未被认领的通知。因此 resume 后的父级没有待处理通知可读：`list_agents` 只告诉它有哪些子级、各自是在线还是仅存于存储；结局本身留在子级自己的 Session 里，一次 `send_message` 会通过 resume 该子级把它取回。已离开注册表的父级不算错误：通知被丢弃，子级自身的 Session 仍是持久记录。投递绝不会阻塞或使拆卸失败——发送被拒只会记录日志，因为为了重试一条通知而保留子级，会把它的整条祖先链永久钉在 `waiting` 上。 + 受继续执行管理的父级 Activation 会在子 agent 能够运行之前，把每个子 agent 的 Session id 记录到 `ownedChildren` 集合中，并且只有在每个所拥有的子 agent Activation 完成 `AgentHandle` dispose 之后才会 dispose（子先于父）。拆卸会先自顶向下传播 Agent 取消，再等待缓慢的后代，而 handle 释放仍保持 child-first。顶层及其他非继续执行的 Agent 没有 Activation，处于该等待图之外。最终结算会在 dispose handle 前等待 best-effort 的 `ctx.sessions.flush(child.session)`。listener rejection 会被记录，但不会使 Activation 失败，因为 listener 是否参与无法标识持久化后端；因此，恢复时持久化状态可能缺失或陈旧。 ## 生命周期事件 @@ -100,11 +108,25 @@ subagent seam 允许一个 agent（智能体）通过具名提供方把工作委 ## 模型体验 +### 结算通知 + +#### 模型看到的内容 + +一条用户角色的父级消息，开头是结果本身——`Background subagent finished and will do no further work unless you send it more.`，或子级被停止、耗尽额度、拒绝任务或失败时的对应句子——随后是 `Its closing message:` 与子级的最终 assistant 内容；若子级没有产出内容，则是 `It left no closing message.`。这是本服务面向父级的唯一直接贡献；委派 schema、父级延续与发现以及子级作用域的 `report` 分别归 `dsh-tool-subagent`、`dsh-tool-subagent-control` 和 `dsh-tool-subagent-report` 所有。 + +#### Token 影响 + +父级请求中，每个已结算的 Activation 一条通知，长度取决于子级的最终消息。既上报又结算的子级会让父级同时支付两份。 + +#### KV Cache 影响 + +在父级中仅追加：通知位于其可复用请求前缀之后。到达空闲父级会启动一次独立的模型请求，到达繁忙父级则不会。 + ### 子级委派范围声明 #### 模型看到的内容 -每个进程内子 agent 的运行时上下文快照都携带下方的 `subagent:delegation` 声明，位于沙箱策略与审批策略语句之后；父级侧的渲染仍归 `dsh-tool-subagent`（委派 schema）、`dsh-tool-subagent-control`（延续与发现）和 `dsh-tool-subagent-report`（子级作用域的 `report`）所有。 +每个进程内子 agent 的运行时上下文快照都携带下方的 `subagent:delegation` 声明，位于沙箱策略与审批策略语句之后。 ##### 委派范围声明 diff --git a/packages/subagent/subagent/src/continuation.ts b/packages/subagent/subagent/src/continuation.ts index 871ec6efdd..073708f4bc 100644 --- a/packages/subagent/subagent/src/continuation.ts +++ b/packages/subagent/subagent/src/continuation.ts @@ -1,7 +1,8 @@ /** * Internal continuable-subagent manager: stable child ids, descriptor * persistence, activation admission, the live ownership graph, cold resume, - * and child-first disposal behind `ctx.subagents`. + * child-first disposal, and settlement delivery to the parent, behind + * `ctx.subagents`. * * A continuable child has one durable Session and at most one process-local * {@link Activation} — one residency epoch for a reconstructed child Agent. An @@ -11,6 +12,12 @@ * residency while the Agent loop owns all turn ordering and execution. No * continuable path creates a Task or an intermediate result-bearing wrapper. * + * Because residency is this manager's alone to end, telling the parent that a + * child settled is its job too. An external `subagent/end` listener cannot do + * it correctly: that payload names no parent, the child handle is already + * disposed by then, and the release that wakes the parent's own settlement + * watcher has already run. See {@link SubagentContinuationManager.notifySettlement}. + * * @module @deepseek-ai/dsh-subagent */ @@ -23,7 +30,7 @@ import type { AgentSetupCommit, CreateAgentOptions, } from '@deepseek-ai/dsh-agent' -import { createUserMessage, errorChain } from '@deepseek-ai/dsh-llm' +import { boundContextSummary, createUserMessage, errorChain } from '@deepseek-ai/dsh-llm' import type { ContentBlock, MessageId, MessageSource } from '@deepseek-ai/dsh-llm' import { SessionId } from '@deepseek-ai/dsh-session' import type { SessionEvent } from '@deepseek-ai/dsh-session' @@ -42,8 +49,8 @@ import { import type { DelegatedPolicyOverrides } from './child-agent.ts' import { assertSubagentMaxDepth } from './depth.ts' import { seedDescriptorTurn } from './descriptor-seed.ts' -import type { ContinuableCreateRequest, ContinuableCreateSpec, SubagentStartRequest } from './types.ts' -import type { ActivationObserver } from './lifecycle.ts' +import type { ContinuableCreateRequest, ContinuableCreateSpec, SubagentResult, SubagentStartRequest } from './types.ts' +import type { ActivationObserver, ActivationTerminal } from './lifecycle.ts' import { SubagentError } from './error.ts' import type SubagentActivationSetupRegistry from './activation-setup-registry.ts' @@ -65,10 +72,28 @@ export interface SubagentReportMessageSource { readonly senderSessionId: SessionId } +/** + * Durable attribution for the runtime's own account of a continuable child + * settling. Deliberately a different kind from + * {@link SubagentReportMessageSource}: a report is content the child chose, + * while this message is the manager stating what became of the child, and a + * transcript that merged them would credit the child with words it never wrote. + */ +export interface SubagentSettledMessageSource { + readonly kind: 'subagent-settled' + /** A runtime account shown without expanding the row (`notice` context form). */ + readonly form: 'notice' + /** One-line account of how the child ended. */ + readonly summary: string + /** Session id of the child that settled. */ + readonly senderSessionId: SessionId +} + declare module '@deepseek-ai/dsh-llm' { interface MessageSourceMap { coordinator: CoordinatorMessageSource 'subagent-report': SubagentReportMessageSource + 'subagent-settled': SubagentSettledMessageSource } } @@ -166,6 +191,13 @@ interface ContinuationHost { interface Activation { /** The durable child this Activation is an epoch of. */ readonly childId: SessionId + /** + * The durable direct parent, stored because settlement delivery must resolve + * that parent after the child handle is gone. {@link ancestry} cannot answer + * it: a `WeakSet` is not enumerable, and the child's own header is only + * reachable through a handle disposal has already released. + */ + readonly parentSession: SessionId /** The provider name recorded in the durable descriptor. */ readonly provider: string /** The retained live Agent handle, disposed exactly once at settlement. */ @@ -197,6 +229,12 @@ interface Activation { * microtask that admits it, so settlement must not treat that gap as quiet. */ readonly accepted: Set + /** + * Whether any delivery to this child was ever accepted. A materialization + * rolled back before its first acceptance is a child the caller was told does + * not exist, so its teardown owes the parent no settlement account. + */ + announced: boolean /** Renewed whenever a settlement watcher must re-observe quiescence. */ poke: PromiseWithResolvers } @@ -243,6 +281,36 @@ function disposalOf(activation: Activation): Promise | undefined { return activation.disposal } +/** + * One line telling a parent that a background child is finished and why, in + * the parent's own task vocabulary. + * @param childId - the durable child the parent knows by id. + * @param stopReason - how the child's last ordinary turn ended. + * @returns the model-facing opening line of the settlement notice. + */ +function settlementSummary(childId: SessionId, stopReason: SubagentResult['stopReason']): string { + const subject = `Background subagent ${childId}` + switch (stopReason) { + case 'completed': + return `${subject} finished and will do no further work unless you send it more.` + case 'aborted': + return `${subject} was stopped before it finished.` + case 'max-tokens': + return `${subject} ran out of room before it finished.` + // A pre-step rejection — a hook deny, a policy plugin — discarded input + // the child had claimed, so the parent must not treat the task as done. + case 'refusal': + return `${subject} declined the task.` + case 'error': + return `${subject} failed before it finished.` + /* v8 ignore next 4 -- `SubagentResult['stopReason']` is merge-extensible, so this arm + * needs a backend that adds a variant; an unnameable ending is reported as unfinished + * rather than silently as success. */ + default: + return `${subject} ended abnormally (${String(stopReason)}) before it finished.` + } +} + /** Whether one settlement attempt opened the disposal transaction. */ type SettlementAttempt = | { readonly settling: false } @@ -576,19 +644,36 @@ export class SubagentContinuationManager { senderSessionId: activation.childId, }, }) - const parentActivation = this.activations.get(parent.id) - if (delivery === 'wakeup' - && parentActivation !== undefined - && parentActivation.handle.agent === parent) { - this.admitWaking(parentActivation, message.id, () => { - this.sendReport(parent, message, delivery) - }) + if (delivery === 'wakeup') { + this.sendWaking(parent, message, () => { this.sendReport(parent, message, delivery) }) } else { this.sendReport(parent, message, delivery) } return message.id } + /** + * Perform one waking send to a parent, accounted against that parent's own + * Activation when it has one. Registering the id before the send is what + * keeps a continuation-managed parent from being judged quiescent in the + * window between `followup()` and the microtask that admits it. + * @param parent - the exact live parent receiving the waking message. + * @param message - the message whose id is accounted. + * @param send - the synchronous waking send to perform. + */ + private sendWaking( + parent: Agent, + message: ReturnType, + send: () => void, + ): void { + const parentActivation = this.activations.get(parent.id) + if (parentActivation !== undefined && parentActivation.handle.agent === parent) { + this.admitWaking(parentActivation, message.id, send) + } else { + send() + } + } + /** Send one report while translating only the parent's own rejection. */ private sendReport( parent: Agent, @@ -745,23 +830,32 @@ export class SubagentContinuationManager { return lineage } - /** Reject new admission once the manager or this exact parent tree began draining. */ - private assertAdmitting(agent: Agent): void { - if (this.draining) { - throw new SubagentError( - 'continuable subagents are draining; the operation was not admitted', - 'DRAINING', - ) - } + /** + * The teardown that closed continuable admission for this agent's lineage. + * `'manager'` is the whole manager draining; an Agent is the exact scoped root + * whose forest is closing. + * @param agent - the agent whose lineage is tested. + * @returns the closing teardown, or `undefined` while admission is open. + */ + private closingTeardownFor(agent: Agent): Agent | 'manager' | undefined { + if (this.draining) return 'manager' const lineage = this.liveLineage(agent) for (const [root, members] of this.closingScopes) { - if (members.has(agent) || lineage.includes(root)) { - throw new SubagentError( - `continuable subagents below parent "${root.id}" are draining; the operation was not admitted`, - 'DRAINING', - ) - } + if (members.has(agent) || lineage.includes(root)) return root } + return undefined + } + + /** Reject new admission once the manager or this exact parent tree began draining. */ + private assertAdmitting(agent: Agent): void { + const closing = this.closingTeardownFor(agent) + if (closing === undefined) return + throw new SubagentError( + closing === 'manager' + ? 'continuable subagents are draining; the operation was not admitted' + : `continuable subagents below parent "${closing.id}" are draining; the operation was not admitted`, + 'DRAINING', + ) } /** @@ -930,6 +1024,10 @@ export class SubagentContinuationManager { const activation: Activation = { childId, + // The durable lineage, not merely the caller: creation stamps this same + // agent into the child's header, and cold resume authorized it against + // the persisted header before materializing. + parentSession: parent.id, provider, handle, ancestry: new WeakSet([handle.agent, ...parentLineage]), @@ -937,6 +1035,7 @@ export class SubagentContinuationManager { observer, disposal: undefined, accepted: new Set(), + announced: false, poke: Promise.withResolvers(), } // After transfer, any failure must dispose the created handle, remove the @@ -1038,9 +1137,13 @@ export class SubagentContinuationManager { // establish it before the message can enter the child's inbox. this.acquireOwnership(parent, activation.childId) const message = createUserMessage({ content, source }) - return this.admitWaking(activation, message.id, () => { + const accepted = this.admitWaking(activation, message.id, () => { activation.handle.agent.followup(message) }) + // Past this point the caller has an id for this child, so its eventual + // settlement is something the parent is owed an account of. + activation.announced = true + return accepted } /** @@ -1261,6 +1364,12 @@ export class SubagentContinuationManager { // makes a racing delivery wait for release rather than cold-resume into the // still-registered agent. this.activations.delete(childId) + // BEFORE releasing ownership, while the parent still counts this child and + // therefore cannot be judged settled. Delivering after the release would + // race a parent watcher that resumes one microtask later, finds itself + // childless and quiet, and disposes an Agent whose `cancel()` clears the + // inbox this notice is sitting in. + this.notifySettlement(activation, activation.observer.terminal(failure)) // Release ownership even on failure: a retained failed child would pin its // ancestors in `waiting` forever. this.releaseOwnership(childId) @@ -1270,6 +1379,75 @@ export class SubagentContinuationManager { if (failure !== undefined) throw failure } + /** + * Tell the durable direct parent that this child produced everything it is + * going to. Unconditional for every child the caller received an id for: it + * does not consider whether the child reported, because the cases that most + * need it — a token ceiling, a model failure, cancellation, teardown — are + * exactly the ones where the child never got to choose. A materialization + * rolled back before its first acceptance stays silent, since the caller was + * told that child was not established. A parent that is no longer live is not + * an error; the child's own Session remains the durable record either way. + * A parent whose own lineage is already closing receives the notice without a + * wake, because teardown is not a reason to start a turn. + * + * Never blocks disposal. A delivery failure is logged and dropped, because + * retaining a child to retry a notice would pin its whole ancestry in + * `waiting` forever. + * @param activation - the settling Activation, still owned by its parent. + * @param terminal - how this epoch ended, as the terminal edge will report it. + */ + private notifySettlement(activation: Activation, terminal: ActivationTerminal): void { + if (!activation.announced) return + try { + const parent = this.ctx.agents.get(activation.parentSession) + if (parent === undefined) return + const summary = settlementSummary(activation.childId, terminal.stopReason) + const message = createUserMessage({ + content: [ + { type: 'text' as const, text: summary }, + ...terminal.output === undefined + ? [{ type: 'text' as const, text: 'It left no closing message.' }] + : [{ type: 'text' as const, text: 'Its closing message:' }, ...terminal.output], + ], + source: { + kind: 'subagent-settled' as const, + form: 'notice' as const, + summary: boundContextSummary(summary), + senderSessionId: activation.childId, + }, + }) + // A parent whose own teardown already began must not be woken. Waking is + // not a queue operation: `followup()` on a quiescent Agent starts a turn, + // and `cancel()` does not arm against a later one, so a notice arriving + // during teardown would spend a model request on an Agent its host is + // about to dispose — once per tree layer, since each layer's own notice + // then wakes the layer above it. Injecting delivers to a parent still + // reading its inbox and records the account in the log either way; it + // does NOT survive that parent's own disposal, whose `keepInbox: false` + // cancel durably clears whatever it never claimed. + if (this.closingTeardownFor(parent) !== undefined) { + parent.inject(message) + return + } + // An idle parent has nothing else to look at, so it gets one ordinary + // turn. A busy parent is steered instead of woken: `Inbox.claim()` takes + // the whole next-step batch at one boundary, so several children settling + // together cost one step rather than one turn each. Steering rather than + // injecting closes the window where a driver retires between this status + // read and the send, which would strand the notice unclaimed. + this.sendWaking(parent, message, () => { + if (parent.status === 'idle') parent.followup(message) + else parent.steer(message) + }) + } catch (error: unknown) { + this.ctx.logger.warn( + `subagent "${activation.childId}" settlement notice was not delivered to its parent: ` + + errorChain(error), + ) + } + } + /** * Request a best-effort final session flush after the child is quiescent. * Listener failure is logged because flush participation cannot identify a diff --git a/packages/subagent/subagent/src/index.ts b/packages/subagent/subagent/src/index.ts index 36acb8de28..c31d197d25 100644 --- a/packages/subagent/subagent/src/index.ts +++ b/packages/subagent/subagent/src/index.ts @@ -119,6 +119,7 @@ export type { SubagentReportDelivery, SubagentReportMessageSource, SubagentReportOptions, + SubagentSettledMessageSource, } from './continuation.ts' export type { ContinuableSetupContribution } from './activation-setup-registry.ts' export type { SubagentDescendantListEntry, SubagentListEntry } from './list-children.ts' diff --git a/packages/subagent/subagent/src/lifecycle.ts b/packages/subagent/subagent/src/lifecycle.ts index 3482a00c0b..2e1d513d5a 100644 --- a/packages/subagent/subagent/src/lifecycle.ts +++ b/packages/subagent/subagent/src/lifecycle.ts @@ -18,12 +18,23 @@ import { randomUUID } from 'node:crypto' import type { Context } from '@deepseek-ai/cordis' import type { Agent } from '@deepseek-ai/dsh-agent' import type { ContentBlock } from '@deepseek-ai/dsh-llm' -import { findLastMessageTurnEnd } from '@deepseek-ai/dsh-session' +import { foldConsumedWork } from '@deepseek-ai/dsh-agent' import type { SessionEvent, SessionId } from '@deepseek-ai/dsh-session' import { finalAssistantOutput } from './assistant-output.ts' import { SubagentRunId } from './types.ts' import type { SubagentResult, SubagentRun, SubagentRunEndInfo, SubagentRunInfo } from './types.ts' +/** + * How one Activation's residency epoch ended, as both the terminal lifecycle + * edge and the manager's own parent delivery report it. + */ +export interface ActivationTerminal { + /** Why this epoch's last ordinary turn ended, or `error` when teardown failed. */ + readonly stopReason: SubagentResult['stopReason'] + /** The epoch's final assistant content, absent when it produced none or failed. */ + readonly output?: ContentBlock[] +} + /** * Lifecycle observer for one Activation's residency epoch, so continuable * children emit the same start/end pair as one-shot runs. Package-private: the @@ -43,6 +54,15 @@ export interface ActivationObserver { * @param child - the quiescent child agent about to be released. */ capture(child: Agent): void + /** + * Resolve the terminal facts {@link settle} will publish, without publishing + * them. The manager's parent delivery must run before the ownership release + * that lets the parent settle, which is earlier than the terminal edge; both + * therefore read one computation instead of restating the failure rule. + * @param failure - the teardown or durability failure, or `undefined` on success. + * @returns this epoch's stop reason and final assistant content. + */ + terminal(failure: unknown): ActivationTerminal /** * Publish the terminal edge exactly once, pairing this epoch's {@link start}, * after the disposal outcome is known. Called only for a resident epoch: a @@ -165,9 +185,12 @@ export function createActivationObserver( let boundary = 0 // Assigned by `capture()`, which the disposal path always runs before // `settle()`; a resident epoch therefore always has its facts by then. - let captured: { stopReason: SubagentResult['stopReason']; output?: ContentBlock[] } = { - stopReason: 'completed', - } + let captured: ActivationTerminal = { stopReason: 'completed' } + // Teardown failure overrides the epoch's own outcome and withholds its + // output: an answer this harness could not durably release is not a result. + const terminal = (failure: unknown): ActivationTerminal => failure === undefined + ? captured + : { stopReason: 'error' } return { start: (child: Agent): void => { boundary = child.session.events.length @@ -181,11 +204,12 @@ export function createActivationObserver( ...output === undefined ? {} : { output }, } }, + terminal, settle: (failure: unknown): void => { - const output = failure === undefined ? captured.output : undefined + const { stopReason, output } = terminal(failure) emit('subagent/end', { ...identity, - stopReason: failure === undefined ? captured.stopReason : 'error', + stopReason, ...output === undefined ? {} : { lastAssistantMessage: output }, }, parent) }, @@ -193,18 +217,24 @@ export function createActivationObserver( } /** - * Why this child's last ordinary turn ended, for the terminal lifecycle edge. - * The child's own `turn/end` is authoritative: teardown succeeding says nothing - * about whether the model errored, hit its token ceiling, or was cancelled, so - * deriving the reason from disposal would report failed work as completed. + * Why this child's epoch ended, for the terminal lifecycle edge and the + * manager's own parent delivery. The child's own log is authoritative: + * teardown succeeding says nothing about whether the model errored, hit its + * token ceiling, or was cancelled, so deriving the reason from disposal would + * report failed work as completed. + * + * {@link foldConsumedWork} supplies both halves the raw turn sequence cannot: + * which turn accounts for the work this epoch consumed, and whether accepted + * work was cancelled after it without any turn opening over it. A recorded + * failure still wins over a cancellation — stopping a child that had already + * failed does not turn its failure into a cancellation. * @param events - this epoch's own event suffix. - * @returns its terminal stop reason; `completed` when no ordinary turn closed. + * @returns its terminal stop reason; `completed` only for an epoch that both + * closed cleanly and had nothing left to run. */ function epochStopReason(events: readonly SessionEvent[]): SubagentResult['stopReason'] { - const reason = findLastMessageTurnEnd(events)?.data.reason - // No ordinary turn closed, so nothing failed either. - if (reason === undefined) return 'completed' - switch (reason.kind) { + const { end, droppedUnrun } = foldConsumedWork(events) + switch (end?.data.reason.kind) { case 'max-tokens': return 'max-tokens' case 'aborted': @@ -212,8 +242,15 @@ function epochStopReason(events: readonly SessionEvent[]): SubagentResult['stopR return 'aborted' case 'error': return 'error' + // A pre-step rejection — a hook deny, a policy plugin — discarded input + // this epoch had claimed: the work was declined, not done. + case 'blocked': + return 'refusal' + // A clean ending and no accounting turn at all share one rule: the epoch + // finished what it was given unless a cancelled queue says otherwise. + case undefined: case 'completed': - return 'completed' + return droppedUnrun ? 'aborted' : 'completed' /* v8 ignore next 3 -- `TurnEndReason` is merge-extensible, so this arm needs a * backend that adds a variant; treating an unnameable reason as success would * report failed work as completed. */ diff --git a/packages/subagent/subagent/tests/continuation.spec.ts b/packages/subagent/subagent/tests/continuation.spec.ts index 50bd71b9cc..828717c1aa 100644 --- a/packages/subagent/subagent/tests/continuation.spec.ts +++ b/packages/subagent/subagent/tests/continuation.spec.ts @@ -15,7 +15,7 @@ import type { GenerateOptions, MessageId, StreamChunk } from '@deepseek-ai/dsh-l import { CallId, createUserMessage, LlmAdapter } from '@deepseek-ai/dsh-llm' import { defineTool } from '@deepseek-ai/dsh-tools' import InvariantService from '@deepseek-ai/dsh-invariants' -import { MockAdapter, textResponse, toolCallResponse } from '../../../core/agent-loop/tests/mock-adapter.ts' +import { MockAdapter, maxTokensResponse, textResponse, toolCallResponse } from '../../../core/agent-loop/tests/mock-adapter.ts' import SubagentService, { SubagentError, SUBAGENT_DESCRIPTOR_VERSION, @@ -142,6 +142,18 @@ async function waitNoActivation(ctx: Context, childId: SessionId): Promise }, { timeout: 5_000 }) } +/** + * Keep the top-level test parent out of a scripted model corpus. Every child + * settlement wakes its parent, so a suite that scripts only child responses + * would otherwise spend them on the parent's own turns. + */ +function parkParent(ctx: Context, parent: Agent): void { + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject !== parent) return next() + return { kind: 'reject' as const } + }) +} + /** Observe calls at the Agent cancellation boundary without a production event. */ function observeCancel(agent: Agent, callback: () => void): void { const cancel = agent.cancel.bind(agent) @@ -531,7 +543,10 @@ describe('SubagentService.followup residency routing', () => { await waitNoActivation(ctx, grandchild.childId) await waitNoActivation(ctx, started.childId) const loaded = await ctx.sessionPersistence.load(started.childId) - expect(userTexts(loaded.events)).toEqual(['child task', 'while waiting']) + // This child is itself a parent, so its grandchild's settlement notice is + // an ordinary later user message in its log. + expect(userTexts(loaded.events).slice(0, 2)).toEqual(['child task', 'while waiting']) + expect(userTexts(loaded.events).slice(2).join('\n')).toContain('finished and will do no further work') }) it('rejects a parent that is not the durable direct parent', async () => { @@ -1182,6 +1197,7 @@ describe('continuable review regressions', () => { it('reports this epoch\'s own output, captured while the child was still live', async () => { const { ctx, parent } = await setup([textResponse('first answer'), textResponse('second answer')]) + parkParent(ctx, parent) const ends: SubagentRunEndInfo[] = [] ctx.on('subagent/end', (info) => { ends.push(info) }) @@ -1254,9 +1270,10 @@ describe('continuable review regressions', () => { await waitNoActivation(ctx, started.childId) await vi.waitFor(() => { expect(ends).toHaveLength(1) }) - // Reading the whole session would resurrect 'first answer' here. + // Reading the whole session would resurrect 'first answer' here. The + // rejection discarded the claimed follow-up, so the epoch reads as refused. expect(ends[0]!.lastAssistantMessage).toBeUndefined() - expect(ends[0]!.stopReason).toBe('completed') + expect(ends[0]!.stopReason).toBe('refusal') }) it('reports handle-disposal failure on the terminal edge', async () => { @@ -1439,11 +1456,13 @@ describe('continuable review regressions', () => { await waitNoActivation(ctx, started.childId) }) - it('reports completed when no ordinary turn closed', async () => { + it('reports a prompt a pre-step rejection discarded as refusal', async () => { const { ctx, parent } = await setup([]) + parkParent(ctx, parent) const ends: SubagentRunEndInfo[] = [] ctx.on('subagent/end', (info) => { ends.push(info) }) - // Block admission so the child's only turn never opens. + // A UserPromptSubmit deny or a policy plugin: the child claims its prompt, + // the rejection discards it, and no step ever runs. ctx.on('agent/pre-step', async ({ agent: subject }, next) => { if (subject === parent) return next() return { kind: 'reject' } @@ -1452,8 +1471,10 @@ describe('continuable review regressions', () => { const started = await ctx.subagents.startContinuable(startSpec(parent)) await waitNoActivation(ctx, started.childId) + // The parent would otherwise believe a vetoed delivery was done and never + // resend it — the one failure the settlement promise says cannot happen. await vi.waitFor(() => { expect(ends).toHaveLength(1) }) - expect(ends[0]!.stopReason).toBe('completed') + expect(ends[0]!.stopReason).toBe('refusal') }) it('retains the Activation while an accepted message is still in the inbox', async () => { @@ -1482,15 +1503,538 @@ describe('continuable review regressions', () => { expect(ctx.agents.get(started.childId)).toBe(child) releaseFirst.resolve(undefined) await waitNoActivation(ctx, started.childId) - expect(adapter.requests).toHaveLength(2) + // Two child turns; the third request is the parent's own turn on the + // settlement notice. + expect(adapter.requests.filter(request => request.sessionId === started.childId)).toHaveLength(2) const loaded = await ctx.sessionPersistence.load(started.childId) expect(hasUserText(loaded.events, 'queued')).toBe(true) }) }) +/** Every settlement notice this agent received, in order, as flat text. */ +function settlementNotices(agent: Agent): { sender: string; text: string; summary: string }[] { + const logged = agent.session.events.flatMap(event => event.type === 'user/message' ? [event.data] : []) + return [...logged, ...agent.inbox.nextStep, ...agent.inbox.nextTurn].flatMap((message) => { + if (message.source.kind !== 'subagent-settled') return [] + return [{ + sender: message.source.senderSessionId, + summary: message.source.summary, + text: message.content.flatMap(block => block.type === 'text' ? [block.text] : []).join('\n'), + }] + }) +} + +describe('continuable settlement delivery', () => { + it('tells the parent what the child finished with, without being asked', async () => { + const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')]) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await waitNoActivation(ctx, started.childId) + + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + const notice = settlementNotices(parent)[0]! + expect(notice.sender).toBe(started.childId) + expect(notice.text).toBe( + `Background subagent ${started.childId} finished and will do no further work unless you send it more.` + + '\nIts closing message:\nthe answer', + ) + // The collapsed row states the outcome without the child's content. + expect(notice.summary).toBe( + `Background subagent ${started.childId} finished and will do no further work unless you send it more.`, + ) + }) + + it('delivers even when the child already reported for itself', async () => { + const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')]) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + const child = await vi.waitFor(() => { + const live = ctx.agents.get(started.childId) + expect(live).toBeDefined() + return live! + }) + await ctx.subagents.reportFrom(child, message('an explicit report'), { + delivery: 'quiet', + signal: testSignal, + }) + await waitNoActivation(ctx, started.childId) + + // The contract is unconditional precisely so the parent-side tool + // description can promise it; bookkeeping "did it report?" would make the + // promise conditional on a channel this manager does not own. + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + }) + + it('delivers the terminal reason when the child never had a chance to report', async () => { + const { ctx, parent } = await setup([maxTokensResponse('half an ans'), textResponse('parent ack')]) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await waitNoActivation(ctx, started.childId) + + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} ran out of room before it finished.` + + '\nIts closing message:\nhalf an ans', + ) + }) + + it('tells the parent a policy-rejected delivery was declined, not finished', async () => { + const { ctx, parent } = await setup([textResponse('parent ack')]) + // A pre-step rejection on the child — a UserPromptSubmit deny, a policy + // plugin — discards the claimed prompt without running it. + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject === parent) return next() + return { kind: 'reject' } + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await waitNoActivation(ctx, started.childId) + + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} declined the task.` + + '\nIt left no closing message.', + ) + }) + + it('reports a turn that failed before reaching its first step', async () => { + const releaseFirst = Promise.withResolvers() + const adapter = new GatedAdapter([ + { chunks: textResponse('the answer'), gate: releaseFirst.promise }, + { chunks: textResponse('parent ack') }, + ]) + const { ctx, parent } = await setupWith(adapter) + // The shipped durability checkpoint (`dsh-session-checkpoint-policy`) is + // fail-closed at the step boundary, so a rejected write ends the turn after + // it claimed its messages and before it entered a step. + ctx.on('agent/pre-step', async ({ agent: subject, turn }, next) => { + if (subject.session.header.parentSession === undefined || turn < 2) return next() + throw new Error('ENOSPC: no space left on device') + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await followup(ctx, parent, started.childId, message('second task')) + releaseFirst.resolve(undefined) + await waitNoActivation(ctx, started.childId) + + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + // The parent must not be told the child finished: the delivery it is still + // waiting on was claimed out of the inbox and then swallowed by the failure. + const child = await ctx.sessionPersistence.load(started.childId) + expect(hasUserText(child.events, 'second task')).toBe(false) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} failed before it finished.` + + '\nIts closing message:\nthe answer', + ) + }) + + it('reports accepted work cut short before its first step as stopped', async () => { + const releaseFirst = Promise.withResolvers() + const releaseCheckpoint = Promise.withResolvers() + const atCheckpoint = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('the answer'), gate: releaseFirst.promise }]) + const { ctx, parent } = await setupWith(adapter) + // A step-boundary participant — the shipped durability checkpoint, a hook, + // prompt assembly — holding the child's second turn open before its first + // step, which is where teardown cancellation then catches it. + ctx.on('agent/pre-step', async ({ agent: subject, turn }, next) => { + if (subject.session.header.parentSession === undefined || turn < 2) return next() + atCheckpoint.resolve(undefined) + await releaseCheckpoint.promise + return next() + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + // Queued while turn 1 still runs, so turn 2 opens and claims it without a + // second model call: the Activation is mid-turn when the drain cancels it. + await followup(ctx, parent, started.childId, message('second task')) + releaseFirst.resolve(undefined) + await atCheckpoint.promise + const drained = drainManager(ctx) + releaseCheckpoint.resolve(undefined) + await drained + + // Turn 2 leaves a balanced no-step `aborted` end, so the log alone would + // answer with turn 1's clean completion and tell the parent its still-unrun + // task had finished. + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} was stopped before it finished.` + + '\nIts closing message:\nthe answer', + ) + }) + + it('reports a child stopped before it ever reached the model as stopped', async () => { + const releaseCheckpoint = Promise.withResolvers() + const atCheckpoint = Promise.withResolvers() + const { ctx, parent } = await setupWith(new GatedAdapter([])) + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject.session.header.parentSession === undefined) return next() + atCheckpoint.resolve(undefined) + await releaseCheckpoint.promise + return next() + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await atCheckpoint.promise + const drained = drainManager(ctx) + releaseCheckpoint.resolve(undefined) + await drained + + // This epoch closed no stepped turn at all, which on its own reads as "had + // nothing to report"; only the interruption distinguishes it from a child + // that genuinely finished with no output. + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} was stopped before it finished.` + + '\nIt left no closing message.', + ) + }) + + it('reports a child an ancestor interrupted before its first step as stopped', async () => { + const atCheckpoint = Promise.withResolvers() + const releaseCheckpoint = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('parent ack') }]) + const { ctx, parent } = await setupWith(adapter) + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject.session.header.parentSession === undefined) return next() + atCheckpoint.resolve(undefined) + await releaseCheckpoint.promise + return next() + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await atCheckpoint.promise + // The shipped interrupt path: nothing about it runs inside this manager, so + // no pre-cancel sample could see it — the child's own log has to say so. + ctx.subagents.interrupt(started.childId, { kind: 'ancestor', agent: parent }) + releaseCheckpoint.resolve(undefined) + await waitNoActivation(ctx, started.childId) + + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} was stopped before it finished.` + + '\nIt left no closing message.', + ) + }) + + it('reports accepted work cancelled before any turn could open as stopped', async () => { + const releaseChild = Promise.withResolvers() + const releaseGrandchild = Promise.withResolvers() + const releaseMaintenance = Promise.withResolvers() + const adapter = new GatedAdapter([ + { chunks: textResponse('the answer'), gate: releaseChild.promise }, + { chunks: textResponse('grandchild'), gate: releaseGrandchild.promise }, + ]) + const { ctx, parent } = await setupWith(adapter) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + const child = await vi.waitFor(() => { + const live = ctx.agents.get(started.childId) + expect(live).toBeDefined() + return live! + }) + // A descendant keeps the child resident once its own turn closes, so the + // maintenance phase below is reachable without racing settlement. + const grandchild = await ctx.subagents.startContinuable(startSpec(child)) + await vi.waitFor(() => { expect(ctx.agents.get(grandchild.childId)).toBeDefined() }) + releaseChild.resolve(undefined) + await vi.waitFor(() => { expect(child.status).toBe('idle') }) + + // Context maintenance folds into `idle` and defers waking work, so this + // delivery is accepted with no turn to claim it. + const maintaining = child.runMaintenance(async () => { await releaseMaintenance.promise }) + await followup(ctx, parent, started.childId, message('never runs')) + const drained = drainManager(ctx) + releaseMaintenance.resolve(undefined) + releaseGrandchild.resolve(undefined) + await maintaining + await drained + + // Turn 1 closed cleanly and no later turn opened, so the cancelled queue is + // the only record that this epoch was cut short. + expect(hasUserText(child.session.events, 'never runs')).toBe(false) + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} was stopped before it finished.` + + '\nIts closing message:\nthe answer', + ) + }) + + it('withholds an outcome the harness could not durably release', async () => { + const { ctx, parent } = await setup([textResponse('the answer'), textResponse('parent ack')]) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + const manager = (ctx.subagents as unknown as { + continuations: { activations: Map } }> } + }).continuations + const activation = await vi.waitFor(() => { + const live = manager.activations.get(started.childId) + expect(live).toBeDefined() + return live! + }) + const dispose = activation.handle.dispose.bind(activation.handle) + activation.handle.dispose = async () => { + await dispose() + throw new Error('scope unwind failed') + } + + await waitNoActivation(ctx, started.childId) + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(1) }) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} failed before it finished.\nIt left no closing message.`, + ) + }) + + it('gives an idle parent one ordinary turn on the notice', async () => { + const { ctx, parent, adapter } = await setup([textResponse('the answer'), textResponse('parent ack')]) + const turnStarts: number[] = [] + ctx.on('session/event', (session, event) => { + if (session.id === parent.id && event.type === 'turn/start') turnStarts.push(event.data.turn) + }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await waitNoActivation(ctx, started.childId) + await vi.waitFor(() => { + expect(adapter.requests.filter(request => request.sessionId === parent.id)).toHaveLength(1) + }) + expect(turnStarts).toEqual([1]) + }) + + it('batches simultaneous notices into one step of a busy parent', async () => { + const releaseChildren = Promise.withResolvers() + const releaseParent = Promise.withResolvers() + const adapter = new GatedAdapter([ + { chunks: textResponse('parent works'), gate: releaseParent.promise }, + { chunks: textResponse('first child'), gate: releaseChildren.promise }, + { chunks: textResponse('second child'), gate: releaseChildren.promise }, + { chunks: textResponse('parent reacts') }, + ]) + const { ctx, parent } = await setupWith(adapter) + // Open a parent turn first, so both notices arrive while it is running. + parent.followup(createUserMessage({ content: message('start working'), source: { kind: 'user' } })) + await vi.waitFor(() => { expect(parent.status).toBe('running') }) + + const first = await ctx.subagents.startContinuable(startSpec(parent)) + const second = await ctx.subagents.startContinuable(startSpec(parent)) + releaseChildren.resolve(undefined) + await waitNoActivation(ctx, first.childId) + await waitNoActivation(ctx, second.childId) + + // Both notices are waiting for the same step boundary, not two turns. + expect(parent.inbox.nextStep).toHaveLength(2) + expect(parent.inbox.nextTurn).toHaveLength(0) + const turnStarts: number[] = [] + ctx.on('session/event', (session, event) => { + if (session.id === parent.id && event.type === 'turn/start') turnStarts.push(event.data.turn) + }) + releaseParent.resolve(undefined) + await vi.waitFor(() => { expect(settlementNotices(parent)).toHaveLength(2) }) + expect(turnStarts).toEqual([]) + // Both children released together, so which settles first is not ordered. + expect(new Set(settlementNotices(parent).map(entry => entry.sender))) + .toEqual(new Set([first.childId, second.childId])) + }) + + it('holds a maintaining parent live until it can read the notice', async () => { + const releaseFirst = Promise.withResolvers() + const releaseSecond = Promise.withResolvers() + const adapter = new GatedAdapter([ + { chunks: textResponse('outer') }, + { chunks: textResponse('first inner'), gate: releaseFirst.promise }, + { chunks: textResponse('second inner'), gate: releaseSecond.promise }, + { chunks: textResponse('outer reacts') }, + { chunks: textResponse('root reacts') }, + ]) + const { ctx, parent } = await setupWith(adapter) + const outer = await ctx.subagents.startContinuable(startSpec(parent)) + const middle = await vi.waitFor(() => { + const live = ctx.agents.get(outer.childId) + expect(live).toBeDefined() + return live! + }) + const first = await ctx.subagents.startContinuable(startSpec(middle)) + const second = await ctx.subagents.startContinuable(startSpec(middle)) + await vi.waitFor(() => { expect(middle.status).toBe('idle') }) + + // `Agent.status` folds maintenance into `idle`, and a waking send behind it + // only arms a deferred wake. The first child's release moves the middle + // Activation's settlement watcher onto its quiescence race; the second one + // then arrives at exactly the point where an unaccounted delivery would be + // judged quiet, settled, and cancelled — clearing the inbox it sits in. + const maintaining = Promise.withResolvers() + const maintenance = middle.runMaintenance(async () => { await maintaining.promise }) + releaseFirst.resolve(undefined) + await waitNoActivation(ctx, first.childId) + releaseSecond.resolve(undefined) + await waitNoActivation(ctx, second.childId) + expect(ctx.agents.get(outer.childId)).toBe(middle) + + maintaining.resolve(undefined) + await maintenance + await vi.waitFor(() => { expect(settlementNotices(middle)).toHaveLength(2) }) + expect(settlementNotices(middle).map(entry => entry.sender)) + .toEqual([first.childId, second.childId]) + await waitNoActivation(ctx, outer.childId) + }) + + it('delivers before releasing the ownership that lets the parent settle', async () => { + const releaseChild = Promise.withResolvers() + const adapter = new GatedAdapter([ + { chunks: textResponse('outer') }, + { chunks: textResponse('inner'), gate: releaseChild.promise }, + { chunks: textResponse('outer reacts') }, + ]) + const { ctx, parent } = await setupWith(adapter) + const outer = await ctx.subagents.startContinuable(startSpec(parent)) + const middle = await vi.waitFor(() => { + const live = ctx.agents.get(outer.childId) + expect(live).toBeDefined() + return live! + }) + const inner = await ctx.subagents.startContinuable(startSpec(middle)) + await vi.waitFor(() => { expect(middle.status).toBe('idle') }) + + const manager = (ctx.subagents as unknown as { + continuations: { activations: Map }> } + }).continuations + let ownedAtDelivery: SessionId[] | undefined + ctx.on('agent/inbox/inserted', ({ agent, message }) => { + if (agent !== middle || message.source.kind !== 'subagent-settled') return + ownedAtDelivery = [...manager.activations.get(middle.id)!.ownedChildren] + }) + + releaseChild.resolve(undefined) + await waitNoActivation(ctx, inner.childId) + // Still owned at delivery: the parent is structurally unable to settle in + // the window the notice crosses, rather than winning a race against it. + expect(ownedAtDelivery).toEqual([inner.childId]) + await waitNoActivation(ctx, outer.childId) + }) + + it('does not wake a parent whose own teardown already began', async () => { + const hold = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }]) + const { ctx, parent } = await setupWith(adapter) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() }) + + const drained = drainManager(ctx) + hold.resolve(undefined) + await drained + + // Delivered and durably logged, but no turn: waking a parent the host is + // about to dispose spends a model request nothing reads. What happens to the + // message when that parent is disposed next is pinned by the test below. + expect(settlementNotices(parent)).toHaveLength(1) + expect(settlementNotices(parent)[0]!.text).toBe( + `Background subagent ${started.childId} was stopped before it finished.` + + '\nIt left no closing message.', + ) + expect(parent.session.events.some(event => event.type === 'agent/inbox/spliced')).toBe(true) + expect(parent.session.events.some(event => event.type === 'turn/start')).toBe(false) + expect(parent.status).toBe('idle') + }) + + it('does not wake a parent below a scoped teardown root', async () => { + const hold = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }]) + const { ctx, parent } = await setupWith(adapter) + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() }) + + const drained = ctx.subagents.drainContinuableDescendants([parent]) + hold.resolve(undefined) + await drained + + expect(settlementNotices(parent)).toHaveLength(1) + expect(parent.session.events.some(event => event.type === 'turn/start')).toBe(false) + }) + + it('records but cannot deliver a teardown notice once the parent is disposed too', async () => { + const hold = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('interrupted'), gate: hold.promise }]) + const { ctx } = await setupWith(adapter) + const parentId = SessionId('closing-parent') + const host = await ctx.agents.create({ + sessionId: parentId, + agentOptions: { provider: 'mock', model: 'mock' }, + }) + const started = await ctx.subagents.startContinuable(startSpec(host.agent)) + await vi.waitFor(() => { expect(ctx.agents.get(started.childId)).toBeDefined() }) + + const drained = ctx.subagents.drainContinuableDescendants([host.agent]) + hold.resolve(undefined) + await drained + expect(settlementNotices(host.agent)).toHaveLength(1) + + // Disposal is a `keepInbox: false` cancel, so it durably cancels the notice + // it never claimed. Teardown delivery therefore reaches a parent that is + // still resident — a resumed one reads the log, not a pending message — and + // no wording anywhere may promise otherwise. + await host.dispose() + const resumed = await ctx.agents.resume({ + resumeSessionId: parentId, + agentOptions: { provider: 'mock', model: 'mock' }, + }) + expect(settlementNotices(resumed.agent)).toEqual([]) + await resumed.dispose() + // The account is still in the durable log: delivered, then cancelled unread. + const persisted = await ctx.sessionPersistence.load(parentId) + expect(persisted.events.flatMap(event => event.type === 'agent/inbox/spliced' + ? [{ inserted: event.data.inserted.length, removed: event.data.removedCount ?? 0 }] + : [])).toEqual([{ inserted: 1, removed: 0 }, { inserted: 0, removed: 1 }]) + }) + + it('drops the notice without disturbing teardown when the parent is gone', async () => { + const releaseChild = Promise.withResolvers() + const adapter = new GatedAdapter([{ chunks: textResponse('answer'), gate: releaseChild.promise }]) + const { ctx } = await setupWith(adapter) + const warnings: string[] = [] + ctx.logger.warn = (text: string) => { warnings.push(text) } + const host = await ctx.agents.create({ + sessionId: SessionId('disposable-parent'), + agentOptions: { provider: 'mock', model: 'mock' }, + }) + const started = await ctx.subagents.startContinuable(startSpec(host.agent)) + const ends: SubagentRunEndInfo[] = [] + ctx.on('subagent/end', (info) => { ends.push(info) }) + + releaseChild.resolve(undefined) + await host.dispose() + await waitNoActivation(ctx, started.childId) + await vi.waitFor(() => { expect(ends).toHaveLength(1) }) + expect(warnings).toEqual([]) + }) + + it('logs a rejected notice instead of failing the child\'s teardown', async () => { + const { ctx, parent } = await setup([textResponse('the answer')]) + const warnings: string[] = [] + ctx.logger.warn = (text: string) => { warnings.push(text) } + vi.spyOn(parent, 'followup').mockImplementation(() => { + throw new Error('parent closed during delivery') + }) + const ends: SubagentRunEndInfo[] = [] + ctx.on('subagent/end', (info) => { ends.push(info) }) + + const started = await ctx.subagents.startContinuable(startSpec(parent)) + await waitNoActivation(ctx, started.childId) + await vi.waitFor(() => { expect(ends).toHaveLength(1) }) + expect(ends[0]!.stopReason).toBe('completed') + expect(warnings.some(warning => warning.includes('settlement notice was not delivered'))).toBe(true) + }) + + it('stays silent about a child the caller was told does not exist', async () => { + const { ctx, parent } = await setup([]) + const drains: Promise[] = [] + ctx.on('subagent/start', () => { drains.push(drainManager(ctx)) }) + + await expect(ctx.subagents.startContinuable(startSpec(parent))) + .rejects.toMatchObject({ code: 'DRAINING' }) + await Promise.all(drains) + expect(settlementNotices(parent)).toEqual([]) + }) +}) + describe('continuable lifecycle observation', () => { it('emits one paired start/end per residency epoch', async () => { const { ctx, parent } = await setup([textResponse('first'), textResponse('second')]) + parkParent(ctx, parent) const starts: SubagentRunInfo[] = [] const ends: SubagentRunEndInfo[] = [] ctx.on('subagent/start', (info) => { starts.push(info) }) @@ -1510,6 +2054,8 @@ describe('continuable lifecycle observation', () => { expect(starts.map(info => info.provider)).toEqual(['spawn', 'spawn']) // Each end pairs its own start's runId. expect(ends.map(info => info.runId)).toEqual(starts.map(info => info.runId)) + // Both epochs ran their own scripted response; neither exhausted the corpus. + expect(ends.map(info => info.stopReason)).toEqual(['completed', 'completed']) }) }) diff --git a/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts b/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts index 1c12602dd2..21cb242844 100644 --- a/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts +++ b/packages/subagent/tool-subagent-control/tests/list-agents.spec.ts @@ -16,6 +16,7 @@ import type { GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm' import { LlmAdapter } from '@deepseek-ai/dsh-llm' import { MockAdapter, textResponse } from '../../../core/agent-loop/tests/mock-adapter.ts' import * as tool from '../src/list-agents.ts' +import { parkParent } from './park-parent.ts' /** One scripted response that may wait on a caller-released gate before streaming. */ interface GatedEntry { @@ -63,6 +64,7 @@ async function setupWith(adapter: MockAdapter | GatedAdapter) { await ctx.plugin(tool) ctx.llm.registerAdapter(['mock'], adapter) const parent = ctx.agentLoop.create(SessionId('parent'), { provider: 'mock', model: 'mock' }) + parkParent(ctx, parent) return { ctx, parent, adapter } } diff --git a/packages/subagent/tool-subagent-control/tests/park-parent.ts b/packages/subagent/tool-subagent-control/tests/park-parent.ts new file mode 100644 index 0000000000..7f61f55d51 --- /dev/null +++ b/packages/subagent/tool-subagent-control/tests/park-parent.ts @@ -0,0 +1,22 @@ +/** + * Shared suite helper: keep this package's stand-in parent out of a scripted + * model corpus. + * @module park-parent + */ + +import type { Context } from '@deepseek-ai/cordis' +import type { SessionId } from '@deepseek-ai/dsh-session' + +/** + * Reject every step of the stand-in parent. Each child settlement wakes its + * parent, and these suites size their scripts for child turns only; the tests + * assert on delivery rather than on the parent's own turn. + * @param ctx - the booted test context. + * @param parent - the stand-in parent whose turns must not reach the model. + */ +export function parkParent(ctx: Context, parent: { id: SessionId }): void { + ctx.on('agent/pre-step', async ({ agent: subject }, next) => { + if (subject.id !== parent.id) return next() + return { kind: 'reject' as const } + }) +} diff --git a/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts b/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts index 4c378fcbc6..4960992068 100644 --- a/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts +++ b/packages/subagent/tool-subagent-control/tests/tool-subagent-control.spec.ts @@ -15,6 +15,7 @@ import type { GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm' import { LlmAdapter } from '@deepseek-ai/dsh-llm' import { MockAdapter, textResponse } from '../../../core/agent-loop/tests/mock-adapter.ts' import * as tool from '../src/index.ts' +import { parkParent } from './park-parent.ts' /** One scripted response that may wait on a caller-released gate before streaming. */ interface GatedEntry { @@ -62,6 +63,7 @@ async function setupWith(adapter: MockAdapter | GatedAdapter) { await ctx.plugin(tool) ctx.llm.registerAdapter(['mock'], adapter) const parent = ctx.agentLoop.create(SessionId('parent'), { provider: 'mock', model: 'mock' }) + parkParent(ctx, parent) return { ctx, parent, adapter } } diff --git a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts index 53a84294cf..ac106a5680 100644 --- a/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts +++ b/packages/subagent/tool-subagent-report/tests/tool-subagent-report.spec.ts @@ -549,6 +549,9 @@ describe('dsh-tool-subagent-report result independence', () => { expect(ctx.agents.get(started.childId) === undefined).toBe(true) }, { timeout: 5_000 }) + // The parent does learn the child settled — that account is the + // continuation service's, carried under its own `subagent-settled` source. + // Nothing turns the child's final answer into a report it did not send. expect(reports(parent)).toEqual([]) expect(userTexts((await ctx.sessionPersistence.load(started.childId)).events)).toEqual(['child task']) expect(ctx.get('tasks')).toBeUndefined() diff --git a/packages/subagent/tool-subagent/README.i18n.yaml b/packages/subagent/tool-subagent/README.i18n.yaml index 8c41c6413a..0292c0a51c 100644 --- a/packages/subagent/tool-subagent/README.i18n.yaml +++ b/packages/subagent/tool-subagent/README.i18n.yaml @@ -2,5 +2,5 @@ # side as of the last confirmed-consistent state. Both languages carry equal authority; # after editing either side, bring the other along and re-record with: # pnpm run verify-translation-pairing --write packages/subagent/tool-subagent/README.md -README.md: ac3ec0563cce9128608ca31860b034a103dc1a3a -README.zh.md: d64831d7cf64800ad3307ce6cb7f294500a0a6f0 +README.md: a46300cc5dffcbc420ed9525d8f081c670771294 +README.zh.md: 985d33beb2ac7de60527b457c4786a0257232611 diff --git a/packages/subagent/tool-subagent/README.md b/packages/subagent/tool-subagent/README.md index ac3ec0563c..a46300cc5d 100644 --- a/packages/subagent/tool-subagent/README.md +++ b/packages/subagent/tool-subagent/README.md @@ -10,7 +10,7 @@ Each plugin instance binds one `provider` to one `toolName`; the model receives A foreground call passes the execution signal through startup and execution, awaits `run.result`, and always awaits `run.dispose()` before returning. Only `completed` returns the canonical `{ kind: 'foreground', runId, output: JsonValue[] }`, rendered as the same final text; abort, refusal, token limit, and other failures become errored tool results whose message appends the child's preserved partial text (the `SubagentResult.output` selection) after the stop-reason headline, so a truncated answer is never reported as success yet never silently lost. If result collection and disposal both reject, the errored result preserves both diagnostics. -With `run_in_background: true`, `backgroundMode` selects the route. `one-shot` registers a plain parent-owned Task and returns canonical `{ kind: 'background', taskId }`, rendered as `started background subagent task `, even when the provider supports continuable children; generic task tools own its later status, collection, cancellation, and notices. `continuable` requires a provider with the `prepareContinuable` capability, calls `ctx.subagents.startContinuable()`, and returns `{ kind: 'continuable', subagentId }`, rendered as `started subagent `. The continuable route resolves at inbox acceptance: the child owns its own turns from there, so this call neither waits for nor collects a result, and the child does not report back — its transcript by that id is the source of its output, and the optional global `send_message` tool sends it more work. Starting continuable work does not require `send_message` to be loaded. See the [background subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md), and the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md). +With `run_in_background: true`, `backgroundMode` selects the route. `one-shot` registers a plain parent-owned Task and returns canonical `{ kind: 'background', taskId }`, rendered as `started background subagent task `, even when the provider supports continuable children; generic task tools own its later status, collection, cancellation, and notices. `continuable` requires a provider with the `prepareContinuable` capability, calls `ctx.subagents.startContinuable()`, and returns `{ kind: 'continuable', subagentId }`, rendered as `started subagent `. The continuable route resolves at inbox acceptance: the child owns its own turns from there, so this call neither waits for nor collects a result. The child's transcript by that id remains the source of its detailed output, and the optional global `send_message` tool sends it more work. The parent is not left guessing when to look, though: the continuation service delivers one settlement notice to it whenever a continuable child's Activation ends, which is why the schema tells the model it will be told and must not poll. Starting continuable work does not require `send_message` to be loaded. See the [background subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md), the [continuable subagents Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md), and the [merged-service Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md). `toolFilter` changes the child's global tool layer but is not a parent-derived authority ceiling. See the [agent-scope security non-goal](../../../.agents/notes/implemented/architecture/2026-07-08-agent-scope-contexts.md#security-and-authority-are-non-goals). @@ -37,7 +37,7 @@ Foreground and background calls are exclusive. Children may share the parent's w #### What the model sees -The generated default [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent) under this instance's configured name while its provider exists. Provider context inheritance changes the tool and prompt descriptions; enabled background mode adds `run_in_background`, and continuable mode describes starting a background subagent that keeps its conversation and returns its subagent id, while one-shot mode describes a background task id collected with `task_output` and stopped with `task_kill`. +The generated default [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent) under this instance's configured name while its provider exists. Provider context inheritance changes the tool and prompt descriptions; enabled background mode adds `run_in_background`, and continuable mode describes starting a background subagent that keeps its conversation, returns its subagent id, and reports its own completion — so the model is told never to poll or wait on it — while one-shot mode describes a background task id collected with `task_output` and stopped with `task_kill`. #### Token effect @@ -65,11 +65,11 @@ Append-only; newly visible content follows the reusable request prefix and does #### What the model sees -Start returns exactly `started subagent ` in configured continuable mode, or `started background subagent task ` in configured one-shot mode. In one-shot mode the generic task surface provides later status, final output, cancellation responses, and notices. In continuable mode the child does not report back; an independently loaded `send_message` tool delivers follow-ups, and the child's transcript by its id is the source of its output. +Start returns exactly `started subagent ` in configured continuable mode, or `started background subagent task ` in configured one-shot mode. In one-shot mode the generic task surface provides later status, final output, cancellation responses, and notices. In continuable mode this tool returns no result of its own; the child's settlement reaches the parent as a [service-owned notice](../subagent/README.md#settlement-notice), an independently loaded `send_message` tool delivers follow-ups, and the child's transcript by its id is the source of its detailed output. #### Token effect -The acknowledgement is retained; a one-shot final output enters parent history only when collected or injected, while a continuable child's output never returns through this tool. +The acknowledgement is retained; a one-shot final output enters parent history only when collected or injected, while a continuable child's output never returns through this tool — its settlement notice arrives independently of any tool result. #### KV Cache effect @@ -77,6 +77,6 @@ Append-only; newly visible content follows the reusable request prefix and does ## Known Limitations and Deferred Work -- **Background runs expose no result through this tool** — a one-shot task's final output is collected through the generic task surface, and a continuable child's output stays in its own session, read by its subagent id. +- **Background runs expose no result through this tool** — a one-shot task's final output is collected through the generic task surface, and a continuable child's output stays in its own session, read by its subagent id. The settlement notice states how that child ended and carries its closing message, but it is not this call's return value and cannot be awaited here. - **Duplicate names across waiting instances are detected late** (`TODO(subagent-dup-toolname)`) — preventing provider-registration rollback requires a registry of intended names. - **Child policy is fixed per instance** — another model, persona, tool filter, or depth cap requires another distinctly named tool. diff --git a/packages/subagent/tool-subagent/README.zh.md b/packages/subagent/tool-subagent/README.zh.md index d64831d7cf..985d33beb2 100644 --- a/packages/subagent/tool-subagent/README.zh.md +++ b/packages/subagent/tool-subagent/README.zh.md @@ -10,7 +10,7 @@ 前台调用会让执行信号贯穿启动和执行，等待 `run.result`，并且在返回前总会等待 `run.dispose()`。只有 `completed` 会返回规范值 `{ kind: 'foreground', runId, output: JsonValue[] }`，并渲染为相同的最终文本；中止、拒绝、token 上限和其他失败都会变成出错的工具结果，其消息在终止原因标题之后附带子代理保留下来的部分文本（即 `SubagentResult.output` 的选取结果）——被截断的回答不会被报告为成功，也绝不会被悄悄丢弃。如果结果收集与 dispose（资源释放）都 reject，出错的结果会保留两项诊断信息。 -设置 `run_in_background: true` 后，`backgroundMode` 会选择路由。`one-shot` 会注册一个归父级所有的普通 Task，并返回规范值 `{ kind: 'background', taskId }`，渲染为 `started background subagent task `，即使提供方支持可继续子 agent 也不例外；通用 Task 工具负责其后续状态、收集、取消和通知。`continuable` 要求提供方具备 `prepareContinuable` 能力，调用 `ctx.subagents.startContinuable()`，并返回 `{ kind: 'continuable', subagentId }`，渲染为 `started subagent `。可继续路由在 inbox 接受时结算：子 agent 自此拥有自己的轮次，因此该调用既不等待也不收集结果，而且子 agent 不会回报——通过该 id 查看其 transcript（文本记录）即是其输出来源，可选的全局 `send_message` 工具则向其发送更多工作。启动可继续工作不要求加载 `send_message`。见 [后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续的 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md)和[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)。 +设置 `run_in_background: true` 后，`backgroundMode` 会选择路由。`one-shot` 会注册一个归父级所有的普通 Task，并返回规范值 `{ kind: 'background', taskId }`，渲染为 `started background subagent task `，即使提供方支持可继续子 agent 也不例外；通用 Task 工具负责其后续状态、收集、取消和通知。`continuable` 要求提供方具备 `prepareContinuable` 能力，调用 `ctx.subagents.startContinuable()`，并返回 `{ kind: 'continuable', subagentId }`，渲染为 `started subagent `。可继续路由在 inbox 接受时结算：子 agent 自此拥有自己的轮次，因此该调用既不等待也不收集结果。通过该 id 查看其 transcript（文本记录）仍是其详细输出的来源，可选的全局 `send_message` 工具则向其发送更多工作。不过父级无需猜测何时查看：每当可继续子 agent 的 Activation 结束，继续执行服务都会向父级投递一条结算通知——正因如此，schema 才告诉模型它会被通知，且不得轮询。启动可继续工作不要求加载 `send_message`。见 [后台 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-08-background-subagent-tasks.md)、[可继续的 subagent Agent Note](../../../.agents/notes/implemented/feature/2026-07-28-continuable-subagent-conversations.md)和[服务合并 Agent Note](../../../.agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md)。 `toolFilter` 会改变子 agent 的全局工具层，但不是从父级派生的权限上限。见 [agent 作用域的安全非目标](../../../.agents/notes/implemented/architecture/2026-07-08-agent-scope-contexts.md#security-and-authority-are-non-goals)。 @@ -37,7 +37,7 @@ #### 模型看到的内容 -当提供方存在时，以当前实例配置的名称公开已生成的默认 [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent)。提供方是否继承上下文会改变工具描述和提示词描述；启用后台模式会添加 `run_in_background`，可继续模式描述为启动一个保留其对话并返回子 agent id 的后台子 agent，而一次性模式描述为返回一个用 `task_output` 收集、用 `task_kill` 停止的后台任务 id。 +当提供方存在时，以当前实例配置的名称公开已生成的默认 [`subagent` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-subagent)。提供方是否继承上下文会改变工具描述和提示词描述；启用后台模式会添加 `run_in_background`，可继续模式描述为启动一个保留其对话、返回子 agent id 并会自行报告完成的后台子 agent——因此模型被告知绝不要轮询或等待它——而一次性模式描述为返回一个用 `task_output` 收集、用 `task_kill` 停止的后台任务 id。 #### Token 影响 @@ -65,11 +65,11 @@ #### 模型看到的内容 -在配置的可继续模式下，启动时返回内容恰为 `started subagent `；在配置的一次性模式下，则返回 `started background subagent task `。一次性模式下，通用 Task 接口提供后续状态、最终输出、取消响应和通知。可继续模式下，子 agent 不会回报；独立加载的 `send_message` 工具会投递后续消息，而通过其 id 查看子 agent 的 transcript 即是其输出来源。 +在配置的可继续模式下，启动时返回内容恰为 `started subagent `；在配置的一次性模式下，则返回 `started background subagent task `。一次性模式下，通用 Task 接口提供后续状态、最终输出、取消响应和通知。可继续模式下，本工具不返回自己的结果；子 agent 的结算会以[服务负责的通知](../subagent/README.md#settlement-notice)到达父级，独立加载的 `send_message` 工具会投递后续消息，而通过其 id 查看子 agent 的 transcript 即是其详细输出来源。 #### Token 影响 -确认消息会被保留；一次性最终输出只在收集或注入时进入父级历史，而可继续子 agent 的输出绝不会通过本工具返回。 +确认消息会被保留；一次性最终输出只在收集或注入时进入父级历史，而可继续子 agent 的输出绝不会通过本工具返回——其结算通知独立于任何工具结果到达。 #### KV Cache 影响 @@ -77,6 +77,6 @@ ## 已知限制与暂缓事项 -- **后台运行不通过本工具公开结果**：一次性任务的最终输出通过通用 Task 接口收集，可继续子 agent 的输出留在其自身会话中，按其 subagent id 读取。 +- **后台运行不通过本工具公开结果**：一次性任务的最终输出通过通用 Task 接口收集，可继续子 agent 的输出留在其自身会话中，按其 subagent id 读取。结算通知会说明该子 agent 如何结束并携带其收尾消息，但它不是本次调用的返回值，也无法在此等待。 - **等待中实例的重复名称发现较晚**（`TODO(subagent-dup-toolname)`）：若要阻止提供方注册回滚，需要一份预期名称注册表。 - **每个实例的子 agent 策略固定**：其他模型、persona、工具过滤器或深度上限都需要另一个名称不同的工具。 diff --git a/packages/subagent/tool-subagent/src/index.ts b/packages/subagent/tool-subagent/src/index.ts index 13781bde91..bda7f8ae89 100644 --- a/packages/subagent/tool-subagent/src/index.ts +++ b/packages/subagent/tool-subagent/src/index.ts @@ -262,12 +262,13 @@ export function apply(ctx: Context, config: Config): void { disposeTool = ctx.tools.register(defineTool({ name: config.toolName ?? 'subagent', description: wording.description + (backgroundEnabled - // The return channel is a separately installed capability this package - // cannot observe, so this describes only this call's result. + // The completion notice is the continuation service's own behavior, not + // a separately installed capability, so this promise holds whenever the + // continuable background path is reachable at all. ? continuable ? ' Set `run_in_background: true` to start a background subagent that keeps its conversation:' - + ' you receive only its subagent id, never its result, and it works on its own. Use this for' - + ' work whose result you do not need returned by this call; `send_message` sends it more work.' + + ' this call returns only its subagent id, and the subagent works on its own from there. You' + + ' are told when it finishes, so never poll or wait on it; `send_message` sends it more work.' : ' Set `run_in_background: true` to return a task id; collect with `task_output` and stop with `task_kill`.' : ''), parameters: { @@ -286,7 +287,7 @@ export function apply(ctx: Context, config: Config): void { type: 'boolean' as const, description: continuable ? 'Run as a background subagent that keeps its conversation and return only its subagent id. ' - + 'This call never returns its result; send it more work with send_message.' + + 'This call does not wait for it; you are told when it finishes. Send it more work with send_message.' : 'Run as a background task and return its id; collect with task_output or stop with task_kill.', }, } : {}, diff --git a/scripts/type-equiv.manifest.json b/scripts/type-equiv.manifest.json index c1b1b2753a..9bf4d8ceb0 100644 --- a/scripts/type-equiv.manifest.json +++ b/scripts/type-equiv.manifest.json @@ -1165,6 +1165,11 @@ "symbol": "SubagentReportDelivery", "source": "packages/subagent/subagent/src/continuation.ts" }, + { + "doc": "docs/subsystems/subagent.md", + "symbol": "SubagentSettledMessageSource", + "source": "packages/subagent/subagent/src/continuation.ts" + }, { "doc": "docs/subsystems/subagent.md", "symbol": "SubagentReportOptions", From 567fcadba011088cadec538873ae837c65496fb0 Mon Sep 17 00:00:00 2001 From: Hypatia May Date: Tue, 11 Aug 2026 12:38:57 +0800 Subject: [PATCH 06/16] fix(subagent): report resumable children as ready MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit list_agents called a storage-only continuable child 'complete', inviting parents to poll it for a result that will never appear there. Rename the status to 'ready' — resumable, not terminal — and rewrite the tool description to say a parent is told when a child finishes, so listing is for recall, not polling. --- ...list-agents-residency-vocabulary.i18n.yaml | 6 +++ ...-08-06-list-agents-residency-vocabulary.md | 40 +++++++++++++++++++ ...-06-list-agents-residency-vocabulary.zh.md | 40 +++++++++++++++++++ ...subagent-catalog-and-list-agents.i18n.yaml | 4 +- ...urable-subagent-catalog-and-list-agents.md | 6 +-- ...ble-subagent-catalog-and-list-agents.zh.md | 6 +-- docs/subsystems/subagent.i18n.yaml | 4 +- docs/subsystems/subagent.md | 2 +- docs/subsystems/subagent.zh.md | 2 +- docs/tool-catalog.i18n.yaml | 4 +- docs/tool-catalog.md | 2 +- docs/tool-catalog.zh.md | 2 +- .../advanced-toolchain/session.1.jsonl | 2 +- .../advanced-toolchain/session.2.jsonl | 2 +- .../system-prompt.expected.md | 4 +- .../tool-schemas.expected.json | 2 +- .../both-mode-turn/tool-schemas.expected.json | 2 +- .../code-mode-turn/system-prompt.expected.md | 4 +- .../lsp-definition/tool-schemas.expected.json | 2 +- .../tool-schemas.expected.json | 2 +- .../tool-schemas.expected.json | 2 +- .../pty-tools/tool-schemas.expected.json | 2 +- .../tool-schemas.expected.json | 2 +- .../tool-schemas.expected.json | 2 +- .../tool-schemas.1.expected.json | 2 +- .../tool-schemas.1.expected.json | 2 +- .../session.1.jsonl | 2 +- .../session.2.jsonl | 2 +- .../subagent-list-agents/session.jsonl | 2 +- .../tool-schemas.1.expected.json | 2 +- .../snapshots/subagent-mixed/session.1.jsonl | 2 +- .../snapshots/subagent-mixed/session.2.jsonl | 2 +- .../snapshots/subagent-multi/session.1.jsonl | 2 +- .../snapshots/subagent-multi/session.2.jsonl | 2 +- .../tool-schemas.1.expected.json | 2 +- .../text-turn/tool-schemas.expected.json | 2 +- .../web-fetch/tool-schemas.expected.json | 2 +- .../advanced-toolchain/session.1.jsonl | 8 ++-- .../advanced-toolchain/session.2.jsonl | 8 ++-- .../advanced-toolchain/session.jsonl | 26 ++++++------ .../compaction-recovery/session.jsonl | 12 +++--- .../tests/snapshots/pty-tools/session.jsonl | 32 +++++++-------- packages/subagent/subagent/README.i18n.yaml | 4 +- packages/subagent/subagent/README.md | 2 +- packages/subagent/subagent/README.zh.md | 2 +- .../tool-subagent-control/README.i18n.yaml | 4 +- .../subagent/tool-subagent-control/README.md | 6 +-- .../tool-subagent-control/README.zh.md | 6 +-- .../tool-subagent-control/src/list-agents.ts | 20 ++++++---- .../tests/list-agents.spec.ts | 22 ++++++++-- 50 files changed, 215 insertions(+), 109 deletions(-) create mode 100644 .agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml create mode 100644 .agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md create mode 100644 .agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.zh.md diff --git a/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml new file mode 100644 index 0000000000..5f01146dbe --- /dev/null +++ b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.i18n.yaml @@ -0,0 +1,6 @@ +# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each +# side as of the last confirmed-consistent state. Both languages carry equal authority; +# after editing either side, bring the other along and re-record with: +# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md +2026-08-06-list-agents-residency-vocabulary.md: 01fff958921465909f5cd300dc355fdac6ec0772 +2026-08-06-list-agents-residency-vocabulary.zh.md: d68b9857de23ccb0b68f0bfafba153fde63303a7 diff --git a/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md new file mode 100644 index 0000000000..01fff95892 --- /dev/null +++ b/.agents/notes/implemented/bug-fix/2026-08-06-list-agents-residency-vocabulary.md @@ -0,0 +1,40 @@ +# Agent Note: `list_agents` uses `ready` for resumable children + +Status: implemented + +English | [中文](2026-08-06-list-agents-residency-vocabulary.zh.md) + +## Problem + +`list_agents` projected a continuable child's process residency as `running | idle | complete`. `complete` reads as a terminal unit of work with a result somewhere, but the underlying fact says only that no Activation is resident: the conversation is intact, `send_message` can continue it, and nothing about the child's outcome is being claimed. A model that reads `complete` reasonably looks for a result to collect or sends replacement work to a conversation it believes has ended. + +The word is especially misleading alongside [manager-owned settlement delivery](../feature/2026-08-06-manager-owned-subagent-settlement-delivery.md). Completion reaches the parent as a notice; listing exists to recall durable conversations, not to poll for that notice. + +## Decision + +The model-facing projection reports `running | idle | ready`: + +- **`running`** means the resident Agent has an active driver. +- **`idle`** means the Agent is resident between turns and may be waiting on agents it started. +- **`ready`** means only the durable conversation remains. `send_message` starts the next turn on the same conversation; the status is resumable rather than terminal and does not mean a result is waiting to be collected. + +The tool description states those distinctions and directs the model away from polling: it says the parent is told when a child finishes and that listing is for recalling which children it started. `send_message` remains the authoritative delivery check because either snapshot may race another process or a later message. + +The service layer is unchanged. `SubagentListEntry.activity` retains `'running' | 'inactive'`, which accurately describes corpus residency for consumers such as a UI. The model-facing adapter maps `inactive` to `ready` because that word communicates the action available to the model without inventing an outcome. + +## Alternatives considered + +**Keep `complete` and qualify it in the description.** A description saying that `complete` does not mean complete fights the rendered status on every read. The line the model scans must carry the correct distinction itself. + +**Use `active | dormant`.** This removes the useful distinction between a resident Agent that is between turns and a storage-only conversation, and makes the storage-only state sound unavailable. `ready` states the useful fact: the same conversation accepts another turn. + +**Drop the status entirely.** Residency remains useful when a parent decides whether to send more work. Removing it trades one misleading status for no signal. + +**Rename the service activity values.** `running | inactive` is correct at the service layer and has non-model consumers. Renaming it would churn a general contract to fix one adapter's presentation; the [durable catalog note](../feature/2026-07-22-durable-subagent-catalog-and-list-agents.md) continues to own that service vocabulary. + +## Consequences + +- The rendered line uses ` [running] —