diff --git a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.i18n.yaml index cca3debb14..0b27ea5d9f 100644 --- a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.i18n.yaml +++ b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.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/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md -2026-08-10-minimal-preset-owns-rl-composition.md: d83083de74f1d3203822d82e4e04560819412b48 -2026-08-10-minimal-preset-owns-rl-composition.zh.md: 36aeea0d8af6e371eddd6f09e36bb48a0348959e +2026-08-10-minimal-preset-owns-rl-composition.md: 65d24f9a03eedffac34f0c0141a8e2f643a48b7b +2026-08-10-minimal-preset-owns-rl-composition.zh.md: 72eaa0e5bba46743abd336e64316c886fa58631a diff --git a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md index d83083de74..65d24f9a03 100644 --- a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md +++ b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md @@ -14,13 +14,13 @@ The split also hid other drift. The preset mounted one-shot Bash rather than the The shipped Web `minimal` preset is the sole Web owner of the RL agent composition. It declares an entry-local PTY registry and local backend, persistent `bash` with the RL environment description and 300-second timeout, and `str_replace_editor`. Tool presentation remains a deployment choice. The later [bare two-tool runtime decision](../feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md) supersedes this note's original compaction and filesystem-provider choices: the current preset mounts an entry-local `fs-local` provider and no compaction backend. The editor accepts no `requireAbsolutePath` setting because absolute paths are its unconditional contract. -The preset persona is exactly `You are a helpful software engineer assistant.` and sets `complete: true`. A complete `PromptSection` participates in ordinary assembly so tools, contexts, variables, and cooperative listeners still resolve; after the `system-prompt/assemble` waterfall, the prompt registry restores a detached copy of that section as the sole system-prompt section. Multiple effective complete sections reject assembly. This final registry constraint prevents harness identity, Web orientation, tool guidance, or an assembly listener from appending prompt text. +The preset persona is exactly `You are a helpful software engineer assistant.`, sets `complete: true`, and suppresses runtime context for its agent scope. A complete `PromptSection` participates in ordinary assembly so tools, variables, and cooperative listeners still resolve; after the `system-prompt/assemble` waterfall, the prompt registry restores a detached copy of that section as the sole system-prompt section and discards every dynamic context contribution. Multiple effective complete sections reject assembly. These final registry constraints prevent harness identity, Web orientation, tool guidance, an assembly listener, sandbox policy, approval policy, delegation, or another dynamic context provider from adding model input. The process-wide `core-web.cordis.yml` patch is absent. Browser UI, workspace attachment, persistence, subprocess, sandbox, permission, model routing, and other cross-session services remain host-owned. Selecting `minimal` changes one agent's model-facing composition and shadows the host filesystem provider only for that agent, without changing other sessions in the Web process. ## Verification -System-prompt and persona package tests prove final complete-section enforcement, including waterfall mutation and duplicate rejection. The shipped-preset composition test asserts the exact prompt, Bash description, absolute editor schema, and two-tool catalog under the default native presentation. The keyless Web replay sends a real request through a `minimal` agent while global identity, Web-orientation text, and a test section are registered, asserts the entry-local filesystem is bare and compaction is absent, then executes two persistent Bash calls to prove environment and cwd state survive and executes the editor through an absolute path. +System-prompt and persona package tests prove final complete-section and runtime-context suppression, including waterfall mutation and duplicate rejection. The shipped-preset composition test asserts the exact prompt, Bash description, absolute editor schema, and two-tool catalog under the default native presentation. The keyless Web replay sends a real request through a `minimal` agent while global identity, Web-orientation text, dynamic policy contexts, and a test section are registered, asserts that no runtime-context snapshot exists, the entry-local filesystem is bare, and compaction is absent, then executes two persistent Bash calls to prove environment and cwd state survive and executes the editor through an absolute path. The standalone [`minimal.cordis.yml`](../../../../examples/jsonrpc-agent/minimal.cordis.yml) is the complete two-tool composition for the bundled JSON-RPC runtime. The [bare two-tool runtime decision](../feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md) owns its launch-specific environment configuration, bare filesystem, and absence of compaction. Its keyless SDK replay asserts the assembled system prompt and two-tool catalog, executes persistent Bash across calls, and exercises the editor; the Python SDK tutorial provides the runnable entry point. diff --git a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.zh.md b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.zh.md index 36aeea0d8a..72eaa0e5bb 100644 --- a/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.zh.md +++ b/.agents/notes/implemented/bug-fix/2026-08-10-minimal-preset-owns-rl-composition.zh.md @@ -14,13 +14,13 @@ Status: implemented 随附的 Web `minimal` preset 是 RL agent 组合在 Web 中的唯一所有者。它声明 entry 本地的 PTY 注册表与本地后端、带 RL 环境描述且超时为 300 秒的持久 `bash`，以及 `str_replace_editor`。工具呈现仍由部署选择。后续的[裸双工具运行时决策](../feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md)取代了本记录最初的压缩与文件系统提供方选择：当前 preset 挂载 entry 本地的 `fs-local` 提供方，不挂载压缩后端。编辑器不接受 `requireAbsolutePath` 设置，因为要求绝对路径是它的无条件约定。 -preset persona 恰好是 `You are a helpful software engineer assistant.`，并设置 `complete: true`。complete `PromptSection` 参与常规组装，因此工具、上下文、变量和协作式监听器仍会解析；`system-prompt/assemble` waterfall（瀑布式事件）结束后，提示词注册表会将该段落的独立副本恢复为唯一的系统提示词段落。存在多个有效 complete 段时，组装会被拒绝。这项最终注册表约束可防止 harness 身份、Web 定位、工具引导或组装监听器追加提示词文本。 +preset persona 恰好是 `You are a helpful software engineer assistant.`，它设置 `complete: true`，并为其 agent 作用域抑制 runtime context。complete `PromptSection` 参与常规组装，因此工具、变量和协作式监听器仍会解析；`system-prompt/assemble` waterfall（瀑布式事件）结束后，提示词注册表会将该段落的独立副本恢复为唯一的系统提示词段落，并丢弃每个动态上下文贡献。存在多个有效 complete 段时，组装会被拒绝。这些最终注册表约束可防止 harness 身份、Web 定位、工具引导、组装监听器、沙箱策略、批准策略、委派或其他动态上下文提供方添加模型输入。 进程级 `core-web.cordis.yml` patch 不再存在。浏览器 UI、workspace 附加、持久化、子进程、沙箱、权限、模型路由及其他跨会话服务仍由宿主持有。选择 `minimal` 会改变一个 agent 面向模型的组合，并且仅为该 agent 遮蔽宿主文件系统提供方，不会改变 Web 进程中的其他会话。 ## 验证 -系统提示词与 persona 包测试证明了 complete 段的最终约束，包括 waterfall 修改与重复项拒绝。交付 preset 组合测试在默认原生呈现下断言精确的提示词、Bash 描述、要求绝对路径的编辑器 schema 和双工具目录。无密钥 Web 回放通过 `minimal` agent 发送一个真实请求，同时注册全局身份、Web 定位文本和一个测试段落；它断言 entry 本地文件系统是裸后端且压缩不存在，随后执行两次持久 Bash 调用，证明环境与 cwd 状态能够保留，并通过绝对路径执行编辑器。 +系统提示词与 persona 包测试证明了 complete 段最终约束与 runtime-context 抑制，包括 waterfall 修改与重复项拒绝。交付 preset 组合测试在默认原生呈现下断言精确的提示词、Bash 描述、要求绝对路径的编辑器 schema 和双工具目录。无密钥 Web 回放通过 `minimal` agent 发送一个真实请求，同时注册全局身份、Web 定位文本、动态策略上下文和一个测试段落；它断言不存在 runtime-context 快照、entry 本地文件系统是裸后端且压缩不存在，随后执行两次持久 Bash 调用，证明环境与 cwd 状态能够保留，并通过绝对路径执行编辑器。 独立的 [`minimal.cordis.yml`](../../../../examples/jsonrpc-agent/minimal.cordis.yml) 是内置 JSON-RPC 运行时的完整双工具组合。[裸双工具运行时决策](../feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md)说明其启动方式专属的环境配置、裸文件系统和无压缩选择。其无密钥 SDK 回放会断言组装后的系统提示词与双工具目录，跨调用执行持久 Bash，并使用编辑器；Python SDK 教程提供可运行的入口。 diff --git a/.agents/notes/implemented/feature/2026-06-15-code-mode.i18n.yaml b/.agents/notes/implemented/feature/2026-06-15-code-mode.i18n.yaml index 701a26bf10..da4b80d9d5 100644 --- a/.agents/notes/implemented/feature/2026-06-15-code-mode.i18n.yaml +++ b/.agents/notes/implemented/feature/2026-06-15-code-mode.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/2026-06-15-code-mode.md -2026-06-15-code-mode.md: bc019c0b9acc6d834da41ceff2b1d1930fb34027 -2026-06-15-code-mode.zh.md: c19562b828089b6deeb048d30a08ce8681f150da +2026-06-15-code-mode.md: adabf80ab03bdae57bba84957cdacf648c16c2f6 +2026-06-15-code-mode.zh.md: 24d83b9f896a73b449852d28c3e0a376fb6ace5a diff --git a/.agents/notes/implemented/feature/2026-06-15-code-mode.md b/.agents/notes/implemented/feature/2026-06-15-code-mode.md index bc019c0b9a..adabf80ab0 100644 --- a/.agents/notes/implemented/feature/2026-06-15-code-mode.md +++ b/.agents/notes/implemented/feature/2026-06-15-code-mode.md @@ -87,6 +87,8 @@ The worker runtime provides containment, not a security boundary: model code can The SDK instructs the model to write an async body in the loaded runtime's language (an erasable-TypeScript body by default; a Python `async` body under a Python runtime — see the [language-dispatch note](2026-07-31-code-mode-language-dispatch.md)), call tools through `await tools.name(args)`, catch rejected tool calls when needed, and return or log only the output that should re-enter context. Both flavors state the same contract in their own primitive: independent read-only calls MAY overlap under `Promise.all` (TypeScript) or `asyncio.gather` (Python), mutating calls run alone in submission order, and dependent work sequences with `await`. The declaration prefix can be as large as native schemas, especially in `'both'`, but remains stable for provider caching. +The transport's own `description` and both SDK instruction flavors open by naming `code` and `description` as the call's two required arguments. Prose that describes the call as passing a program leaves the second argument discoverable only through the parameter schema, and a model that emits `{code}` alone loses the whole written program to an `INVALID_ARGS` rejection. + ## Consequences Deployments switching to `'code'` must update any native-only `toolOrder`. Assembly listeners own the integrity of any rewritten protocol messages. Sub-dispatch starts in submission order under a bounded overlap pool, while per-call contexts retain their source, envelope, and metadata through the outer result. diff --git a/.agents/notes/implemented/feature/2026-06-15-code-mode.zh.md b/.agents/notes/implemented/feature/2026-06-15-code-mode.zh.md index c19562b828..24d83b9f89 100644 --- a/.agents/notes/implemented/feature/2026-06-15-code-mode.zh.md +++ b/.agents/notes/implemented/feature/2026-06-15-code-mode.zh.md @@ -87,6 +87,8 @@ worker 运行时只能约束程序的运行，而不构成安全边界：模型 SDK 指示模型编写一个所加载运行时语言的异步函数体（默认可擦除 TypeScript；Python 运行时下为 Python `async` 函数体——见[语言分发 note](2026-07-31-code-mode-language-dispatch.md)），通过 `await tools.name(args)` 调用工具，在需要时捕获被拒绝的工具调用，并仅 return 或 log 应重新进入上下文的输出。两种 flavor 用各自的原语陈述同一约定：相互独立的只读调用可以（MAY）在 `Promise.all`（TypeScript）或 `asyncio.gather`（Python）下重叠，有副作用的调用按提交顺序单独运行，有依赖的工作用 `await` 排序。声明前缀可能与原生 schema 一样大，尤其在 `'both'` 下，但对提供方缓存保持稳定。 +传输自身的 `description` 与两种 flavor 的 SDK 说明都以点名 `code` 和 `description` 这两个必填参数开头。把该调用描述成「传入一个程序」的散文会让第二个参数只能从参数 schema 中发现，而只发出 `{code}` 的模型会因 `INVALID_ARGS` 被拒，连同已写好的整个程序一起丢失。 + ## 后果 切换到 `'code'` 的部署必须更新任何仅限 native 的 `toolOrder`。组装监听器有责任维护任何被重写的协议消息的完整性。子分发在有界的重叠池下按提交顺序启动，而每次调用的上下文会通过外层结果保留其 source、信封与元数据。 diff --git a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.i18n.yaml b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.i18n.yaml index 5455415389..695a161749 100644 --- a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.i18n.yaml +++ b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.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-current-sandbox-policy-context.md -2026-07-30-current-sandbox-policy-context.md: ab231d19aec4852369c0f1a13fc50c69ab8128f1 -2026-07-30-current-sandbox-policy-context.zh.md: 4ac493dcfff5c286d569a46c34686dc4b8746618 +2026-07-30-current-sandbox-policy-context.md: b4c657ba52ff26e2e295e3635201424a155738d0 +2026-07-30-current-sandbox-policy-context.zh.md: bd32b021ef8622dc6843a163b21714a263bc8bed diff --git a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.md b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.md index ab231d19ae..b4c657ba52 100644 --- a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.md +++ b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.md @@ -12,7 +12,7 @@ The sandbox policy already enforced and logged each session's file-effect mode, `dsh-sandbox-policy`, the owner of mode and workspace-root resolution, registers one `sandbox:policy` cache-safe context contribution. Every agent request resolves the active session directly through `ctx.sandboxPolicy.resolve({ session })`; there is no denial-history scan or process-local “last told” state. -The policy contribution is capability-neutral and present for every agent session. It does not maintain a second inventory of mounted backends or tools; model-visible schemas remain the authority for available operations, while the context conditions its claims on any available operation that the DSH file sandbox enforces. The [capability-neutral policy context decision](../simplification/2026-07-31-capability-neutral-sandbox-policy-context.md) supersedes the earlier family-registration mechanism while retaining this note's cache-safe delivery and durable snapshot design. +The policy contribution is capability-neutral and available to every agent session by default. A composition may suppress the complete runtime-context channel when its model interface intentionally excludes dynamic context; this does not disable policy enforcement. The contribution does not maintain a second inventory of mounted backends or tools; model-visible schemas remain the authority for available operations, while the context conditions its claims on any available operation that the DSH file sandbox enforces. The [capability-neutral policy context decision](../simplification/2026-07-31-capability-neutral-sandbox-policy-context.md) supersedes the earlier family-registration mechanism while retaining this note's cache-safe delivery and durable snapshot design. The contribution states only facts shared by every enforcement dialect. `read-only` says an available sandbox-enforced operation cannot modify files in the standing mode and directs the model to try an available tool normally, then follow any denial and escalation guidance that tool returns. `workspace-write` states the canonical session workspace with non-exclusive wording and summarizes, without enumerating, that some platform temporary areas may also be writable. `danger-full-access` says the DSH file sandbox does not restrict file modifications by available operations. Backend-selected temporary paths, `/dev/null`, runner readiness, exact tool availability, and other policy domains are absent because `resolve()` cannot establish them at request assembly. diff --git a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.zh.md b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.zh.md index 4ac493dcff..bd32b021ef 100644 --- a/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.zh.md +++ b/.agents/notes/implemented/feature/2026-07-30-current-sandbox-policy-context.zh.md @@ -12,7 +12,7 @@ Status: implemented `dsh-sandbox-policy` 负责解析模式与工作区根目录，并注册一项缓存安全的 `sandbox:policy` 上下文贡献。每次 agent（智能体）请求都通过 `ctx.sandboxPolicy.resolve({ session })` 直接解析当前会话；不存在拒绝历史扫描或进程本地的「上次告知」状态。 -该策略贡献不依赖具体能力，并存在于每个 agent 会话中。它不会另行维护一份已挂载后端或工具清单；模型可见的 schema 仍是可用操作的权威来源，而上下文会将其声明限定在 DSH 文件沙箱所强制执行的任何可用操作上。[不依赖具体能力的策略上下文决策](../simplification/2026-07-31-capability-neutral-sandbox-policy-context.md)取代了较早的家族注册机制，同时保留本 Agent Note 的缓存安全交付与持久快照设计。 +该策略贡献不依赖具体能力，并默认可用于每个 agent 会话。当模型接口有意排除动态上下文时，组合可以抑制完整的 runtime-context 通道；这不会禁用策略强制机制。该贡献不会另行维护一份已挂载后端或工具清单；模型可见的 schema 仍是可用操作的权威来源，而上下文会将其声明限定在 DSH 文件沙箱所强制执行的任何可用操作上。[不依赖具体能力的策略上下文决策](../simplification/2026-07-31-capability-neutral-sandbox-policy-context.md)取代了较早的家族注册机制，同时保留本 Agent Note 的缓存安全交付与持久快照设计。 该贡献只说明所有强制执行方言所共有的事实。`read-only` 表明受沙箱强制执行的可用操作在常驻模式下无法修改文件，并指示模型正常尝试可用工具，随后遵循该工具返回的任何拒绝与升权引导。`workspace-write` 用非排他措辞说明规范化的会话工作区，并概述某些平台临时区域可能也可写，而不逐一列举。`danger-full-access` 表明 DSH 文件沙箱不会限制可用操作修改文件。后端选择的临时路径、`/dev/null`、runner 就绪状态、确切的工具可用情况和其他策略领域都不会出现，因为 `resolve()` 无法在请求组装时确定它们。 diff --git a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.i18n.yaml b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.i18n.yaml index 8133a3f08a..2907d9c241 100644 --- a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.i18n.yaml +++ b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.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/2026-08-11-minimal-profiles-bare-two-tool-runtime.md -2026-08-11-minimal-profiles-bare-two-tool-runtime.md: 92d00a77de3e378abccc2ea67ba38a0cc76e0eca -2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md: 1528b241436171145839240c7415ef18acab337f +2026-08-11-minimal-profiles-bare-two-tool-runtime.md: 2f26dda5ddc905076dbc2f6c681f462973bde793 +2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md: ffdd8f0ddb86535468dbcb4384a6ba18c97b7124 diff --git a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md index 92d00a77de..2f26dda5dd 100644 --- a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md +++ b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.md @@ -12,7 +12,7 @@ The two launch paths also have different configuration owners. Web mounts a per- ## Decision -Both shipped minimal profiles expose exactly persistent `bash` and `str_replace_editor`, mount no context-compaction provider, and run the editor against `@deepseek-ai/dsh-fs-local`. The Web preset isolates `ctx.fs` inside the agent entry and mounts `fs-local` beside the editor, so other Web agents retain the host filesystem provider. Its persona remains the fixed complete prompt owned by the earlier [minimal-preset composition decision](../bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md). +Both shipped minimal profiles expose exactly persistent `bash` and `str_replace_editor`, mount no context-compaction provider, suppress every `dsh-system-prompt` runtime-context contribution for fresh sessions, and run the editor against `@deepseek-ai/dsh-fs-local`. The Web preset isolates `ctx.fs` inside the agent entry and mounts `fs-local` beside the editor, so other Web agents retain the host filesystem provider. Its persona remains the fixed complete prompt owned by the earlier [minimal-preset composition decision](../bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md) and applies runtime-context suppression only to that agent scope. The standalone spine forwards the same setting to its process-owned system-prompt service. Sandbox and approval services remain mounted and enforce their policies; only their model-facing dynamic context is absent. The standalone [`minimal.cordis.yml`](../../../../examples/jsonrpc-agent/minimal.cordis.yml) remains a complete JSON-RPC process composition. It mounts `dsh-sdk-jsonrpc-server`, the local PTY and subprocess services required by persistent Bash, `fs-local`, the two tool consumers, and uncompressed JSONL persistence. It does not mount `token-meter`, `compaction-basic`, `fs-sandbox`, or `fs-observation-policy`. Persistent Bash still consumes the deployment's danger-full-access sandbox policy; the editor is not confined by that policy. @@ -20,9 +20,9 @@ The standalone [`minimal.cordis.yml`](../../../../examples/jsonrpc-agent/minimal ## Verification -The Web replay boots the complete Web host, creates the agent through the preset service, and asserts that the scoped filesystem is bare, no scoped compaction service exists, and the assembled request contains exactly the fixed prompt and two tools. It then executes persistent Bash and the editor against the real scoped services. +The Web replay boots the complete Web host, creates the agent through the preset service, and asserts that the scoped filesystem is bare, no scoped compaction service exists, no system-prompt-owned runtime-context message was appended, and the assembled request contains exactly the fixed prompt and two tools. It then executes persistent Bash and the editor against the real scoped services. -The SDK replay boots the real JSON-RPC agent process through the SDK client, injects an environment-selected prompt, asserts the assembled prompt and exact two-tool catalog, and executes both tools. Python SDK bundled-runtime coverage initializes the standalone configuration through each available packaged carrier with environment-selected model, model capacity, and prompt values. Cordis validation checks that both configurations resolve their declared plugins and configuration fields. +The SDK replay boots the real JSON-RPC agent process through the SDK client, injects an environment-selected prompt, asserts the assembled prompt, exact two-tool catalog, and absence of every system-prompt-owned runtime-context message, and executes both tools. Python SDK bundled-runtime coverage initializes the standalone configuration through each available packaged carrier with environment-selected model, model capacity, and prompt values. Cordis validation checks that both configurations resolve their declared plugins and configuration fields. ## Alternatives considered @@ -36,4 +36,4 @@ The SDK replay boots the real JSON-RPC agent process through the SDK client, inj ## Consequences -Minimal sessions never summarize or replace earlier history; callers must keep turns within the selected model's context capacity. The editor can address any absolute path visible to the runtime process, independently of the persistent shell's sandbox policy. The two launch paths share their model-facing tool and no-compaction guarantees while retaining different prompt and model configuration appropriate to their owners. The Python SDK path continues to communicate only through the bundled stdio JSON-RPC runtime. +Minimal sessions never summarize or replace earlier history and never add a runtime-context snapshot; callers must keep turns within the selected model's context capacity and must not rely on model-visible narration of standing sandbox or approval policy. The editor can address any absolute path visible to the runtime process, independently of the persistent shell's sandbox policy. The two launch paths share their model-facing tool, no-context, and no-compaction guarantees while retaining different prompt and model configuration appropriate to their owners. The Python SDK path continues to communicate only through the bundled stdio JSON-RPC runtime. diff --git a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md index 1528b24143..ffdd8f0ddb 100644 --- a/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md +++ b/.agents/notes/implemented/feature/2026-08-11-minimal-profiles-bare-two-tool-runtime.zh.md @@ -12,7 +12,7 @@ Web `minimal` preset 与独立 JSON-RPC minimal 组合对外提供持久 `bash` ## 决策 -两种随附 minimal profile 都只对外提供持久 `bash` 与 `str_replace_editor`，不挂载上下文压缩提供方，并让编辑器使用 `@deepseek-ai/dsh-fs-local`。Web preset 在 agent entry 内隔离 `ctx.fs`，将 `fs-local` 与编辑器一起挂载，因此其他 Web agent 仍使用宿主文件系统提供方。其 persona 继续采用较早的 [minimal preset 组合决策](../bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md)所拥有的固定 complete 提示词。 +两种随附 minimal profile 都只对外提供持久 `bash` 与 `str_replace_editor`，不挂载上下文压缩提供方，为新建会话抑制每个 `dsh-system-prompt` runtime-context 贡献，并让编辑器使用 `@deepseek-ai/dsh-fs-local`。Web preset 在 agent entry 内隔离 `ctx.fs`，将 `fs-local` 与编辑器一起挂载，因此其他 Web agent 仍使用宿主文件系统提供方。其 persona 继续采用较早的 [minimal preset 组合决策](../bug-fix/2026-08-10-minimal-preset-owns-rl-composition.md)所拥有的固定 complete 提示词，并仅为该 agent 作用域实施 runtime-context 抑制。独立 spine 将同一设置转发给其进程拥有的 system-prompt 服务。沙箱与批准服务仍保持挂载并强制其策略；只有它们面向模型的动态上下文缺席。 独立的 [`minimal.cordis.yml`](../../../../examples/jsonrpc-agent/minimal.cordis.yml) 仍是完整的 JSON-RPC 进程组合。它挂载 `dsh-sdk-jsonrpc-server`、持久 Bash 所需的本地 PTY 和子进程服务、`fs-local`、两个工具消费方，以及未压缩的 JSONL 持久化。它不挂载 `token-meter`、`compaction-basic`、`fs-sandbox` 或 `fs-observation-policy`。持久 Bash 仍消费部署的 danger-full-access 沙箱策略；编辑器不受该策略限制。 @@ -20,9 +20,9 @@ Web `minimal` preset 与独立 JSON-RPC minimal 组合对外提供持久 `bash` ## 验证 -Web 回放会启动完整 Web 宿主，通过 preset 服务创建 agent，并断言作用域文件系统为裸后端、不存在作用域压缩服务，而且组装请求只包含固定提示词与两个工具。随后，它通过真实作用域服务执行持久 Bash 和编辑器。 +Web 回放会启动完整 Web 宿主，通过 preset 服务创建 agent，并断言作用域文件系统为裸后端、不存在作用域压缩服务、没有追加 system-prompt 拥有的 runtime-context 消息，而且组装请求只包含固定提示词与两个工具。随后，它通过真实作用域服务执行持久 Bash 和编辑器。 -SDK 回放通过 SDK 客户端启动真实 JSON-RPC agent 进程，注入由环境选择的提示词，断言组装提示词与精确双工具目录，并执行两个工具。Python SDK 内置运行时覆盖会通过每种可用的打包载体，使用环境选择的模型、模型容量和提示词值初始化独立配置。Cordis 校验会检查两份配置能否解析声明的插件和配置字段。 +SDK 回放通过 SDK 客户端启动真实 JSON-RPC agent 进程，注入由环境选择的提示词，断言组装提示词与精确双工具目录，另外断言不存在任何 system-prompt 拥有的 runtime-context 消息，并执行两个工具。Python SDK 内置运行时覆盖会通过每种可用的打包载体，使用环境选择的模型、模型容量和提示词值初始化独立配置。Cordis 校验会检查两份配置能否解析声明的插件和配置字段。 ## 考虑过的替代方案 @@ -36,4 +36,4 @@ SDK 回放通过 SDK 客户端启动真实 JSON-RPC agent 进程，注入由环 ## 后果 -Minimal 会话不会摘要或替换较早历史；调用方必须让会话轮次保持在所选模型的上下文容量内。编辑器可以访问运行时进程可见的任何绝对路径，且不受持久 shell 沙箱策略影响。两条启动路径共享面向模型的工具与无压缩保证，同时保留适合各自所有者的不同提示词和模型配置。Python SDK 路径继续仅通过内置 stdio JSON-RPC 运行时通信。 +Minimal 会话不会摘要或替换较早历史，也不会添加 runtime-context 快照；调用方必须让会话轮次保持在所选模型的上下文容量内，且不得依赖模型可见的常驻沙箱或批准策略说明。编辑器可以访问运行时进程可见的任何绝对路径，且不受持久 shell 沙箱策略影响。两条启动路径共享面向模型的工具、无上下文与无压缩保证，同时保留适合各自所有者的不同提示词和模型配置。Python SDK 路径继续仅通过内置 stdio JSON-RPC 运行时通信。 diff --git a/apps/cli/config/agent-presets/minimal/agent.cordis.yml b/apps/cli/config/agent-presets/minimal/agent.cordis.yml index 1a2336b222..b6ed406f54 100644 --- a/apps/cli/config/agent-presets/minimal/agent.cordis.yml +++ b/apps/cli/config/agent-presets/minimal/agent.cordis.yml @@ -1,15 +1,16 @@ # The `minimal` agent preset: a fixed-prompt, two-tool coding-agent composition. # # The persona is the complete system prompt, so global identity, Web orientation, -# tool guidance, and later assembly listeners cannot add prompt text. The model -# composes only the persistent `bash` and `str_replace_editor` tools. Context -# compaction is deliberately absent. +# tool guidance, and later assembly listeners cannot add prompt text. Runtime +# context snapshots are suppressed for this preset, and the model composes only +# persistent `bash` and `str_replace_editor`. Context compaction is absent. - id: persona name: '@deepseek-ai/dsh-persona' config: text: You are a helpful software engineer assistant. complete: true + includeRuntimeContext: false # The PTY registry is an agent-owned service, so it lives in an entry-local # realm. The backend still consumes the host sandbox policy and subprocess diff --git a/apps/web/tests/minimal-preset.snapshot.ts b/apps/web/tests/minimal-preset.snapshot.ts index 4a402077a3..011f71ea8d 100644 --- a/apps/web/tests/minimal-preset.snapshot.ts +++ b/apps/web/tests/minimal-preset.snapshot.ts @@ -55,6 +55,9 @@ describe('minimal agent preset', () => { const requestHeader = agentHandle.agent.session.requestHeader() if (requestHeader === undefined) throw new Error('the minimal agent issued no model request') + expect(agentHandle.agent.session.events.some(event => event.type === 'user/message' + && event.data.source.kind === 'plugin' + && event.data.source.plugin === '@deepseek-ai/dsh-system-prompt')).toBe(false) const presetFileSystem = scaffold.ctx.agentPresets.serviceFor(agentHandle.agent, 'fs') expect(presetFileSystem).toBeDefined() expect(presetFileSystem?.sandboxMode).toBeUndefined() diff --git a/docs/config-catalog.i18n.yaml b/docs/config-catalog.i18n.yaml index 96397ace10..7e94316a6a 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: 09c6064c2c6bb3c1da6283a885db33a5b4a674df -config-catalog.zh.md: 0bd95a5ab24b7acf85ba7ca00125ea9135e7dd82 +config-catalog.md: 6b60fad7395ba4de28921390641a332ea9667217 +config-catalog.zh.md: b4378b32b925adce221ca0bac837a6630eb30a24 diff --git a/docs/config-catalog.md b/docs/config-catalog.md index 09c6064c2c..6b60fad739 100644 --- a/docs/config-catalog.md +++ b/docs/config-catalog.md @@ -196,9 +196,10 @@ Source: [`packages/preset/agent-presets/src/preset.ts:52`](../packages/preset/ag /** * Bundle config: each field forwarded verbatim to the child that owns it — * `agents` to the agent loop (an app that pre-creates no agents, like the ACP - * bridge, simply omits it), `includeHarnessIdentity`, `persona`, and `toolOrder` - * to the system-prompt plugin (the fixed opener, deployment persona, and explicit - * model-facing tool order), the `tools` object to the tool registry (its presentation `mode`), + * bridge, simply omits it), `includeHarnessIdentity`, `includeRuntimeContext`, + * `persona`, and `toolOrder` to the system-prompt plugin (the fixed opener, + * dynamic-context policy, deployment persona, and explicit model-facing tool + * order), the `tools` object to the tool registry (its presentation `mode`), * `dshHome` to bash environment and local skill discovery, `sessionTitle` to * the fallback title service, `skills` to the * skill registry/local provider/tool consumer, `workspaceContext` to the @@ -222,6 +223,8 @@ export interface Config { maxParallelToolCalls?: AgentLoopConfig['maxParallelToolCalls'] /** Whether the system prompt includes the fixed Harness identity (default true). */ includeHarnessIdentity?: SystemPromptConfig['includeHarnessIdentity'] + /** Whether model history includes dynamic runtime-context snapshots (default true). */ + includeRuntimeContext?: SystemPromptConfig['includeRuntimeContext'] /** The deployment persona (see dsh-system-prompt's `Config`). */ persona?: SystemPromptConfig['persona'] /** The explicit model-facing tool order (see dsh-system-prompt's `Config`). */ @@ -275,7 +278,7 @@ export interface GoalConfig { Depends on: [`AgentLoopConfig`](#deepseek-aidsh-agent-loop) · [`GoalDomainConfig`](#deepseek-aidsh-goal) · [`InvariantConfig`](#deepseek-aidsh-invariants) · [`JobsConfig`](#deepseek-aidsh-jobs-local) · [`SessionTitleConfig`](#deepseek-aidsh-session-title) · [`SkillFileSystem`](../packages/skill/skill-filesystem/src/index.ts) · [`SkillRegistryConfig`](#deepseek-aidsh-skill) · [`SystemPromptConfig`](#deepseek-aidsh-system-prompt) · [`toolBash`](../packages/shell/tool-bash/src/index.ts) · [`toolGoal`](../packages/goal/tool-goal/src/index.ts) · [`toolJobs`](../packages/jobs/tool-jobs/src/index.ts) · [`ToolsConfig`](#deepseek-aidsh-tools) · [`toolSkill`](../packages/skill/tool-skill/src/index.ts) · [`workspaceContext`](../packages/context/agent-instructions/src/index.ts) -Source: [`packages/examples/agent-spine-demo/src/index.ts:91`](../packages/examples/agent-spine-demo/src/index.ts) +Source: [`packages/examples/agent-spine-demo/src/index.ts:92`](../packages/examples/agent-spine-demo/src/index.ts) ## `@deepseek-ai/dsh-agent-tool-presentation` @@ -1260,6 +1263,8 @@ export interface Config { text: string /** Make this persona the complete system prompt, suppressing every other section. */ complete?: boolean + /** Suppress dynamic runtime-context snapshots for this persona's agent scope. */ + includeRuntimeContext?: boolean } ``` @@ -2063,6 +2068,8 @@ Source: [`packages/e2b/subprocess-e2b/src/index.ts:25`](../packages/e2b/subproce export interface Config { /** Include the fixed DeepSeek Harness identity before the deployment persona (default true). */ includeHarnessIdentity?: boolean + /** Include dynamic runtime-context snapshots in model history (default true). */ + includeRuntimeContext?: boolean /** * Deployment-wide order-0 persona template. A scoped section named * `deployment:persona` shadows it; `{{variable}}` references are strict. diff --git a/docs/config-catalog.zh.md b/docs/config-catalog.zh.md index 0bd95a5ab2..b4378b32b9 100644 --- a/docs/config-catalog.zh.md +++ b/docs/config-catalog.zh.md @@ -198,9 +198,10 @@ export type PresetTrust = 'system' | 'user' /** * Bundle config: each field forwarded verbatim to the child that owns it — * `agents` to the agent loop (an app that pre-creates no agents, like the ACP - * bridge, simply omits it), `includeHarnessIdentity`, `persona`, and `toolOrder` - * to the system-prompt plugin (the fixed opener, deployment persona, and explicit - * model-facing tool order), the `tools` object to the tool registry (its presentation `mode`), + * bridge, simply omits it), `includeHarnessIdentity`, `includeRuntimeContext`, + * `persona`, and `toolOrder` to the system-prompt plugin (the fixed opener, + * dynamic-context policy, deployment persona, and explicit model-facing tool + * order), the `tools` object to the tool registry (its presentation `mode`), * `dshHome` to bash environment and local skill discovery, `sessionTitle` to * the fallback title service, `skills` to the * skill registry/local provider/tool consumer, `workspaceContext` to the @@ -224,6 +225,8 @@ export interface Config { maxParallelToolCalls?: AgentLoopConfig['maxParallelToolCalls'] /** Whether the system prompt includes the fixed Harness identity (default true). */ includeHarnessIdentity?: SystemPromptConfig['includeHarnessIdentity'] + /** Whether model history includes dynamic runtime-context snapshots (default true). */ + includeRuntimeContext?: SystemPromptConfig['includeRuntimeContext'] /** The deployment persona (see dsh-system-prompt's `Config`). */ persona?: SystemPromptConfig['persona'] /** The explicit model-facing tool order (see dsh-system-prompt's `Config`). */ @@ -277,7 +280,7 @@ export interface GoalConfig { 依赖：[`AgentLoopConfig`](#deepseek-aidsh-agent-loop) · [`GoalDomainConfig`](#deepseek-aidsh-goal) · [`InvariantConfig`](#deepseek-aidsh-invariants) · [`JobsConfig`](#deepseek-aidsh-jobs-local) · [`SessionTitleConfig`](#deepseek-aidsh-session-title) · [`SkillFileSystem`](../packages/skill/skill-filesystem/src/index.ts) · [`SkillRegistryConfig`](#deepseek-aidsh-skill) · [`SystemPromptConfig`](#deepseek-aidsh-system-prompt) · [`toolBash`](../packages/shell/tool-bash/src/index.ts) · [`toolGoal`](../packages/goal/tool-goal/src/index.ts) · [`toolJobs`](../packages/jobs/tool-jobs/src/index.ts) · [`ToolsConfig`](#deepseek-aidsh-tools) · [`toolSkill`](../packages/skill/tool-skill/src/index.ts) · [`workspaceContext`](../packages/context/agent-instructions/src/index.ts) -来源：[`packages/examples/agent-spine-demo/src/index.ts:91`](../packages/examples/agent-spine-demo/src/index.ts) +来源：[`packages/examples/agent-spine-demo/src/index.ts:92`](../packages/examples/agent-spine-demo/src/index.ts) ## `@deepseek-ai/dsh-agent-tool-presentation` @@ -1262,6 +1265,8 @@ export interface Config { text: string /** Make this persona the complete system prompt, suppressing every other section. */ complete?: boolean + /** Suppress dynamic runtime-context snapshots for this persona's agent scope. */ + includeRuntimeContext?: boolean } ``` @@ -2065,6 +2070,8 @@ export interface Config { export interface Config { /** Include the fixed DeepSeek Harness identity before the deployment persona (default true). */ includeHarnessIdentity?: boolean + /** Include dynamic runtime-context snapshots in model history (default true). */ + includeRuntimeContext?: boolean /** * Deployment-wide order-0 persona template. A scoped section named * `deployment:persona` shadows it; `{{variable}}` references are strict. diff --git a/docs/subsystems/system-prompt.i18n.yaml b/docs/subsystems/system-prompt.i18n.yaml index 3ac169b4e0..0b69a5ca8c 100644 --- a/docs/subsystems/system-prompt.i18n.yaml +++ b/docs/subsystems/system-prompt.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/system-prompt.md -system-prompt.md: f03d11866bcdcf3bcb5c5fe96931b059cb8d0336 -system-prompt.zh.md: 23b3e0b20cc4772baf22727d2e57866aa7cbbeb0 +system-prompt.md: 8a6006d0b96552b52d0b050c304a11a95add8d9e +system-prompt.zh.md: 02be9eed795d3a6c8e43403048f135d493b6509d diff --git a/docs/subsystems/system-prompt.md b/docs/subsystems/system-prompt.md index f03d11866b..8a6006d0b9 100644 --- a/docs/subsystems/system-prompt.md +++ b/docs/subsystems/system-prompt.md @@ -117,6 +117,14 @@ section(section: PromptSection): () => void */ context(context: PromptContext): () => void +/** + * Suppress every dynamic runtime-context contribution in the calling + * context's scope without changing the services that own or enforce those + * facts. Multiple suppressors remain independently disposable. + * @returns the exact Cordis effect disposer. + */ +suppressRuntimeContext(): () => void + /** * Register a tool-schema provider in the calling context's scope. Global and * matching scoped providers both contribute; returning the reserved @@ -148,7 +156,7 @@ variable(name: string, provider: (context: AssembleContext) => string | undefine async assemble(context: AssembleContext = {}): Promise ``` -Source: [`packages/core/system-prompt/src/index.ts:334`](../../packages/core/system-prompt/src/index.ts) +Source: [`packages/core/system-prompt/src/index.ts:338`](../../packages/core/system-prompt/src/index.ts) diff --git a/docs/subsystems/system-prompt.zh.md b/docs/subsystems/system-prompt.zh.md index 23b3e0b20c..02be9eed79 100644 --- a/docs/subsystems/system-prompt.zh.md +++ b/docs/subsystems/system-prompt.zh.md @@ -117,6 +117,14 @@ section(section: PromptSection): () => void */ context(context: PromptContext): () => void +/** + * Suppress every dynamic runtime-context contribution in the calling + * context's scope without changing the services that own or enforce those + * facts. Multiple suppressors remain independently disposable. + * @returns the exact Cordis effect disposer. + */ +suppressRuntimeContext(): () => void + /** * Register a tool-schema provider in the calling context's scope. Global and * matching scoped providers both contribute; returning the reserved @@ -148,7 +156,7 @@ variable(name: string, provider: (context: AssembleContext) => string | undefine async assemble(context: AssembleContext = {}): Promise ``` -Source: [`packages/core/system-prompt/src/index.ts:334`](../../packages/core/system-prompt/src/index.ts) +Source: [`packages/core/system-prompt/src/index.ts:338`](../../packages/core/system-prompt/src/index.ts) diff --git a/docs/tool-catalog.i18n.yaml b/docs/tool-catalog.i18n.yaml index 9cbc5707a4..a2083a076c 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: ea89508500ec6e73e6e56c4e5d0ac2c35897b342 -tool-catalog.zh.md: 076cc63151e3b9ee8b3828ded65df9ebf09c9430 +tool-catalog.md: 3d73ed1ef5346f620808e9d6b04299291ba1ed4c +tool-catalog.zh.md: 90573928a24630e802d98441bc0e340baf582b23 diff --git a/docs/tool-catalog.md b/docs/tool-catalog.md index ea89508500..3d73ed1ef5 100644 --- a/docs/tool-catalog.md +++ b/docs/tool-catalog.md @@ -116,7 +116,7 @@ ask_user_question pauses the tool call until the active UI provider returns a hu ### `run_code` -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. +Execute a TypeScript program against the available tools. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. Call tools as `await tools.name(args)` per the declarations in the system prompt. Only what you print or return comes back — curate it. ```json { diff --git a/docs/tool-catalog.zh.md b/docs/tool-catalog.zh.md index 076cc63151..90573928a2 100644 --- a/docs/tool-catalog.zh.md +++ b/docs/tool-catalog.zh.md @@ -118,7 +118,7 @@ ask_user_question 会暂停工具调用，直到当前 UI 提供方返回人类 ### `run_code` -针对可用工具执行 TypeScript 程序。请编写异步函数的**函数体**（仅使用可擦除语法；支持顶层 `await` 和 `return`），并根据系统提示词中的声明，以 `await tools.name(args)` 形式调用工具。只有打印或返回的内容会传回，请谨慎筛选。 +针对可用工具执行 TypeScript 程序。接受两个必填参数：`code`，即异步函数的**函数体**（仅使用可擦除语法；支持顶层 `await` 和 `return`）；以及 `description`，简要说明该程序做什么。请根据系统提示词中的声明，以 `await tools.name(args)` 形式调用工具。只有打印或返回的内容会传回，请谨慎筛选。 ```json { 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 5e266ddbc0..1d2682498e 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 @@ -131,7 +131,7 @@ Use subagent in the background by default. Start independent delegations togethe ## Writing code for run_code -Pass `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: +`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. Inside the program: - 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. - 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. 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 eba901f611..aec8cea715 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 @@ -443,7 +443,7 @@ }, { "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.", + "description": "Execute a TypeScript program against the available tools. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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": { diff --git a/examples/acp-agent/tests/snapshots/both-mode-turn/system-prompt.expected.md b/examples/acp-agent/tests/snapshots/both-mode-turn/system-prompt.expected.md index 9ed911462a..57603571d8 100644 --- a/examples/acp-agent/tests/snapshots/both-mode-turn/system-prompt.expected.md +++ b/examples/acp-agent/tests/snapshots/both-mode-turn/system-prompt.expected.md @@ -25,7 +25,7 @@ Use subagent in the background by default. Start independent delegations togethe ## Writing code for run_code -Pass `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: +`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. Inside the program: - 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. - 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. 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 ae72cfca19..c76cd2ef49 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 @@ -246,7 +246,7 @@ }, { "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.", + "description": "Execute a TypeScript program against the available tools. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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": { 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 e0398b0322..0659f8e2a5 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 @@ -27,7 +27,7 @@ Use subagent in the background by default. Start independent delegations togethe ## Writing code for run_code -Pass `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: +`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. Inside the program: - 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. - 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. diff --git a/examples/acp-agent/tests/snapshots/code-mode-turn/tool-schemas.expected.json b/examples/acp-agent/tests/snapshots/code-mode-turn/tool-schemas.expected.json index a9ee29aa7a..2a94d503e2 100644 --- a/examples/acp-agent/tests/snapshots/code-mode-turn/tool-schemas.expected.json +++ b/examples/acp-agent/tests/snapshots/code-mode-turn/tool-schemas.expected.json @@ -2,7 +2,7 @@ "initial": [ { "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.", + "description": "Execute a TypeScript program against the available tools. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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": { 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 dfe9a23a15..8fb937dee2 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":"77f5992c-b7ed-4453-96de-83c51b1e8041"}]}} +{"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":"c66e310e-2597-4d01-85c8-2d70a9d831c0"}]}} {"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":"77f5992c-b7ed-4453-96de-83c51b1e8041"},"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":"16344357-da11-40dc-a87f-b390a2c9686d"},"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":"c66e310e-2597-4d01-85c8-2d70a9d831c0"},"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":"fd0a0587-8df4-46b2-809a-317346a4c0f4"},"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\n\n## Writing code for run_code\n\n`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. 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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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":"ce699f54-ef07-41e3-86e9-ef57995549f0"},"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":"02dd8a61-a39a-46d0-8f6f-457533271cae"},"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 6d26883f39..e05489ba1f 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":"893169a9-49fc-4849-96ce-18bd7132f2e3"}]}} +{"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":"8b3cd23c-82f1-4903-8a3f-b9082059b40c"}]}} {"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":"893169a9-49fc-4849-96ce-18bd7132f2e3"},"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":"61a17dbb-19d2-4053-8251-ff867a552522"},"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":"8b3cd23c-82f1-4903-8a3f-b9082059b40c"},"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":"823e5037-9e96-4ef5-8c5b-cbe73b993ee2"},"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\n\n## Writing code for run_code\n\n`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. 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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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":"9091fda5-4d10-4d7d-8cd2-53aa493c9dd1"},"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":"cd78f077-1fad-4cdc-ab56-09d39d9095cd"},"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 a7183e045c..19e4ec8138 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: define a host-only dynamic Cordis Package named Snapshot Marker; run and inspect snap-1/pkg-1 through run_code; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; remove snap-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"eeafe3ec-26b4-4e44-ae07-96422e3691ac"}]}} +{"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: define a host-only dynamic Cordis Package named Snapshot Marker; run and inspect snap-1/pkg-1 through run_code; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; remove snap-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"8a0ac233-283c-4eb4-8bbd-5c50b7e99afe"}]}} {"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: define a host-only dynamic Cordis Package named Snapshot Marker; run and inspect snap-1/pkg-1 through run_code; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; remove snap-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"eeafe3ec-26b4-4e44-ae07-96422e3691ac"},"surfaceOp":"append"} +{"type":"user/message","seq":4,"time":1785498583779,"data":{"content":[{"type":"text","text":"Run this advanced flow exactly once: define a host-only dynamic Cordis Package named Snapshot Marker; run and inspect snap-1/pkg-1 through run_code; delegate once to a direct spawn child; run one workflow that delegates to another spawn child; remove snap-1; then reply with exactly ADVANCED_HEADLESS_OK."}],"source":{"kind":"user"},"role":"user","id":"8a0ac233-283c-4eb4-8bbd-5c50b7e99afe"},"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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 DeepSeek Harness.\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-observation-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-observation-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 job id you start. You are notified in-session when a job finishes — do not busy-poll or sleep on one; keep working on independent steps and do not duplicate a running job's work. Before giving a final answer, collect every still-relevant job with job_output (set wait: true only when you are genuinely blocked on it), and job_kill jobs 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\n# Dynamic Cordis Plugins\n\nDynamic Cordis plugins temporarily extend the current DSH process. A Plugin uses apply(ctx) to consume Services, listen to Events, provide Services, register model Tools, or register browser UI in Slots.\n\n- Plugin and Package definitions exist only in the current process. define itself does not modify repository source, configuration, or disk, and definitions do not survive a process restart.\n- The restricted execution environment prevents accidental misuse; it is not a security boundary for malicious code. Services obtained by dynamic code connect to the real runtime.\n\n## Make the user-facing plan clear first\n\n- Dynamic Cordis Plugins are one available implementation mechanism, not the default for every request. Consider whether one could help only when the user intends to design or create something, or when a temporary interface could materially aid the current work. The presence of these instructions or Tools, and discussion of Cordis itself, do not make a request a dynamic-Plugin task.\n- When Cordis is a plausible fit, infer the intended work target and lifetime from the request and conversation. Use it only when the outcome belongs to the current running harness and should be delivered as a temporary runtime extension. If that distinction is materially ambiguous, ask at most one concise question about the intended result or lifetime. Otherwise proceed with the matching workflow; do not require the user to know or choose Cordis as an implementation mechanism.\n- Once a dynamic Plugin is appropriate, decide whether the task creates a new Plugin or modifies the Plugin named by the user with @pluginId. Proceed directly when the goal is clear; do not ask for repeated confirmation.\n- Choose Host, Client, or both from the requested outcome. Do not propose a Client/browser UI when the task does not need visible page behavior, and do not avoid Client when the requested outcome is visual, interactive, or depends on page state. Host versus Client is an implementation choice; do not make the user choose it.\n- When a design direction or a potentially useful interface would materially affect the result, ask at most one concise outcome or creative-preference question and offer a few candidate directions. Otherwise proceed directly; do not conduct a multi-round interview or a complex questionnaire.\n- cordis_define only defines and presents code; it does not run it. After definition, explain the pluginId and packageId returned by the Host and whether the next step is a run or update.\n- cordis_run may require user approval. When it returns awaiting-approval, explain that the user must allow or reject it in the UI. Do not wait, retry, or claim that it is running.\n- When it returns starting, explain that the request has entered the asynchronous flow and the Client is still activating. starting does not mean success. Wait for the system to report the final result through steering context.\n- Do not request approval again after the user rejects it. After a technical failure, fix the same Plugin from its diagnostics; do not silently create a replacement Plugin.\n\n## Recommended workflow and Tools\n\nBefore creating, modifying, or repairing a Plugin, load the cordis-plugin-development Skill. The Skill provides requirement navigation, capability composition, complete examples, and troubleshooting. Treat Inspect Provider results as the source of truth for exact APIs.\n\n1. cordis_inspect_list: discover the current Host and Client Providers and their read-only query methods.\n2. cordis_inspect_query: use the returned platform, provider, method, and schema to query exact Service, Event, Builtin, Slot, Theme token, or Tool information.\n3. cordis_inspect_self: inspect the current Session's Plugins, Packages, version pointers, source, and diagnostics. Source is returned only when both pluginId and packageId are specified.\n4. cordis_define: create the first Package for a new Plugin or append an immutable Package to an existing Plugin. It defines code but does not run it.\n5. cordis_run: activate an exact Package. Use run for the first activation, restarting current, or rollback; use update to switch versions.\n6. cordis_stop: remove the current Run and pending approval request while retaining definitions, grants, and version pointers.\n7. cordis_undefine: permanently stop and delete a Plugin and all of its Packages. Use it only after confirming that the user no longer needs them.\n\n- Inspect and Catalog data only confirm capabilities, names, signatures, types, and registration protocols before code is written; they do not replace business APIs.\n- Query Service.listService and Event.listEvents without input to choose from their compact signature directories, then query the exact service or event before using it. Exact queries return the structured contract and only its referenced types.\n- At runtime, a Plugin must call real Services or listen to real Events. Do not cache, display, or depend on Inspect results as business data.\n\n## Identity, versions, and approval\n\n- pluginId identifies a Plugin that can be modified over time. For a new Plugin, submit only a semantic idPrefix of 3–6 lowercase English letters; the Host allocates the final ID.\n- packageId identifies one immutable Host/Client source version under a Plugin. To change code, define a new Package; never overwrite an old version.\n- pluginRunId identifies one activation attempt and connects its approval, Host/Client loading, private RPC, Run card, and errors.\n- currentPackageId is the most recent fully successful Package. Stopping, starting an update, or failing an update does not clear it.\n- nextPackageId is the target awaiting approval, being attempted, awaiting Client activation, or most recently failed.\n- A single check mark authorizes only the current Package; double check marks authorize future versions of the same Plugin. A grant remains in effect after a technical failure.\n- An update stops the old Run before starting the target Package. Failure does not automatically restart the old version; retry next with update or roll back to current with run.\n\nWhen the user enters @pluginId, the system injects identity, the default base Package, version pointers, and runtime status, but not source code:\n\n1. Call cordis_inspect_self(pluginId, packageId) to read the target source.\n2. Use cordis_define in existing mode to append a Package to the same Plugin.\n3. Call cordis_run in run or update mode according to the version relationship.\n\nNever silently create another Plugin for @pluginId. If the reference is unavailable because it was removed, belongs to another Session, or was lost on process restart, tell the user directly.\n\n## High-frequency errors that must be avoided\n\n### Services: ctx.get and inject\n\n- Read an optional Service with ctx.get('serviceName') by default and handle undefined.\n- Declare inject: ['serviceName'] on the returned Plugin object only when the Service is a hard dependency and the Plugin must enter waiting until Cordis reactivates it after the Service appears.\n- Read ctx.serviceName only after declaring that Service in inject. Never access an undeclared Service as a ctx property.\n\n```js\nreturn {\n inject: ['requiredService'],\n apply(ctx) {\n ctx.requiredService.someMethod()\n const optionalService = ctx.get('optionalService')\n if (optionalService !== undefined) optionalService.someMethod()\n },\n}\n```\n\n### Code: use plain JavaScript only\n\n- Host and Client code is not transformed by TypeScript, JSX, or a bundler.\n- Do not use TypeScript types, as, decorators, import, require, or JSX.\n- Client React code must use React.createElement(...); never write .\n- Do not assume that process, Buffer, window, document, fetch, native timers, or any other global is available. Query the corresponding platform's Builtins and Services first.\n\n### Data: do not serialize live data\n\n- Services, Events, Slots, Sessions, and their derived Cordis/DSH objects are internal live data, not ordinary JSON that can be dumped.\n- Do not apply JSON.stringify, structuredClone, recursive enumeration, full copying, or whole-object display to live data.\n- Read only the leaf fields required by the task, then construct the smallest owned data object without Host references.\n\n### Lifecycle: every side effect must be reversible\n\n- Services, Events, Tools, handlers, timers, Slots, styles, and theme overrides must all belong to the current Fiber.\n- Use ctx.effect(), ctx.on(), or official APIs that return a disposer so stop, update, or undefine removes every side effect.\n- The cordis-plugin-development Skill contains complete timer, Waterfall, Slot, theme, Tool, RPC, and React examples and troubleshooting guidance.\n\n## Host and Client\n\n- Host runs in the DSH Node.js process and is appropriate for files, networking, commands, Agent/Session access, Host Events, Services, model Tools, and JSON methods callable by the Client.\n- Client runs in the browser page and is appropriate for themes, layout, current page state, Tool cards, and Slot UI.\n- Host and Client communicate through Package-private JSON methods: Host uses harness.handle(method, handler), and Client uses host.call(method, args). The direction is Client→Host, and only lossless JSON may cross it.\n- Client UI must be registered in a queried Slot; apply() cannot directly return a React Element. Query Slots.listSubTree without root to choose from the compact purpose/topology tree, then query the exact root for its full registration contract and props before writing code.\n- See the Skill and Inspect Providers for Run-specific panels and exact Slot registration patterns.\n\n## Asynchronous results and recovery\n\n- Do not wait inside a Tool for approval or browser work that can happen only after the current turn ends.\n- Asynchronous success, rejection, and runtime errors update Run state and notify you through steering context.\n- After a technical failure, use cordis_inspect_self to read the exact Package source and its message/stack. Define a corrected Package under the same Plugin and retry autonomously.\n- Use the cordis-plugin-development Skill for other failure causes, repair procedures, and complete extension patterns.\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\nUse subagent in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set `run_in_background: false` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.\n\n## Writing code for run_code\n\n`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. 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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies. */\n run_in_background?: boolean;\n } & Record;\n /** Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs. */\n cordis_define: {\n plugin: {\n kind: \"new\";\n /** Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix. */\n idPrefix: string;\n } | {\n kind: \"existing\";\n /** Exact ID of an existing Plugin; the new Package is appended to that instance. */\n pluginId: string;\n };\n /** Short, readable Package name. */\n name: string;\n /** One-sentence, user-facing description of the Package purpose. */\n purpose: string;\n code: {\n /** Plain JavaScript function body that returns the Host-half Cordis Plugin. */\n host?: string;\n /** Plain JavaScript function body that returns the browser Client-half Cordis Plugin. */\n client?: string;\n };\n } & Record;\n /** List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call. */\n cordis_inspect_list: Record;\n /** Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props. */\n cordis_inspect_query: {\n /** Runtime platform that owns the Provider. */\n platform: \"host\" | \"client\";\n /** Exact Provider ID returned by cordis_inspect_list. */\n provider: string;\n /** Exact method name declared by the Provider manifest. */\n method: string;\n /** Optional query input; it must satisfy the method input schema. */\n input?: JsonValue;\n } & Record;\n /** Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers. */\n cordis_inspect_self: {\n /** Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin. */\n pluginId?: string;\n /** Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned. */\n packageId?: string;\n } & Record;\n /** Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it. */\n cordis_run: {\n /** Stable Plugin ID returned by cordis_define. */\n pluginId: string;\n /** Exact immutable Package ID to activate under that Plugin. */\n packageId: string;\n /** Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package. */\n mode: \"run\" | \"update\";\n } & Record;\n /** Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal. */\n cordis_stop: {\n /** Stable dynamic Plugin ID to stop. */\n pluginId: string;\n } & Record;\n /** Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead. */\n cordis_undefine: {\n /** Stable dynamic Plugin ID to remove permanently. */\n pluginId: 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 /** Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops. */\n job_kill: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Optional short reason, recorded in the log and forwarded to the job. */\n reason?: string;\n } & Record;\n /** List your background jobs (running and finished) with their ids, kinds, and statuses. */\n job_list: Record;\n /** Read a background job. Stream jobs return only output since the previous read; final-output jobs 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 job_output: {\n /** Job id returned by the tool that started the background work. */\n job_id: string;\n /** Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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 /** 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result. */\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 /** Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it. */\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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result. */\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 /** 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 jobId: 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_define: {\n pluginId: string;\n packageId: string;\n name: string;\n purpose: string;\n hasHostHalf: boolean;\n hasClientHalf: boolean;\n };\n cordis_inspect_list: JsonValue;\n cordis_inspect_query: JsonValue;\n cordis_inspect_self: JsonValue;\n cordis_run: JsonValue;\n cordis_stop: {\n pluginId: string;\n };\n cordis_undefine: {\n pluginId: string;\n wasRunning: boolean;\n };\n edit: {\n path: string;\n before: string;\n after: string;\n };\n interrupt_agent: {\n accepted: boolean;\n };\n job_kill: {\n outcome: \"cancellation-requested\" | \"already-finished\";\n job: {\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 job_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 job_output: {\n text: string;\n job: {\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 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 jobId: 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 jobId: string;\n } | {\n kind: \"continuable\";\n subagentId: string;\n } | {\n kind: \"foreground\";\n runId: string;\n output: JsonValue[];\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 job id immediately; read its output with `job_output` and stop it with `job_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 job id immediately (collect with job_output, stop with job_kill). No timeout applies."}},"required":["command","description"]}},{"name":"cordis_define","description":"Define an immutable Cordis Package. For a new Plugin, use kind:\"new\" and provide only a semantic prefix of 3–6 lowercase English letters; the Host returns the final pluginId and packageId. To modify an existing Plugin, use kind:\"existing\" with its exact pluginId to append a Package without overwriting older versions. Provide at least one of code.host and code.client. Each value is a plain JavaScript function body that returns a Cordis Plugin; no TypeScript, JSX, or import transformation occurs. Query Inspect before depending on a Service, Event, Builtin, Slot, or token. Define only validates parameters and syntax and records source: it does not request approval, execute apply, or change currentPackageId. On success, call cordis_run with the returned IDs.","parameters":{"type":"object","properties":{"plugin":{"oneOf":[{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"new"},"idPrefix":{"type":"string","description":"Suggested semantic prefix of 3–6 lowercase English letters; the Host adds a unique numeric suffix."}},"required":["kind","idPrefix"]},{"type":"object","additionalProperties":false,"properties":{"kind":{"type":"string","const":"existing"},"pluginId":{"type":"string","description":"Exact ID of an existing Plugin; the new Package is appended to that instance."}},"required":["kind","pluginId"]}]},"name":{"type":"string","description":"Short, readable Package name."},"purpose":{"type":"string","description":"One-sentence, user-facing description of the Package purpose."},"code":{"type":"object","additionalProperties":false,"properties":{"host":{"type":"string","description":"Plain JavaScript function body that returns the Host-half Cordis Plugin."},"client":{"type":"string","description":"Plain JavaScript function body that returns the browser Client-half Cordis Plugin."}}}},"required":["plugin","name","purpose","code"]}},{"name":"cordis_inspect_list","description":"List every Cordis Inspect Provider currently known to the Host, including local Host Providers and the latest manifests synchronized from the Client. Each entry includes its platform, purpose, read-only methods, and input/output schemas. Call this Tool before creating or modifying a Package, then select the provider and method for cordis_inspect_query from its result. Do not guess names or treat an Inspect method as a business Service that Plugin code can call.","parameters":{"type":"object","properties":{}}},{"name":"cordis_inspect_query","description":"Run a read-only query explicitly declared by an Inspect Provider. platform, provider, and method must come from cordis_inspect_list, and input must satisfy that method's schema. Use this Tool before cordis_define to read exact Service methods, Event modes, Builtin signatures, Tool schemas, theme tokens, or live Slot trees and props. Host queries run locally. A Client query waits for the first valid page response and remains pending until a page answers or the Tool is cancelled. This Tool cannot invoke business Service methods or modify the runtime. For Service.listService and Event.listEvents, query without input to navigate the compact signature directory, then query the exact service or event for its structured contract and referenced types. For Slots.listSubTree, query without root to navigate the compact tree, then query the exact root for its complete registration contract and props.","parameters":{"type":"object","properties":{"platform":{"type":"string","description":"Runtime platform that owns the Provider.","enum":["host","client"]},"provider":{"type":"string","description":"Exact Provider ID returned by cordis_inspect_list."},"method":{"type":"string","description":"Exact method name declared by the Provider manifest."},"input":{"description":"Optional query input; it must satisfy the method input schema."}},"required":["platform","provider","method"]}},{"name":"cordis_inspect_self","description":"Inspect dynamic Cordis objects owned by the current Session at increasing levels of detail. With no IDs, list only Plugin summaries. With pluginId alone, return version pointers, the latest Run, and every Package summary. Only pluginId plus packageId returns that immutable Package's Host/Client source and runtime diagnostics. packageId cannot be supplied alone. Query an exact Package before handling @pluginId, repairing an asynchronous failure, or defining an updated version. This Tool is read-only: it neither executes code nor changes version pointers.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define or injected by @pluginId; omit it to list every current Plugin."},"packageId":{"type":"string","description":"Exact immutable Package ID owned by pluginId; when specified, source and diagnostics are returned."}}}},{"name":"cordis_run","description":"Activate one exact Package of a dynamic Plugin. Use mode:\"run\" for the first activation, restarting currentPackageId, or rollback. When current exists, use mode:\"update\" to switch to a different Package, even if the Plugin is currently stopped. An unauthorized Client Package creates an approval request and returns awaiting-approval; an authorized Package returns starting and continues asynchronously in the browser. Neither result waits for the final outcome inside the Tool. currentPackageId changes only after complete success; on failure, the old current and target next remain. Asynchronous success, rejection, or technical failure is reported through state and steering. After a technical failure, read diagnostics with cordis_inspect_self, correct the same Plugin, and retry autonomously. Do not request approval again after the user rejects it.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable Plugin ID returned by cordis_define."},"packageId":{"type":"string","description":"Exact immutable Package ID to activate under that Plugin."},"mode":{"type":"string","description":"Use run for the first activation, restarting current, or rollback; use update to switch from current to a different Package.","enum":["run","update"]}},"required":["pluginId","packageId","mode"]}},{"name":"cordis_stop","description":"Stop the current Run of a dynamic Plugin and cancel unfinished approval or activation requests. Retain the Plugin, every immutable Package, grants, currentPackageId, and nextPackageId so it can later run or update directly. Stopping an already stopped Plugin succeeds idempotently. Use this Tool to disable effects temporarily; use cordis_undefine for permanent removal.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to stop."}},"required":["pluginId"]}},{"name":"cordis_undefine","description":"Permanently remove a dynamic Plugin owned by the current Session. If it is running or awaiting approval, first stop it and cancel the request, then delete every Package, grant, and version pointer. After this returns, its pluginId, packageIds, @ reference, and Package business views are invalid; historical cards retain only a \"Plugin removed\" record. Do not call this Tool when versions must remain available for restart or rollback; use cordis_stop instead.","parameters":{"type":"object","properties":{"pluginId":{"type":"string","description":"Stable dynamic Plugin ID to remove permanently."}},"required":["pluginId"]}},{"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":"job_kill","description":"Request cancellation of a running background job by job id. Returns immediately; the job settles as killed once its work actually stops.","parameters":{"type":"object","properties":{"job_id":{"type":"string","description":"Job 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 job."}},"required":["job_id"]}},{"name":"job_list","description":"List your background jobs (running and finished) with their ids, kinds, and statuses.","parameters":{"type":"object","properties":{}}},{"name":"job_output","description":"Read a background job. Stream jobs return only output since the previous read; final-output jobs 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":{"job_id":{"type":"string","description":"Job id returned by the tool that started the background work."},"wait":{"type":"boolean","description":"Block until the job reaches a terminal status or the timeout expires. A timed-out wait returns [status: running] and leaves the job 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":["job_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. Takes two required arguments: `code`, the BODY of an async function (erasable syntax only; top-level `await` and `return` work), and `description`, a short summary of what the program does. 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) to offload focused, independent work — research, a scoped implementation, an analysis — so it does not consume this conversation's context. The subagent returns its result, not its intermediate steps. Give it a complete, standalone prompt: it does not see this conversation. This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result.","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":"Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it."}},"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). 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 its result, not its intermediate steps. This call waits for the subagent and returns its result.","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":"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-define","name":"cordis_define","argumentsDelta":"{\"plugin\":{\"kind\":\"new\",\"idPrefix\":\"snap\"},\"name\":\"Snapshot Marker\",\"purpose\":\"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\"code\":{\"host\":\"return { 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-define","name":"cordis_define","arguments":"{\"plugin\":{\"kind\":\"new\",\"idPrefix\":\"snap\"},\"name\":\"Snapshot Marker\",\"purpose\":\"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\"code\":{\"host\":\"return { 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-define","name":"cordis_define","arguments":"{\"plugin\":{\"kind\":\"new\",\"idPrefix\":\"snap\"},\"name\":\"Snapshot Marker\",\"purpose\":\"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\"code\":{\"host\":\"return { apply() {} }\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"79c1bd38-584a-4756-8854-211919937987"},"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-define","name":"cordis_define","arguments":"{\"plugin\":{\"kind\":\"new\",\"idPrefix\":\"snap\"},\"name\":\"Snapshot Marker\",\"purpose\":\"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\"code\":{\"host\":\"return { apply() {} }\"}}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e0351488-7bca-48d6-b7af-87d533858f47"},"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-define","name":"cordis_define","arguments":"{\"plugin\":{\"kind\":\"new\",\"idPrefix\":\"snap\"},\"name\":\"Snapshot Marker\",\"purpose\":\"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\"code\":{\"host\":\"return { apply() {} }\"}}"}} -{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-define"},"content":[{"type":"tool-result","toolCallId":"advanced-define","content":[{"type":"text","text":"Defined snap-1/pkg-1 (Snapshot Marker); it is not running yet. Use cordis_run to activate this Package."}],"isError":false}],"role":"user","id":"8255947c-3d5d-4872-8359-914555595bd3"},"meta":{"pluginId":"snap-1","packageId":"pkg-1"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"tool/result","seq":15,"time":1785730501413,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"advanced-define"},"content":[{"type":"tool-result","toolCallId":"advanced-define","content":[{"type":"text","text":"Defined snap-1/pkg-1 (Snapshot Marker); it is not running yet. Use cordis_run to activate this Package."}],"isError":false}],"role":"user","id":"9fd9ba62-1b6d-4bb7-98c6-97815e983026"},"meta":{"pluginId":"snap-1","packageId":"pkg-1"}},"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"}}} @@ -22,13 +22,13 @@ {"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\":\"const run = await tools.cordis_run({ pluginId: 'snap-1', packageId: 'pkg-1', mode: 'run' });\\nconst inspected = await tools.cordis_inspect_self({ pluginId: 'snap-1' });\\nreturn { run, inspected };\",\"description\":\"Run and inspect the dynamic Cordis Package\"}"}}}} {"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\":\"const run = await tools.cordis_run({ pluginId: 'snap-1', packageId: 'pkg-1', mode: 'run' });\\nconst inspected = await tools.cordis_inspect_self({ pluginId: 'snap-1' });\\nreturn { run, inspected };\",\"description\":\"Run and inspect the dynamic Cordis Package\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e73246e2-257d-4d38-a8cd-5e5f713d4ad3"},"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\":\"const run = await tools.cordis_run({ pluginId: 'snap-1', packageId: 'pkg-1', mode: 'run' });\\nconst inspected = await tools.cordis_inspect_self({ pluginId: 'snap-1' });\\nreturn { run, inspected };\",\"description\":\"Run and inspect the dynamic Cordis Package\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"6f6c3cc3-350b-4afb-a3d1-8f91b9494628"},"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\":\"const run = await tools.cordis_run({ pluginId: 'snap-1', packageId: 'pkg-1', mode: 'run' });\\nconst inspected = await tools.cordis_inspect_self({ pluginId: 'snap-1' });\\nreturn { run, inspected };\",\"description\":\"Run and inspect the dynamic Cordis Package\"}"}} {"type":"tool/code-dispatch-start","seq":25,"time":1785730501473,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_run","arguments":{"pluginId":"snap-1","packageId":"pkg-1","mode":"run"}}} {"type":"tool/code-dispatch","seq":26,"time":1785730501474,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:1","name":"cordis_run","arguments":{"pluginId":"snap-1","packageId":"pkg-1","mode":"run"},"isError":false,"content":[{"type":"text","text":"snap-1/pkg-1 is running (run-1)."}]}} {"type":"tool/code-dispatch-start","seq":27,"time":1786558903189,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:2","name":"cordis_inspect_self","arguments":{"pluginId":"snap-1"}}} {"type":"tool/code-dispatch","seq":28,"time":1786558903189,"data":{"rootCallId":"advanced-code","parentCallId":"advanced-code","subCallId":"advanced-code:code:2","name":"cordis_inspect_self","arguments":{"pluginId":"snap-1"},"isError":false,"content":[{"type":"text","text":"{\n \"mode\": \"plugin\",\n \"pluginId\": \"snap-1\",\n \"name\": \"Snapshot Marker\",\n \"packageCount\": 1,\n \"state\": \"running\",\n \"currentPackageId\": \"pkg-1\",\n \"activeRun\": {\n \"pluginRunId\": \"run-1\",\n \"packageId\": \"pkg-1\"\n },\n \"packages\": [\n {\n \"packageId\": \"pkg-1\",\n \"name\": \"Snapshot Marker\",\n \"purpose\": \"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\n \"hasHostHalf\": true,\n \"hasClientHalf\": false,\n \"isCurrent\": true,\n \"isNext\": false\n }\n ]\n}"}]}} -{"type":"tool/result","seq":29,"time":1786558903193,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"{\n \"run\": {\n \"status\": \"running\",\n \"pluginId\": \"snap-1\",\n \"packageId\": \"pkg-1\",\n \"pluginRunId\": \"run-1\",\n \"currentPackageId\": \"pkg-1\",\n \"host\": {\n \"status\": \"running\",\n \"provides\": [],\n \"waitingFor\": []\n },\n \"client\": {\n \"status\": \"absent\",\n \"waitingFor\": []\n }\n },\n \"inspected\": {\n \"mode\": \"plugin\",\n \"pluginId\": \"snap-1\",\n \"name\": \"Snapshot Marker\",\n \"packageCount\": 1,\n \"state\": \"running\",\n \"currentPackageId\": \"pkg-1\",\n \"activeRun\": {\n \"pluginRunId\": \"run-1\",\n \"packageId\": \"pkg-1\"\n },\n \"packages\": [\n {\n \"packageId\": \"pkg-1\",\n \"name\": \"Snapshot Marker\",\n \"purpose\": \"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\n \"hasHostHalf\": true,\n \"hasClientHalf\": false,\n \"isCurrent\": true,\n \"isNext\": false\n }\n ]\n }\n}"}],"isError":false}],"role":"user","id":"acb0a833-2535-41ff-9e1d-926c092329b6"}},"sourceEventSeqs":[24],"surfaceOp":"append"} +{"type":"tool/result","seq":29,"time":1786558903193,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"advanced-code"},"content":[{"type":"tool-result","toolCallId":"advanced-code","content":[{"type":"text","text":"{\n \"run\": {\n \"status\": \"running\",\n \"pluginId\": \"snap-1\",\n \"packageId\": \"pkg-1\",\n \"pluginRunId\": \"run-1\",\n \"currentPackageId\": \"pkg-1\",\n \"host\": {\n \"status\": \"running\",\n \"provides\": [],\n \"waitingFor\": []\n },\n \"client\": {\n \"status\": \"absent\",\n \"waitingFor\": []\n }\n },\n \"inspected\": {\n \"mode\": \"plugin\",\n \"pluginId\": \"snap-1\",\n \"name\": \"Snapshot Marker\",\n \"packageCount\": 1,\n \"state\": \"running\",\n \"currentPackageId\": \"pkg-1\",\n \"activeRun\": {\n \"pluginRunId\": \"run-1\",\n \"packageId\": \"pkg-1\"\n },\n \"packages\": [\n {\n \"packageId\": \"pkg-1\",\n \"name\": \"Snapshot Marker\",\n \"purpose\": \"Exercise the dynamic Cordis Package lifecycle in the snapshot.\",\n \"hasHostHalf\": true,\n \"hasClientHalf\": false,\n \"isCurrent\": true,\n \"isNext\": false\n }\n ]\n }\n}"}],"isError":false}],"role":"user","id":"40a17cbf-a853-4813-bbb5-7970cfbc7010"}},"sourceEventSeqs":[24],"surfaceOp":"append"} {"type":"step/end","seq":30,"time":1786558903193,"data":{"turn":1,"step":2}} {"type":"step/start","seq":31,"time":1786558903203,"data":{"turn":1,"step":3}} {"type":"assistant/chunk","seq":32,"time":1785037378923,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -36,9 +36,9 @@ {"type":"assistant/chunk","seq":34,"time":1785730501484,"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.\",\"run_in_background\":false}"}}}} {"type":"assistant/chunk","seq":35,"time":1786558903204,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":36,"time":1786558903204,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":37,"time":1786558903204,"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.\",\"run_in_background\":false}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"32da94eb-ef97-4929-8dec-dea761506e90"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} +{"type":"assistant/message","seq":37,"time":1786558903204,"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.\",\"run_in_background\":false}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"8bdea090-4a9a-451f-bb21-459af50472fa"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[32,33,34,35,36],"surfaceOp":"append"} {"type":"tool/call","seq":38,"time":1786558903205,"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.\",\"run_in_background\":false}"}} -{"type":"tool/result","seq":39,"time":1786558903240,"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":"29fcc9e2-dee3-4bf7-b3a9-e167c00a755b"}},"sourceEventSeqs":[38],"surfaceOp":"append"} +{"type":"tool/result","seq":39,"time":1786558903240,"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":"c978c208-8ebd-4dd6-b997-606ca7de787e"}},"sourceEventSeqs":[38],"surfaceOp":"append"} {"type":"step/end","seq":40,"time":1786558903240,"data":{"turn":1,"step":3}} {"type":"step/start","seq":41,"time":1786558903249,"data":{"turn":1,"step":4}} {"type":"assistant/chunk","seq":42,"time":1785037378946,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -46,13 +46,13 @@ {"type":"assistant/chunk","seq":44,"time":1785730501522,"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":45,"time":1786558903250,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":46,"time":1786558903250,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":47,"time":1786558903250,"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":"6fff1bc7-3e35-415b-ae02-e018d360fe09"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[42,43,44,45,46],"surfaceOp":"append"} +{"type":"assistant/message","seq":47,"time":1786558903250,"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":"e55b2b2e-497c-45d2-8115-16f317ae573f"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[42,43,44,45,46],"surfaceOp":"append"} {"type":"tool/call","seq":48,"time":1786558903250,"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-workflow/run-start","seq":49,"time":1786558903255,"data":{"runId":"cd3d2666-94a1-4285-804e-c99630bc7b51","name":"advanced-headless-snapshot"}} {"type":"tool-workflow/agent-start","seq":50,"time":1786558903485,"data":{"runId":"cd3d2666-94a1-4285-804e-c99630bc7b51","seq":1,"label":"workflow-child","phase":"Delegate","childId":"33333333-3333-4333-8333-333333333333"}} {"type":"tool-workflow/agent-end","seq":51,"time":1786558903518,"data":{"runId":"cd3d2666-94a1-4285-804e-c99630bc7b51","seq":1,"outcome":"completed"}} {"type":"tool-workflow/run-end","seq":52,"time":1786558903521,"data":{"runId":"cd3d2666-94a1-4285-804e-c99630bc7b51","stopReason":"completed"}} -{"type":"tool/result","seq":53,"time":1786558903521,"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":"82895074-35a8-470a-a622-b951d2fc2665"}},"sourceEventSeqs":[48],"surfaceOp":"append"} +{"type":"tool/result","seq":53,"time":1786558903521,"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":"f85f58fc-8d7e-4c9f-a0fe-caff480a9fec"}},"sourceEventSeqs":[48],"surfaceOp":"append"} {"type":"step/end","seq":54,"time":1786558903521,"data":{"turn":1,"step":4}} {"type":"step/start","seq":55,"time":1786558903530,"data":{"turn":1,"step":5}} {"type":"assistant/chunk","seq":56,"time":1786359174239,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} @@ -60,9 +60,9 @@ {"type":"assistant/chunk","seq":58,"time":1786359174239,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"advanced-undefine","name":"cordis_undefine","arguments":"{\"pluginId\":\"snap-1\"}"}}}} {"type":"assistant/chunk","seq":59,"time":1786558903530,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":60,"time":1786558903530,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":61,"time":1786558903530,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-undefine","name":"cordis_undefine","arguments":"{\"pluginId\":\"snap-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"b1219ecf-66a4-4de1-b822-a39b6651bd90"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[56,57,58,59,60],"surfaceOp":"append"} +{"type":"assistant/message","seq":61,"time":1786558903530,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"advanced-undefine","name":"cordis_undefine","arguments":"{\"pluginId\":\"snap-1\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"68bd993f-a0bd-4ea8-ad16-6b1a19e09bd3"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[56,57,58,59,60],"surfaceOp":"append"} {"type":"tool/call","seq":62,"time":1786558903531,"data":{"turn":1,"step":5,"callId":"advanced-undefine","name":"cordis_undefine","arguments":"{\"pluginId\":\"snap-1\"}"}} -{"type":"tool/result","seq":63,"time":1786558903537,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-undefine"},"content":[{"type":"tool-result","toolCallId":"advanced-undefine","content":[{"type":"text","text":"Removed dynamic Plugin snap-1 and all of its Packages."}],"isError":false}],"role":"user","id":"30c8e3bb-2ada-4576-9764-6e87f5b8fc94"}},"sourceEventSeqs":[62],"surfaceOp":"append"} +{"type":"tool/result","seq":63,"time":1786558903537,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"advanced-undefine"},"content":[{"type":"tool-result","toolCallId":"advanced-undefine","content":[{"type":"text","text":"Removed dynamic Plugin snap-1 and all of its Packages."}],"isError":false}],"role":"user","id":"2d90e3b5-2a4c-4408-a1a0-3d009786a07b"}},"sourceEventSeqs":[62],"surfaceOp":"append"} {"type":"step/end","seq":64,"time":1786558903537,"data":{"turn":1,"step":5}} {"type":"step/start","seq":65,"time":1786558903547,"data":{"turn":1,"step":6}} {"type":"assistant/chunk","seq":66,"time":1786359174249,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} @@ -70,6 +70,6 @@ {"type":"assistant/chunk","seq":68,"time":1786359174249,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"ADVANCED_HEADLESS_OK"}}}} {"type":"assistant/chunk","seq":69,"time":1786558903548,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} {"type":"assistant/chunk","seq":70,"time":1786558903548,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":71,"time":1786558903548,"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":"9b46dd3a-e3ea-4f25-a246-814f31826e9e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[66,67,68,69,70],"surfaceOp":"append"} +{"type":"assistant/message","seq":71,"time":1786558903548,"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":"f3823367-2e25-43d2-a129-70dc492b2a90"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[66,67,68,69,70],"surfaceOp":"append"} {"type":"step/end","seq":72,"time":1786558903549,"data":{"turn":1,"step":6}} {"type":"turn/end","seq":73,"time":1786558903549,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/examples/jsonrpc-agent/README.i18n.yaml b/examples/jsonrpc-agent/README.i18n.yaml index b8cf6969df..78a96ee6ac 100644 --- a/examples/jsonrpc-agent/README.i18n.yaml +++ b/examples/jsonrpc-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 examples/jsonrpc-agent/README.md -README.md: 5166d94e92e2e4bcdda5026fa65ba6b77b87fe40 -README.zh.md: f84ab95132e820cf0fcf45bff4ae30d9bccb55c1 +README.md: ec94fa7ceec5a51585241851eaee2cfcf938738f +README.zh.md: 64a5e607bd37e97b60cb0a4f6ee89a7c766d7464 diff --git a/examples/jsonrpc-agent/README.md b/examples/jsonrpc-agent/README.md index 5166d94e92..ec94fa7cee 100644 --- a/examples/jsonrpc-agent/README.md +++ b/examples/jsonrpc-agent/README.md @@ -30,7 +30,7 @@ Pass the config path through the Python SDK's `cordis` option or `DSH_CORDIS_CON ## Minimal variant -[`minimal.cordis.yml`](minimal.cordis.yml) is the complete standalone counterpart of the Web `minimal` preset. `DSH_SYSTEM_PROMPT` selects its system prompt, with `You are a helpful software engineer assistant.` as the fallback, and no context-compaction plugin is mounted. Its model-facing tools are exactly: +[`minimal.cordis.yml`](minimal.cordis.yml) is the complete standalone counterpart of the Web `minimal` preset. `DSH_SYSTEM_PROMPT` selects its system prompt, with `You are a helpful software engineer assistant.` as the fallback. It suppresses every system-prompt runtime-context contribution for fresh sessions and mounts no context-compaction plugin. Its model-facing tools are exactly: - owner-scoped persistent `bash` - `str_replace_editor` with `view`, `create`, `str_replace`, and `insert` diff --git a/examples/jsonrpc-agent/README.zh.md b/examples/jsonrpc-agent/README.zh.md index f84ab95132..64a5e607bd 100644 --- a/examples/jsonrpc-agent/README.zh.md +++ b/examples/jsonrpc-agent/README.zh.md @@ -30,7 +30,7 @@ ## 极简变体 -[`minimal.cordis.yml`](minimal.cordis.yml) 是 Web `minimal` preset 的完整独立版本。`DSH_SYSTEM_PROMPT` 选择它的系统提示词，未设置时使用 `You are a helpful software engineer assistant.`，且不挂载上下文压缩插件。面向模型的工具严格只有： +[`minimal.cordis.yml`](minimal.cordis.yml) 是 Web `minimal` preset 的完整独立版本。`DSH_SYSTEM_PROMPT` 选择它的系统提示词，未设置时使用 `You are a helpful software engineer assistant.`。它为新建会话抑制每个 system-prompt runtime-context 贡献，且不挂载上下文压缩插件。面向模型的工具严格只有： - 所有者作用域内持久化的 `bash` - 提供 `view`、`create`、`str_replace` 与 `insert` 的 `str_replace_editor` diff --git a/examples/jsonrpc-agent/minimal.cordis.yml b/examples/jsonrpc-agent/minimal.cordis.yml index 8d6e9b8181..e23d52a866 100644 --- a/examples/jsonrpc-agent/minimal.cordis.yml +++ b/examples/jsonrpc-agent/minimal.cordis.yml @@ -1,6 +1,7 @@ # Complete unattended minimal-agent composition for the Python SDK. The model # sees one deployment-selected system prompt and only the owner-scoped -# persistent Bash and string-replace editor tools. Context compaction is absent. +# persistent Bash and string-replace editor tools. Runtime-context injection and +# context compaction are absent. - id: sdk-jsonrpc-server name: '@deepseek-ai/dsh-sdk-jsonrpc-server' @@ -47,6 +48,7 @@ name: '@deepseek-ai/dsh-agent-spine-demo' config: includeHarnessIdentity: false + includeRuntimeContext: false persona: !!js process.env.DSH_SYSTEM_PROMPT ?? 'You are a helpful software engineer assistant.' workspaceContext: false skills: diff --git a/examples/jsonrpc-agent/tests/sdk.snapshot.ts b/examples/jsonrpc-agent/tests/sdk.snapshot.ts index 85667cf64d..9b45292df5 100644 --- a/examples/jsonrpc-agent/tests/sdk.snapshot.ts +++ b/examples/jsonrpc-agent/tests/sdk.snapshot.ts @@ -77,8 +77,8 @@ interface SdkScenario { expectedSystem?: string /** Exact model-facing descriptions for selected tools. */ expectedToolDescriptions?: Readonly> - /** Stable policy-context clauses the real assembled request must include or omit. */ - policyContext?: { includes: readonly string[]; excludes: readonly string[] } + /** Expected runtime-context state in the real assembled request. */ + runtimeContext?: false | { includes: readonly string[]; excludes: readonly string[] } } const SCENARIOS: SdkScenario[] = [ @@ -111,10 +111,7 @@ const SCENARIOS: SdkScenario[] = [ expectedTools: { bash: ['command'], str_replace_editor: ['command', 'path'] }, expectedSystem: MINIMAL_SYSTEM_PROMPT, expectedToolDescriptions: { bash: MINIMAL_BASH_DESCRIPTION }, - policyContext: { - includes: ['Current DSH file policy: danger-full-access.', 'file modifications by available operations'], - excludes: ['write and edit tools', 'terminal sessions', 'one-shot bash commands'], - }, + runtimeContext: false, }, ] @@ -182,8 +179,8 @@ function assembledSystem(log: PersistedLog): string { return system } -function assembledPolicyContext(log: PersistedLog): string { - const contexts = log.content.trimEnd().split('\n').flatMap((line) => { +function assembledRuntimeContexts(log: PersistedLog): string[] { + return log.content.trimEnd().split('\n').flatMap((line) => { const event = JSON.parse(line) as { type?: string data?: { source?: { kind?: string; plugin?: string }; content?: Array<{ type?: string; text?: unknown }> } @@ -193,8 +190,6 @@ function assembledPolicyContext(log: PersistedLog): string { || event.data.source.plugin !== '@deepseek-ai/dsh-system-prompt') return [] return event.data.content?.flatMap(block => block.type === 'text' && typeof block.text === 'string' ? [block.text] : []) ?? [] }) - if (contexts.length !== 1) throw new Error(`session log has ${String(contexts.length)} runtime-context snapshots; expected one`) - return contexts[0] as string } function contextOf(logs: readonly { content: string; header: Record }[], cwd: string): NormalizeContext { @@ -447,14 +442,20 @@ describe('TypeScript SDK snapshots over the jsonrpc runtime', () => { if (parent === undefined) throw new Error(`${scenario.name} has no parent session log`) expect(assembledToolDescriptions(parent)).toMatchObject(scenario.expectedToolDescriptions) } - if (scenario.policyContext !== undefined) { + if (scenario.runtimeContext !== undefined) { const parent = ordered[0] if (parent === undefined) throw new Error(`${scenario.name} has no parent session log`) - const context = assembledPolicyContext(parent) - for (const clause of scenario.policyContext.includes) expect(context).toContain(clause) - for (const clause of scenario.policyContext.excludes) expect(context).not.toContain(clause) - const system = assembledSystem(parent) - for (const clause of scenario.policyContext.includes) expect(system).not.toContain(clause) + const contexts = assembledRuntimeContexts(parent) + if (scenario.runtimeContext === false) { + expect(contexts).toEqual([]) + } else { + expect(contexts).toHaveLength(1) + const context = contexts[0] as string + for (const clause of scenario.runtimeContext.includes) expect(context).toContain(clause) + for (const clause of scenario.runtimeContext.excludes) expect(context).not.toContain(clause) + const system = assembledSystem(parent) + for (const clause of scenario.runtimeContext.includes) expect(system).not.toContain(clause) + } } if (scenario.children > 0) { expect(notifications.some(n => n.method === 'subagent.started')).toBe(true) diff --git a/examples/jsonrpc-agent/tests/snapshots/persistent-tools/notifications.expected.jsonl b/examples/jsonrpc-agent/tests/snapshots/persistent-tools/notifications.expected.jsonl index ea8f220308..550d3495f9 100644 --- a/examples/jsonrpc-agent/tests/snapshots/persistent-tools/notifications.expected.jsonl +++ b/examples/jsonrpc-agent/tests/snapshots/persistent-tools/notifications.expected.jsonl @@ -4,76 +4,75 @@ {"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"agent/inbox/spliced","seq":2,"time":0,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}}}} {"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":3,"time":0,"data":{"turn":1,"step":1}}}} {"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"user/message","seq":4,"time":0,"data":{"content":[{"type":"text","text":"Prove that bash state persists. Then create {{cwd}}/note.txt with a tab-indented line, view it, replace that literal tab-indented line, and make the persistent shell exit with code 9."}],"source":{"kind":"user"},"role":"user","id":"{{sessionId}}"},"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"user/message","seq":5,"time":0,"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":"{{sessionId}}"},"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"session/title","seq":6,"time":0,"data":{"title":"Prove that bash state persists.","messageSeqs":[4],"source":{"kind":"fallback"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"request/header","seq":7,"time":0,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"request/context","seq":8,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-1","name":"bash","argumentsDelta":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":13,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":14,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":15,"time":0,"data":{"turn":1,"step":1,"callId":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":16,"time":0,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"bash-1"},"content":[{"type":"tool-result","toolCallId":"bash-1","content":[{"type":"text","text":"COUNT=1 CWD=/tmp"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[15],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":17,"time":0,"data":{"turn":1,"step":1}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":18,"time":0,"data":{"turn":1,"step":2}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":20,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-2","name":"bash","argumentsDelta":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":22,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":23,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":24,"time":0,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":25,"time":0,"data":{"turn":1,"step":2,"callId":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":26,"time":0,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"bash-2"},"content":[{"type":"tool-result","toolCallId":"bash-2","content":[{"type":"text","text":"COUNT=2 CWD=/tmp"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[25],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":27,"time":0,"data":{"turn":1,"step":2}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":28,"time":0,"data":{"turn":1,"step":3}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":30,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-create","name":"str_replace_editor","argumentsDelta":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":32,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":33,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":34,"time":0,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":35,"time":0,"data":{"turn":1,"step":3,"callId":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":36,"time":0,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"editor-create"},"content":[{"type":"tool-result","toolCallId":"editor-create","content":[{"type":"text","text":"New file created successfully at: {{cwd}}/note.txt"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[35],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":37,"time":0,"data":{"turn":1,"step":3}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":38,"time":0,"data":{"turn":1,"step":4}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":39,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":40,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-view","name":"str_replace_editor","argumentsDelta":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":41,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":42,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":43,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":44,"time":0,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":45,"time":0,"data":{"turn":1,"step":4,"callId":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":46,"time":0,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"editor-view"},"content":[{"type":"tool-result","toolCallId":"editor-view","content":[{"type":"text","text":"Here's the content of {{cwd}}/note.txt with line numbers (which has a total of 3 lines):\n 1 target:\n 2 \told\n 3 \n"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[45],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":47,"time":0,"data":{"turn":1,"step":4}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":48,"time":0,"data":{"turn":1,"step":5}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":49,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":50,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-replace","name":"str_replace_editor","argumentsDelta":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":51,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":52,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":53,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":54,"time":0,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":55,"time":0,"data":{"turn":1,"step":5,"callId":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":56,"time":0,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"editor-replace"},"content":[{"type":"tool-result","toolCallId":"editor-replace","content":[{"type":"text","text":"The file {{cwd}}/note.txt has been edited successfully."}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[55],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":57,"time":0,"data":{"turn":1,"step":5}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":58,"time":0,"data":{"turn":1,"step":6}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":59,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":60,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-exit","name":"bash","argumentsDelta":"{\"command\":\"exit 9\"}"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":61,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":62,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":63,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":64,"time":0,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":65,"time":0,"data":{"turn":1,"step":6,"callId":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":66,"time":0,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"bash-exit"},"content":[{"type":"tool-result","toolCallId":"bash-exit","content":[{"type":"text","text":"exit\n[shell exited: code 9]\nThe persistent bash shell was reset; the next bash call starts from the workspace with a fresh current directory and environment."}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[65],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":67,"time":0,"data":{"turn":1,"step":6}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":68,"time":0,"data":{"turn":1,"step":7}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":69,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":70,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"text-delta","index":0,"text":"PERSISTENT_TOOLS_OK"}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":71,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PERSISTENT_TOOLS_OK"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":72,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":73,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":74,"time":0,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"PERSISTENT_TOOLS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":75,"time":0,"data":{"turn":1,"step":7}}}} -{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"turn/end","seq":76,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"session/title","seq":5,"time":0,"data":{"title":"Prove that bash state persists.","messageSeqs":[4],"source":{"kind":"fallback"}}}}} +{"method":"session.event","params":{"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"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"request/context","seq":7,"time":0,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":8,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":9,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-1","name":"bash","argumentsDelta":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":10,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":11,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":12,"time":0,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":13,"time":0,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":14,"time":0,"data":{"turn":1,"step":1,"callId":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":15,"time":0,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"bash-1"},"content":[{"type":"tool-result","toolCallId":"bash-1","content":[{"type":"text","text":"COUNT=1 CWD=/tmp"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[14],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":16,"time":0,"data":{"turn":1,"step":1}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":17,"time":0,"data":{"turn":1,"step":2}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":18,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":19,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-2","name":"bash","argumentsDelta":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":20,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":21,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":22,"time":0,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":23,"time":0,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":24,"time":0,"data":{"turn":1,"step":2,"callId":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":25,"time":0,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"bash-2"},"content":[{"type":"tool-result","toolCallId":"bash-2","content":[{"type":"text","text":"COUNT=2 CWD=/tmp"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[24],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":26,"time":0,"data":{"turn":1,"step":2}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":27,"time":0,"data":{"turn":1,"step":3}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":28,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":29,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-create","name":"str_replace_editor","argumentsDelta":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":30,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":31,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":32,"time":0,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":33,"time":0,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[28,29,30,31,32],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":34,"time":0,"data":{"turn":1,"step":3,"callId":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":35,"time":0,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"editor-create"},"content":[{"type":"tool-result","toolCallId":"editor-create","content":[{"type":"text","text":"New file created successfully at: {{cwd}}/note.txt"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[34],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":36,"time":0,"data":{"turn":1,"step":3}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":37,"time":0,"data":{"turn":1,"step":4}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":38,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":39,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-view","name":"str_replace_editor","argumentsDelta":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":40,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":41,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":42,"time":0,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":43,"time":0,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[38,39,40,41,42],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":44,"time":0,"data":{"turn":1,"step":4,"callId":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":45,"time":0,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"editor-view"},"content":[{"type":"tool-result","toolCallId":"editor-view","content":[{"type":"text","text":"Here's the content of {{cwd}}/note.txt with line numbers (which has a total of 3 lines):\n 1 target:\n 2 \told\n 3 \n"}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[44],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":46,"time":0,"data":{"turn":1,"step":4}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":47,"time":0,"data":{"turn":1,"step":5}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":48,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":49,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-replace","name":"str_replace_editor","argumentsDelta":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":50,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":51,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":52,"time":0,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":53,"time":0,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[48,49,50,51,52],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":54,"time":0,"data":{"turn":1,"step":5,"callId":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":55,"time":0,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"editor-replace"},"content":[{"type":"tool-result","toolCallId":"editor-replace","content":[{"type":"text","text":"The file {{cwd}}/note.txt has been edited successfully."}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[54],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":56,"time":0,"data":{"turn":1,"step":5}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":57,"time":0,"data":{"turn":1,"step":6}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":58,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":59,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-exit","name":"bash","argumentsDelta":"{\"command\":\"exit 9\"}"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":60,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":61,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":62,"time":0,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":63,"time":0,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[58,59,60,61,62],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/call","seq":64,"time":0,"data":{"turn":1,"step":6,"callId":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"tool/result","seq":65,"time":0,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"bash-exit"},"content":[{"type":"tool-result","toolCallId":"bash-exit","content":[{"type":"text","text":"exit\n[shell exited: code 9]\nThe persistent bash shell was reset; the next bash call starts from the workspace with a fresh current directory and environment."}],"isError":false}],"role":"user","id":"{{sessionId}}"}},"sourceEventSeqs":[64],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":66,"time":0,"data":{"turn":1,"step":6}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/start","seq":67,"time":0,"data":{"turn":1,"step":7}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":68,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":69,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"text-delta","index":0,"text":"PERSISTENT_TOOLS_OK"}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":70,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PERSISTENT_TOOLS_OK"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":71,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/chunk","seq":72,"time":0,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"assistant/message","seq":73,"time":0,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"PERSISTENT_TOOLS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"{{sessionId}}"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[68,69,70,71,72],"surfaceOp":"append"}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"step/end","seq":74,"time":0,"data":{"turn":1,"step":7}}}} +{"method":"session.event","params":{"sessionId":"{{sessionId}}","event":{"type":"turn/end","seq":75,"time":0,"data":{"turn":1,"reason":{"kind":"completed"}}}}} {"method":"session.status","params":{"sessionId":"{{sessionId}}","status":"idle"}} diff --git a/examples/jsonrpc-agent/tests/snapshots/persistent-tools/session.jsonl b/examples/jsonrpc-agent/tests/snapshots/persistent-tools/session.jsonl index be559eeb9a..77fe6f03a6 100644 --- a/examples/jsonrpc-agent/tests/snapshots/persistent-tools/session.jsonl +++ b/examples/jsonrpc-agent/tests/snapshots/persistent-tools/session.jsonl @@ -4,75 +4,74 @@ {"type":"agent/inbox/spliced","seq":2,"time":1785821461907,"data":{"target":"next-turn","start":0,"removedCount":1,"inserted":[]}} {"type":"step/start","seq":3,"time":1785331618312,"data":{"turn":1,"step":1}} {"type":"user/message","seq":4,"time":1785498592368,"data":{"content":[{"type":"text","text":"Prove that bash state persists. Then create {{cwd}}/note.txt with a tab-indented line, view it, replace that literal tab-indented line, and make the persistent shell exit with code 9."}],"source":{"kind":"user"},"role":"user","id":"9a08e199-69d7-4b85-bfa4-27b41a92672a"},"surfaceOp":"append"} -{"type":"user/message","seq":5,"time":1785730508088,"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":"bb38bdc2-276e-46ec-87a1-089732acbc8d"},"surfaceOp":"append"} -{"type":"session/title","seq":6,"time":1785730508088,"data":{"title":"Prove that bash state persists.","messageSeqs":[4],"source":{"kind":"fallback"}}} -{"type":"request/header","seq":7,"time":1785498592370,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} -{"type":"request/context","seq":8,"time":1785730508089,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} -{"type":"assistant/chunk","seq":9,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":10,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-1","name":"bash","argumentsDelta":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}} -{"type":"assistant/chunk","seq":11,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} -{"type":"assistant/chunk","seq":12,"time":1785498592372,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":13,"time":1785730508090,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":14,"time":1785730508090,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0d064526-8eff-482d-8525-ac478e1d1791"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[9,10,11,12,13],"surfaceOp":"append"} -{"type":"tool/call","seq":15,"time":1785730508090,"data":{"turn":1,"step":1,"callId":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}} -{"type":"tool/result","seq":16,"time":1785730508425,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"bash-1"},"content":[{"type":"tool-result","toolCallId":"bash-1","content":[{"type":"text","text":"COUNT=1 CWD=/tmp"}],"isError":false}],"role":"user","id":"2c01f81a-01ea-47e2-bf92-f7825b7cc69f"}},"sourceEventSeqs":[15],"surfaceOp":"append"} -{"type":"step/end","seq":17,"time":1785730508425,"data":{"turn":1,"step":1}} -{"type":"step/start","seq":18,"time":1785730508425,"data":{"turn":1,"step":2}} -{"type":"assistant/chunk","seq":19,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":20,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-2","name":"bash","argumentsDelta":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}} -{"type":"assistant/chunk","seq":21,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} -{"type":"assistant/chunk","seq":22,"time":1785498592702,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":23,"time":1785730508426,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":24,"time":1785730508426,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ba4078ce-0e18-419a-b720-339918aecf26"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[19,20,21,22,23],"surfaceOp":"append"} -{"type":"tool/call","seq":25,"time":1785730508426,"data":{"turn":1,"step":2,"callId":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}} -{"type":"tool/result","seq":26,"time":1785730508537,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"bash-2"},"content":[{"type":"tool-result","toolCallId":"bash-2","content":[{"type":"text","text":"COUNT=2 CWD=/tmp"}],"isError":false}],"role":"user","id":"6e3ad5e1-1149-44d5-bd20-d9cc0139c747"}},"sourceEventSeqs":[25],"surfaceOp":"append"} -{"type":"step/end","seq":27,"time":1785730508537,"data":{"turn":1,"step":2}} -{"type":"step/start","seq":28,"time":1785730508537,"data":{"turn":1,"step":3}} -{"type":"assistant/chunk","seq":29,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":30,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-create","name":"str_replace_editor","argumentsDelta":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}} -{"type":"assistant/chunk","seq":31,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}} -{"type":"assistant/chunk","seq":32,"time":1785498592812,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":33,"time":1785730508538,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":34,"time":1785730508538,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e5769b2d-ea91-42fe-a78f-2f7f408f545e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[29,30,31,32,33],"surfaceOp":"append"} -{"type":"tool/call","seq":35,"time":1785730508538,"data":{"turn":1,"step":3,"callId":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}} -{"type":"tool/result","seq":36,"time":1785730508551,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"editor-create"},"content":[{"type":"tool-result","toolCallId":"editor-create","content":[{"type":"text","text":"New file created successfully at: {{cwd}}/note.txt"}],"isError":false}],"role":"user","id":"af41060c-7007-4ada-89d6-8b15a0e8be7c"}},"sourceEventSeqs":[35],"surfaceOp":"append"} -{"type":"step/end","seq":37,"time":1785730508551,"data":{"turn":1,"step":3}} -{"type":"step/start","seq":38,"time":1785730508552,"data":{"turn":1,"step":4}} -{"type":"assistant/chunk","seq":39,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":40,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-view","name":"str_replace_editor","argumentsDelta":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}} -{"type":"assistant/chunk","seq":41,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}} -{"type":"assistant/chunk","seq":42,"time":1785498592825,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":43,"time":1785730508552,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":44,"time":1785730508552,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0e9afabb-10a6-444c-ae22-fcdbb5e14695"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[39,40,41,42,43],"surfaceOp":"append"} -{"type":"tool/call","seq":45,"time":1785730508553,"data":{"turn":1,"step":4,"callId":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}} -{"type":"tool/result","seq":46,"time":1785730508554,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"editor-view"},"content":[{"type":"tool-result","toolCallId":"editor-view","content":[{"type":"text","text":"Here's the content of {{cwd}}/note.txt with line numbers (which has a total of 3 lines):\n 1 target:\n 2 \told\n 3 \n"}],"isError":false}],"role":"user","id":"a4472b37-6311-4880-bce2-cc369f9bc34b"}},"sourceEventSeqs":[45],"surfaceOp":"append"} -{"type":"step/end","seq":47,"time":1785730508554,"data":{"turn":1,"step":4}} -{"type":"step/start","seq":48,"time":1785730508554,"data":{"turn":1,"step":5}} -{"type":"assistant/chunk","seq":49,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":50,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-replace","name":"str_replace_editor","argumentsDelta":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}} -{"type":"assistant/chunk","seq":51,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}} -{"type":"assistant/chunk","seq":52,"time":1785498592826,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":53,"time":1785730508555,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":54,"time":1785730508555,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ba189070-d46e-461e-969b-9bca032bb154"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[49,50,51,52,53],"surfaceOp":"append"} -{"type":"tool/call","seq":55,"time":1785730508555,"data":{"turn":1,"step":5,"callId":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}} -{"type":"tool/result","seq":56,"time":1785730508564,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"editor-replace"},"content":[{"type":"tool-result","toolCallId":"editor-replace","content":[{"type":"text","text":"The file {{cwd}}/note.txt has been edited successfully."}],"isError":false}],"role":"user","id":"17331db9-174b-4699-9c9e-3140921956c4"}},"sourceEventSeqs":[55],"surfaceOp":"append"} -{"type":"step/end","seq":57,"time":1785730508564,"data":{"turn":1,"step":5}} -{"type":"step/start","seq":58,"time":1785730508565,"data":{"turn":1,"step":6}} -{"type":"assistant/chunk","seq":59,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} -{"type":"assistant/chunk","seq":60,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-exit","name":"bash","argumentsDelta":"{\"command\":\"exit 9\"}"}}} -{"type":"assistant/chunk","seq":61,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}} -{"type":"assistant/chunk","seq":62,"time":1785498592838,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":63,"time":1785730508565,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} -{"type":"assistant/message","seq":64,"time":1785730508565,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"7fa07d0f-e70a-460d-b685-bf8a63b6a8a0"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[59,60,61,62,63],"surfaceOp":"append"} -{"type":"tool/call","seq":65,"time":1785730508565,"data":{"turn":1,"step":6,"callId":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}} -{"type":"tool/result","seq":66,"time":1785730508641,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"bash-exit"},"content":[{"type":"tool-result","toolCallId":"bash-exit","content":[{"type":"text","text":"exit\n[shell exited: code 9]\nThe persistent bash shell was reset; the next bash call starts from the workspace with a fresh current directory and environment."}],"isError":false}],"role":"user","id":"ccb91a28-4034-49bf-967d-450f68f7f9b8"}},"sourceEventSeqs":[65],"surfaceOp":"append"} -{"type":"step/end","seq":67,"time":1785730508641,"data":{"turn":1,"step":6}} -{"type":"step/start","seq":68,"time":1785730508642,"data":{"turn":1,"step":7}} -{"type":"assistant/chunk","seq":69,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} -{"type":"assistant/chunk","seq":70,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"text-delta","index":0,"text":"PERSISTENT_TOOLS_OK"}}} -{"type":"assistant/chunk","seq":71,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PERSISTENT_TOOLS_OK"}}}} -{"type":"assistant/chunk","seq":72,"time":1785498592934,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} -{"type":"assistant/chunk","seq":73,"time":1785730508642,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} -{"type":"assistant/message","seq":74,"time":1785730508642,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"PERSISTENT_TOOLS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"43efc58a-46a1-4813-995f-1dc489438942"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[69,70,71,72,73],"surfaceOp":"append"} -{"type":"step/end","seq":75,"time":1785730508642,"data":{"turn":1,"step":7}} -{"type":"turn/end","seq":76,"time":1785730508642,"data":{"turn":1,"reason":{"kind":"completed"}}} +{"type":"session/title","seq":5,"time":1785498592368,"data":{"title":"Prove that bash state persists.","messageSeqs":[4],"source":{"kind":"fallback"}}} +{"type":"request/header","seq":6,"time":1786547087174,"data":{"header":{"config":{"provider":"deepseek-official","model":"deepseek-v4-flash"},"system":"{{system}}","tools":"{{tools}}"},"reason":"initial"}} +{"type":"request/context","seq":7,"time":1786547087175,"data":{"provider":"deepseek-official","model":"deepseek-v4-flash"}} +{"type":"assistant/chunk","seq":8,"time":1786547087175,"data":{"turn":1,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":9,"time":1786547087175,"data":{"turn":1,"step":1,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-1","name":"bash","argumentsDelta":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}} +{"type":"assistant/chunk","seq":10,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} +{"type":"assistant/chunk","seq":11,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":12,"time":1785331618326,"data":{"turn":1,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":13,"time":1786547087175,"data":{"turn":1,"step":1,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0d064526-8eff-482d-8525-ac478e1d1791"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[8,9,10,11,12],"surfaceOp":"append"} +{"type":"tool/call","seq":14,"time":1786547087176,"data":{"turn":1,"step":1,"callId":"bash-1","name":"bash","arguments":"{\"command\":\"cd /tmp && export DSH_EXAMPLE_COUNT=1 && printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}} +{"type":"tool/result","seq":15,"time":1786547094340,"data":{"turn":1,"step":1,"message":{"source":{"kind":"tool","callId":"bash-1"},"content":[{"type":"tool-result","toolCallId":"bash-1","content":[{"type":"text","text":"COUNT=1 CWD=/tmp"}],"isError":false}],"role":"user","id":"2c01f81a-01ea-47e2-bf92-f7825b7cc69f"}},"sourceEventSeqs":[14],"surfaceOp":"append"} +{"type":"step/end","seq":16,"time":1786547094340,"data":{"turn":1,"step":1}} +{"type":"step/start","seq":17,"time":1786547094341,"data":{"turn":1,"step":2}} +{"type":"assistant/chunk","seq":18,"time":1786547094341,"data":{"turn":1,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":19,"time":1786547094341,"data":{"turn":1,"step":2,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-2","name":"bash","argumentsDelta":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}} +{"type":"assistant/chunk","seq":20,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}}}} +{"type":"assistant/chunk","seq":21,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":22,"time":1785331618652,"data":{"turn":1,"step":2,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":23,"time":1786547094341,"data":{"turn":1,"step":2,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ba4078ce-0e18-419a-b720-339918aecf26"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[18,19,20,21,22],"surfaceOp":"append"} +{"type":"tool/call","seq":24,"time":1786547094341,"data":{"turn":1,"step":2,"callId":"bash-2","name":"bash","arguments":"{\"command\":\"DSH_EXAMPLE_COUNT=$((DSH_EXAMPLE_COUNT + 1)); printf \\\"COUNT=%s CWD=%s\\\\n\\\" \\\"$DSH_EXAMPLE_COUNT\\\" \\\"$PWD\\\"\"}"}} +{"type":"tool/result","seq":25,"time":1786547097878,"data":{"turn":1,"step":2,"message":{"source":{"kind":"tool","callId":"bash-2"},"content":[{"type":"tool-result","toolCallId":"bash-2","content":[{"type":"text","text":"COUNT=2 CWD=/tmp"}],"isError":false}],"role":"user","id":"6e3ad5e1-1149-44d5-bd20-d9cc0139c747"}},"sourceEventSeqs":[24],"surfaceOp":"append"} +{"type":"step/end","seq":26,"time":1786547097878,"data":{"turn":1,"step":2}} +{"type":"step/start","seq":27,"time":1786547097878,"data":{"turn":1,"step":3}} +{"type":"assistant/chunk","seq":28,"time":1786547097878,"data":{"turn":1,"step":3,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":29,"time":1786547097878,"data":{"turn":1,"step":3,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-create","name":"str_replace_editor","argumentsDelta":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}} +{"type":"assistant/chunk","seq":30,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}}}} +{"type":"assistant/chunk","seq":31,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":32,"time":1785331618762,"data":{"turn":1,"step":3,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":33,"time":1786547097878,"data":{"turn":1,"step":3,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"e5769b2d-ea91-42fe-a78f-2f7f408f545e"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[28,29,30,31,32],"surfaceOp":"append"} +{"type":"tool/call","seq":34,"time":1786547097879,"data":{"turn":1,"step":3,"callId":"editor-create","name":"str_replace_editor","arguments":"{\"command\":\"create\",\"path\":\"{{cwd}}/note.txt\",\"file_text\":\"target:\\n\\told\\n\"}"}} +{"type":"tool/result","seq":35,"time":1786547097885,"data":{"turn":1,"step":3,"message":{"source":{"kind":"tool","callId":"editor-create"},"content":[{"type":"tool-result","toolCallId":"editor-create","content":[{"type":"text","text":"New file created successfully at: {{cwd}}/note.txt"}],"isError":false}],"role":"user","id":"af41060c-7007-4ada-89d6-8b15a0e8be7c"}},"sourceEventSeqs":[34],"surfaceOp":"append"} +{"type":"step/end","seq":36,"time":1786547097885,"data":{"turn":1,"step":3}} +{"type":"step/start","seq":37,"time":1786547097885,"data":{"turn":1,"step":4}} +{"type":"assistant/chunk","seq":38,"time":1786547097886,"data":{"turn":1,"step":4,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":39,"time":1786547097886,"data":{"turn":1,"step":4,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-view","name":"str_replace_editor","argumentsDelta":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}} +{"type":"assistant/chunk","seq":40,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}}}} +{"type":"assistant/chunk","seq":41,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":42,"time":1785331618784,"data":{"turn":1,"step":4,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":43,"time":1786547097886,"data":{"turn":1,"step":4,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"0e9afabb-10a6-444c-ae22-fcdbb5e14695"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[38,39,40,41,42],"surfaceOp":"append"} +{"type":"tool/call","seq":44,"time":1786547097886,"data":{"turn":1,"step":4,"callId":"editor-view","name":"str_replace_editor","arguments":"{\"command\":\"view\",\"path\":\"{{cwd}}/note.txt\"}"}} +{"type":"tool/result","seq":45,"time":1786547097887,"data":{"turn":1,"step":4,"message":{"source":{"kind":"tool","callId":"editor-view"},"content":[{"type":"tool-result","toolCallId":"editor-view","content":[{"type":"text","text":"Here's the content of {{cwd}}/note.txt with line numbers (which has a total of 3 lines):\n 1 target:\n 2 \told\n 3 \n"}],"isError":false}],"role":"user","id":"a4472b37-6311-4880-bce2-cc369f9bc34b"}},"sourceEventSeqs":[44],"surfaceOp":"append"} +{"type":"step/end","seq":46,"time":1786547097887,"data":{"turn":1,"step":4}} +{"type":"step/start","seq":47,"time":1786547097887,"data":{"turn":1,"step":5}} +{"type":"assistant/chunk","seq":48,"time":1786547097887,"data":{"turn":1,"step":5,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":49,"time":1786547097887,"data":{"turn":1,"step":5,"chunk":{"type":"tool-call-delta","index":0,"id":"editor-replace","name":"str_replace_editor","argumentsDelta":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}} +{"type":"assistant/chunk","seq":50,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}}}} +{"type":"assistant/chunk","seq":51,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":52,"time":1785331618801,"data":{"turn":1,"step":5,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":53,"time":1786547097887,"data":{"turn":1,"step":5,"message":{"role":"assistant","content":[{"type":"tool-call","id":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"ba189070-d46e-461e-969b-9bca032bb154"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[48,49,50,51,52],"surfaceOp":"append"} +{"type":"tool/call","seq":54,"time":1786547097887,"data":{"turn":1,"step":5,"callId":"editor-replace","name":"str_replace_editor","arguments":"{\"command\":\"str_replace\",\"path\":\"{{cwd}}/note.txt\",\"old_str\":\"\\told\",\"new_str\":\"\\tnew\"}"}} +{"type":"tool/result","seq":55,"time":1786547097892,"data":{"turn":1,"step":5,"message":{"source":{"kind":"tool","callId":"editor-replace"},"content":[{"type":"tool-result","toolCallId":"editor-replace","content":[{"type":"text","text":"The file {{cwd}}/note.txt has been edited successfully."}],"isError":false}],"role":"user","id":"17331db9-174b-4699-9c9e-3140921956c4"}},"sourceEventSeqs":[54],"surfaceOp":"append"} +{"type":"step/end","seq":56,"time":1786547097892,"data":{"turn":1,"step":5}} +{"type":"step/start","seq":57,"time":1786547097892,"data":{"turn":1,"step":6}} +{"type":"assistant/chunk","seq":58,"time":1786547097893,"data":{"turn":1,"step":6,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}} +{"type":"assistant/chunk","seq":59,"time":1786547097893,"data":{"turn":1,"step":6,"chunk":{"type":"tool-call-delta","index":0,"id":"bash-exit","name":"bash","argumentsDelta":"{\"command\":\"exit 9\"}"}}} +{"type":"assistant/chunk","seq":60,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}}}} +{"type":"assistant/chunk","seq":61,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":62,"time":1785331618804,"data":{"turn":1,"step":6,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}} +{"type":"assistant/message","seq":63,"time":1786547097893,"data":{"turn":1,"step":6,"message":{"role":"assistant","content":[{"type":"tool-call","id":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"7fa07d0f-e70a-460d-b685-bf8a63b6a8a0"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[58,59,60,61,62],"surfaceOp":"append"} +{"type":"tool/call","seq":64,"time":1786547097893,"data":{"turn":1,"step":6,"callId":"bash-exit","name":"bash","arguments":"{\"command\":\"exit 9\"}"}} +{"type":"tool/result","seq":65,"time":1786547098003,"data":{"turn":1,"step":6,"message":{"source":{"kind":"tool","callId":"bash-exit"},"content":[{"type":"tool-result","toolCallId":"bash-exit","content":[{"type":"text","text":"exit\n[shell exited: code 9]\nThe persistent bash shell was reset; the next bash call starts from the workspace with a fresh current directory and environment."}],"isError":false}],"role":"user","id":"ccb91a28-4034-49bf-967d-450f68f7f9b8"}},"sourceEventSeqs":[64],"surfaceOp":"append"} +{"type":"step/end","seq":66,"time":1786547098003,"data":{"turn":1,"step":6}} +{"type":"step/start","seq":67,"time":1786547098003,"data":{"turn":1,"step":7}} +{"type":"assistant/chunk","seq":68,"time":1786547098003,"data":{"turn":1,"step":7,"chunk":{"type":"block-start","index":0,"blockType":"text"}}} +{"type":"assistant/chunk","seq":69,"time":1786547098003,"data":{"turn":1,"step":7,"chunk":{"type":"text-delta","index":0,"text":"PERSISTENT_TOOLS_OK"}}} +{"type":"assistant/chunk","seq":70,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"PERSISTENT_TOOLS_OK"}}}} +{"type":"assistant/chunk","seq":71,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}} +{"type":"assistant/chunk","seq":72,"time":1785331618807,"data":{"turn":1,"step":7,"chunk":{"type":"finish","reason":{"kind":"stop"}}}} +{"type":"assistant/message","seq":73,"time":1786547098003,"data":{"turn":1,"step":7,"message":{"role":"assistant","content":[{"type":"text","text":"PERSISTENT_TOOLS_OK"}],"source":{"kind":"model","provider":"deepseek-official","model":"deepseek-v4-flash"},"id":"43efc58a-46a1-4813-995f-1dc489438942"},"usage":{"inputTokens":3,"outputTokens":3}},"sourceEventSeqs":[68,69,70,71,72],"surfaceOp":"append"} +{"type":"step/end","seq":74,"time":1786547098003,"data":{"turn":1,"step":7}} +{"type":"turn/end","seq":75,"time":1786547098004,"data":{"turn":1,"reason":{"kind":"completed"}}} diff --git a/packages/core/system-prompt/README.i18n.yaml b/packages/core/system-prompt/README.i18n.yaml index 3b064f7d56..7420408ed2 100644 --- a/packages/core/system-prompt/README.i18n.yaml +++ b/packages/core/system-prompt/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/system-prompt/README.md -README.md: 01af5d3efd9233172dc251072fd2494cc80cbe51 -README.zh.md: e5e8cc25c4f84d88608c30c75050832caf262e8f +README.md: d750a507e628e7609af542227e4528d4d4934ce8 +README.zh.md: 36ee94349b63660ed52eedf7ba098334a6db07fb diff --git a/packages/core/system-prompt/README.md b/packages/core/system-prompt/README.md index 01af5d3efd..d750a507e6 100644 --- a/packages/core/system-prompt/README.md +++ b/packages/core/system-prompt/README.md @@ -9,6 +9,7 @@ System prompt assembly registry. Plugins contribute ordered sections, tool schem | Key | Default | Meaning | |---|---|---| | `includeHarnessIdentity` | `true` | Include the fixed `You are an AI agent powered by DeepSeek Harness.` order-−100 opener. Set false only when a compatibility deployment owns the complete system prompt. | +| `includeRuntimeContext` | `true` | Include ordered dynamic contexts in assembly. When false, context providers are not evaluated and contexts added by `system-prompt/assemble` listeners are discarded after the waterfall; other services and their enforcement remain active. | | `persona` | `''` | The global deployment-persona default: the ONE config-authored prompt fragment, rendered as the order-0 `deployment:persona` section unless an agent-scoped contribution shadows it. A template — complete `{{…}}` groups are interpreted strictly against the registered variables (the shipped loop registers `{{model}}`/`{{cwd}}`), with no escape syntax for literal braces yet. Empty ⇒ the section is dropped at render. | | `toolOrder` | — | Explicit model-facing tool order, as a list of `ToolSchema.name`s with one `''` rest entry (`TOOL_ORDER_REST`): listed tools take their listed position, unlisted tools land at the rest entry in lexicographic name order. Absent ⇒ plain lexicographic name order. Applied to the collected tools BEFORE the `system-prompt/assemble` waterfall — like the sections' `order` sort, it canonicalizes what the registry contributed (registration order is a plugin-load artifact), and a waterfall listener that mutates the list owns the determinism of what it emits. Misconfiguration fails loud: a list without exactly one rest entry, or with duplicates, throws at load; a listed name with no registered tool rejects every `assemble()`; a tool provider returning the reserved rest-entry name also rejects. Under the shipped loop the turn fails before any model request. Why a central list and not per-plugin weights: [Explicit model-facing tool order](../../../.agents/notes/implemented/feature/2026-07-06-explicit-tool-order.md). | @@ -17,9 +18,11 @@ System prompt assembly registry. Plugins contribute ordered sections, tool schem ### Public API - `ctx.systemPrompt.section(section: PromptSection): () => void` Contribute a section. The layer is the calling context's scope: `agent.ctx` contributes to that agent alone, shadowing a same-named global section there. A `complete: true` section becomes the exact complete prompt after the assembly waterfall; more than one effective complete section rejects assembly. Duplicate names within one layer and non-finite orders throw. Disposed with the calling fiber. +- `ctx.systemPrompt.context(context: PromptContext): () => void` Contribute ordered dynamic context for the calling scope. Providers are evaluated for each eligible assembly and become a sourced runtime-context snapshot in model history under the shipped loop. +- `ctx.systemPrompt.suppressRuntimeContext(): () => void` Suppress every dynamic-context contribution for the calling scope. Multiple registrations compose independently; disposing the returned effect restores context when no suppressor remains. - `ctx.systemPrompt.tools(provider: (context: AssembleContext) => ToolProviderResult): () => void` Contribute tool schemas, evaluated at each assembly with that assembly's context. `ToolProviderResult` = `{ schemas, knownNames? }`: `schemas` is the post-restriction visible set; `knownNames` is the pre-restriction universe used by `toolOrder`. A provider must not return a schema named `TOOL_ORDER_REST`. Scoped providers are consulted only for their scope's assemblies. Disposed with the calling fiber. - `ctx.systemPrompt.variable(name: string, provider: (context) => string | undefined): () => void` Contribute a prompt variable, referenced from section text as `{{name}}`. Scoped variables shadow a same-named global for that agent. Duplicate-in-layer or unreferenceable names throw; `undefined` means "no value for this assembly". Disposed with the calling fiber. -- `ctx.systemPrompt.assemble(context?: AssembleContext): Promise` Assemble the prompt for one caller: the global layer merged with `context.scope`'s layer, with tool schemas detached before the transform waterfall. Runs through the scope-filtered `system-prompt/assemble` waterfall, then restores an effective complete section as the sole prompt section. An optional `context.signal` explicitly controls this assembly request; providers and listeners may cooperate with it but must not retain it for another turn. Rejects for multiple complete sections, when a configured `toolOrder` names a tool outside the providers' `knownNames` universe, or when a provider returns the reserved rest-entry name. +- `ctx.systemPrompt.assemble(context?: AssembleContext): Promise` Assemble the prompt for one caller: the global layer merged with `context.scope`'s layer, with tool schemas detached before the transform waterfall. Runs through the scope-filtered `system-prompt/assemble` waterfall, then restores an effective complete section as the sole prompt section and enforces any active runtime-context suppressor. An optional `context.signal` explicitly controls this assembly request; providers and listeners may cooperate with it but must not retain it for another turn. Rejects for multiple complete sections, when a configured `toolOrder` names a tool outside the providers' `knownNames` universe, or when a provider returns the reserved rest-entry name. ### Live events @@ -49,7 +52,7 @@ Design rationale: [the prompt-variables Agent Note](../../../.agents/notes/imple #### What the model sees -By default every assembly starts with the harness identity below, then the configured persona and ordered plugin sections after strict variable interpolation. `includeHarnessIdentity: false` omits only that fixed opener. Empty sections disappear; scoped sections and variables can shadow globals for one agent. The `system-prompt/assemble` waterfall determines the delivered prompt and tool schemas unless one effective section declares itself complete; that exact section then becomes the whole system prompt while the waterfall's contexts, tools, and variables remain. +By default every assembly starts with the harness identity below, then the configured persona and ordered plugin sections after strict variable interpolation. `includeHarnessIdentity: false` omits only that fixed opener. Empty sections disappear; scoped sections and variables can shadow globals for one agent. The `system-prompt/assemble` waterfall determines the delivered prompt and tool schemas unless one effective section declares itself complete; that exact section then becomes the whole system prompt while the waterfall's contexts, tools, and variables remain. Ordered dynamic contexts are separate from system-prompt sections and become sourced user-role snapshots only when present. `includeRuntimeContext: false` or a scoped suppressor removes all such contexts, including listener additions, without disabling the services that own the underlying policy or state. ##### Harness identity diff --git a/packages/core/system-prompt/README.zh.md b/packages/core/system-prompt/README.zh.md index e5e8cc25c4..36ee94349b 100644 --- a/packages/core/system-prompt/README.zh.md +++ b/packages/core/system-prompt/README.zh.md @@ -9,6 +9,7 @@ | 键 | 默认值 | 含义 | |---|---|---| | `includeHarnessIdentity` | `true` | 是否包含顺序为 −100 的固定开场白 `You are an AI agent powered by DeepSeek Harness.`。仅当兼容部署拥有完整系统提示词时设为 false。 | +| `includeRuntimeContext` | `true` | 是否在组装中包含有序动态上下文。设为 false 时不会求值上下文提供方，并会在 waterfall 后丢弃 `system-prompt/assemble` 监听器添加的上下文；其他服务及其强制机制仍然生效。 | | `persona` | `''` | 全局部署 persona 默认值：唯一由配置提供的提示词片段，渲染为顺序为 0 的 `deployment:persona` 段，除非 agent 作用域的贡献将其遮蔽。它是模板，完整的 `{{…}}` 组会严格按已注册变量解释（随附循环注册 `{{model}}`/`{{cwd}}`），目前没有表达字面量花括号的转义语法。为空 ⇒ 渲染时删除该段。 | | `toolOrder` | 无 | 显式的面向模型工具顺序：一个 `ToolSchema.name` 列表，包含一个 `''` 其余项（`TOOL_ORDER_REST`）。已列工具占据列出的位置；未列工具按名称字典序落在其余项位置。缺席 ⇒ 直接按名称字典序排列。在 `system-prompt/assemble` waterfall（瀑布式事件）之前应用于已收集工具；与段的 `order` 排序一样，它会规范化注册表贡献的内容（注册顺序是插件加载产物），而修改列表的 waterfall 监听器拥有其输出的确定性。配置错误会明确失败：列表没有恰好一个其余项或存在重复项，会在加载时抛出；已列名称没有对应已注册工具，会使每次 `assemble()` 被拒绝；工具提供方返回保留的其余项名称也会被拒绝。在随附循环下，轮次会在任何模型请求前失败。为何采用中心列表而非每插件权重，见[显式面向模型工具顺序](../../../.agents/notes/implemented/feature/2026-07-06-explicit-tool-order.md)。 | @@ -17,9 +18,11 @@ ### 公开 API - `ctx.systemPrompt.section(section: PromptSection): () => void`：贡献一个段。层由调用上下文的作用域决定：`agent.ctx` 只为该 agent 贡献，并在该处遮蔽同名全局段。一个 `complete: true` 段会在组装 waterfall 之后成为精确的完整提示词；有效 complete 段超过一个时，组装会被拒绝。同一层中的重复名称和非有限顺序会抛出。随调用 fiber 一并 dispose（资源释放）。 +- `ctx.systemPrompt.context(context: PromptContext): () => void`：为调用作用域贡献有序动态上下文。每次符合条件的组装都会求值提供方，并在随附循环下成为模型历史中带来源的 runtime-context 快照。 +- `ctx.systemPrompt.suppressRuntimeContext(): () => void`：抑制调用作用域的所有动态上下文贡献。多个注册会独立组合；只有当不再存在抑制器时，dispose 返回的 effect 才会恢复上下文。 - `ctx.systemPrompt.tools(provider: (context: AssembleContext) => ToolProviderResult): () => void`：贡献工具 schema；每次组装时使用该次组装的上下文求值。`ToolProviderResult` = `{ schemas, knownNames? }`：`schemas` 是限制后的可见集合；`knownNames` 是限制前由 `toolOrder` 使用的全集。提供方不得返回名为 `TOOL_ORDER_REST` 的 schema。带作用域提供方只在其作用域的组装中查询。随调用 fiber 一并 dispose。 - `ctx.systemPrompt.variable(name: string, provider: (context) => string | undefined): () => void`：贡献提示词变量，在段文本中以 `{{name}}` 引用。带作用域变量会为该 agent 遮蔽同名全局变量。同层重复或无法引用的名称会抛出；`undefined` 表示「本次组装没有值」。随调用 fiber 一并 dispose。 -- `ctx.systemPrompt.assemble(context?: AssembleContext): Promise`：为一个调用方组装提示词：将全局层与 `context.scope` 的层合并，并在变换 waterfall 前分离工具 schema。它经过按作用域筛选的 `system-prompt/assemble` waterfall，之后将一个有效的 complete 段恢复为唯一的提示词段落。可选的 `context.signal` 显式控制本次组装请求；提供方与监听器可以配合该信号，但不得将它保留给另一轮次。存在多个 complete 段、已配置的 `toolOrder` 指名提供方 `knownNames` 全集以外的工具，或提供方返回保留的其余项名称时，调用会被拒绝。 +- `ctx.systemPrompt.assemble(context?: AssembleContext): Promise`：为一个调用方组装提示词：将全局层与 `context.scope` 的层合并，并在变换 waterfall 前分离工具 schema。它经过按作用域筛选的 `system-prompt/assemble` waterfall，之后将一个有效的 complete 段恢复为唯一的提示词段落，并实施任何活动的 runtime-context 抑制器。可选的 `context.signal` 显式控制本次组装请求；提供方与监听器可以配合该信号，但不得将它保留给另一轮次。存在多个 complete 段、已配置的 `toolOrder` 指名提供方 `knownNames` 全集以外的工具，或提供方返回保留的其余项名称时，调用会被拒绝。 @@ -51,7 +54,7 @@ #### 模型看到的内容 -默认情况下，每次组装都从下方 harness 身份开始，然后在严格变量插值后追加已配置 persona 与有序插件段。`includeHarnessIdentity: false` 仅省略这个固定开场白。空段会消失；带作用域的段和变量可以为一个 agent 遮蔽全局项。`system-prompt/assemble` waterfall 决定交付的提示词与工具 schema，除非一个有效段声明自身为 complete；此时，该确切段落会成为完整的系统提示词，而 waterfall 得到的上下文、工具和变量保持不变。 +默认情况下，每次组装都从下方 harness 身份开始，然后在严格变量插值后追加已配置 persona 与有序插件段。`includeHarnessIdentity: false` 仅省略这个固定开场白。空段会消失；带作用域的段和变量可以为一个 agent 遮蔽全局项。`system-prompt/assemble` waterfall 决定交付的提示词与工具 schema，除非一个有效段声明自身为 complete；此时，该确切段落会成为完整的系统提示词，而 waterfall 得到的上下文、工具和变量保持不变。有序动态上下文与系统提示词段落分离，只在存在时才会成为带来源的 user 角色快照。`includeRuntimeContext: false` 或带作用域的抑制器会移除所有这类上下文，包括监听器添加的内容，但不会禁用拥有底层策略或状态的服务。 ##### harness 身份 diff --git a/packages/core/system-prompt/src/index.ts b/packages/core/system-prompt/src/index.ts index 5078119224..ffc052e0b9 100644 --- a/packages/core/system-prompt/src/index.ts +++ b/packages/core/system-prompt/src/index.ts @@ -186,6 +186,8 @@ function compareToolNames(a: ToolSchema, b: ToolSchema): number { export interface Config { /** Include the fixed DeepSeek Harness identity before the deployment persona (default true). */ includeHarnessIdentity?: boolean + /** Include dynamic runtime-context snapshots in model history (default true). */ + includeRuntimeContext?: boolean /** * Deployment-wide order-0 persona template. A scoped section named * `deployment:persona` shadows it; `{{variable}}` references are strict. @@ -302,6 +304,7 @@ type VariableProvider = (context: AssembleContext) => string | undefined class PromptLayer implements ScopeLayer { readonly sections: NamedEntries readonly contexts: NamedEntries + readonly runtimeContextSuppressors = new AnonymousEntries() readonly toolProviders = new AnonymousEntries() readonly variables: NamedEntries @@ -325,6 +328,7 @@ class PromptLayer implements ScopeLayer { isEmpty(): boolean { return this.sections.isEmpty() && this.contexts.isEmpty() + && this.runtimeContextSuppressors.isEmpty() && this.toolProviders.isEmpty() && this.variables.isEmpty() } @@ -334,6 +338,7 @@ class PromptLayer implements ScopeLayer { export class SystemPrompt extends Service { static Config: z = z.object({ includeHarnessIdentity: z.boolean().default(true), + includeRuntimeContext: z.boolean().default(true), persona: z.string().default(''), // Preserve omission because an explicit empty order lacks the rest marker. toolOrder: z.array(z.string()).default(undefined as unknown as string[]), @@ -362,6 +367,7 @@ export class SystemPrompt extends Service { // The fallback narrows the optional input type; the schema already defaults it. text: config.persona ?? '', }) + if (!(config.includeRuntimeContext ?? true)) this.suppressRuntimeContext() } /** @@ -400,6 +406,20 @@ export class SystemPrompt extends Service { ) } + /** + * Suppress every dynamic runtime-context contribution in the calling + * context's scope without changing the services that own or enforce those + * facts. Multiple suppressors remain independently disposable. + * @returns the exact Cordis effect disposer. + */ + suppressRuntimeContext(): () => void { + return this.layers.effect( + this.ctx, + layer => layer.runtimeContextSuppressors.append(true), + { label: 'systemPrompt.suppressRuntimeContext()' }, + ) + } + /** * Register a tool-schema provider in the calling context's scope. Global and * matching scoped providers both contribute; returning the reserved @@ -446,13 +466,16 @@ export class SystemPrompt extends Service { // Keep configuration failures on the declared asynchronous error path. async assemble(context: AssembleContext = {}): Promise { const scope = context.scope + const scopeLayers = this.layers.chainLayers(scope) + const runtimeContextSuppressed = !this.layers.global.runtimeContextSuppressors.isEmpty() + || scopeLayers.some(layer => !layer.runtimeContextSuppressors.isEmpty()) // Scoped variables shadow globals. const variables: Record = {} for (const [name, provider] of this.layers.global.variables.entries()) { variables[name] = provider(context) } // Scope-chain variables, farthest first, so the nearest scope wins a name. - for (const layer of this.layers.chainLayers(scope)) { + for (const layer of scopeLayers) { for (const [name, provider] of layer.variables.entries()) { variables[name] = provider(context) } @@ -463,7 +486,7 @@ export class SystemPrompt extends Service { // Validate order against pre-restriction names while collecting visible schemas. const providers = [ ...this.layers.global.toolProviders.values(), - ...this.layers.chainLayers(scope).flatMap(layer => [...layer.toolProviders.values()]), + ...scopeLayers.flatMap(layer => [...layer.toolProviders.values()]), ] const collected: ToolSchema[] = [] const knownNames = new Set() @@ -495,12 +518,14 @@ export class SystemPrompt extends Service { }) const assembly: PromptAssembly = { sections, - contexts: [...contextByName.values()] - .sort((a, b) => a.order - b.order) - .map(entry => ({ - name: entry.name, - text: typeof entry.text === 'function' ? entry.text(context) : entry.text, - })), + contexts: runtimeContextSuppressed + ? [] + : [...contextByName.values()] + .sort((a, b) => a.order - b.order) + .map(entry => ({ + name: entry.name, + text: typeof entry.text === 'function' ? entry.text(context) : entry.text, + })), tools: orderTools(collected, this.toolOrder, knownNames), variables, } @@ -508,8 +533,12 @@ export class SystemPrompt extends Service { scopeTarget(this, scope), 'system-prompt/assemble', assembly, context, () => Promise.resolve(assembly), ) - if (completeSection === undefined) return transformed - return { ...transformed, sections: [completeSection] } + if (completeSection === undefined && !runtimeContextSuppressed) return transformed + return { + ...transformed, + sections: completeSection === undefined ? transformed.sections : [completeSection], + contexts: runtimeContextSuppressed ? [] : transformed.contexts, + } } } diff --git a/packages/core/system-prompt/tests/scoped.spec.ts b/packages/core/system-prompt/tests/scoped.spec.ts index 1498966cc6..360fef3a14 100644 --- a/packages/core/system-prompt/tests/scoped.spec.ts +++ b/packages/core/system-prompt/tests/scoped.spec.ts @@ -142,6 +142,23 @@ describe('scoped cache-safe context', () => { expect(renderContextSnapshot(await ctx.systemPrompt.assemble({ scope: scopeKeyOf(scope) }))) .toContain('global policy') }) + + it('suppresses all context for one scope and restores it when disposed', async () => { + const ctx = await mount() + const scope = await mintScope(ctx, 'suppressed-context') + const key = scopeKeyOf(scope) + ctx.systemPrompt.context({ name: 'policy', order: 1, text: 'global policy' }) + const dispose = scope.ctx.systemPrompt.suppressRuntimeContext() + + const suppressed = await ctx.systemPrompt.assemble({ scope: key }) + expect(suppressed.contexts).toEqual([]) + const global = await ctx.systemPrompt.assemble() + expect(renderContextSnapshot(global)).toContain('global policy') + + dispose() + expect(renderContextSnapshot(await ctx.systemPrompt.assemble({ scope: key }))) + .toContain('global policy') + }) }) describe('scoped tool providers and toolOrder × restriction', () => { diff --git a/packages/core/system-prompt/tests/system-prompt.spec.ts b/packages/core/system-prompt/tests/system-prompt.spec.ts index f7f5f36727..cf196892a7 100644 --- a/packages/core/system-prompt/tests/system-prompt.spec.ts +++ b/packages/core/system-prompt/tests/system-prompt.spec.ts @@ -49,6 +49,25 @@ describe('SystemPrompt', () => { expect(renderPrompt(assembly)).toBe('You are a helpful software engineer assistant.') }) + it('can suppress runtime context without evaluating providers or accepting waterfall additions', async () => { + const ctx = new Context() + await ctx.plugin(SystemPrompt, { includeRuntimeContext: false }) + let providerCalls = 0 + ctx.systemPrompt.context({ + name: 'policy', + order: 0, + text: () => `policy ${++providerCalls}`, + }) + ctx.on('system-prompt/assemble', async (assembly, _context, next) => { + assembly.contexts.push({ name: 'late', text: 'late context' }) + return next() + }) + + const assembly = await ctx.systemPrompt.assemble() + expect(assembly.contexts).toEqual([]) + expect(providerCalls).toBe(0) + }) + it('tolerates a schema-bypassing direct construction (persona omitted)', async () => { // ctx.plugin validates + defaults the config first; a direct construction // skips the schema, so the ctor's `?? ''` narrowing is what fires. diff --git a/packages/core/tools/README.i18n.yaml b/packages/core/tools/README.i18n.yaml index 20b3510645..ede5781a7a 100644 --- a/packages/core/tools/README.i18n.yaml +++ b/packages/core/tools/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/tools/README.md -README.md: 120931c9f4b4f5e1c39c3ddd8da4b8cae42dbe69 -README.zh.md: 92ea290e5f1fc2e66630a1ba63744d3b380d340a +README.md: 60841513ad5ad439ae8851dffc34e8d250acaa67 +README.zh.md: 6e26a802c2c16a9d48e29236d358b15eafddcc07 diff --git a/packages/core/tools/README.md b/packages/core/tools/README.md index 120931c9f4..60841513ad 100644 --- a/packages/core/tools/README.md +++ b/packages/core/tools/README.md @@ -155,7 +155,7 @@ Code Mode exposes the generated [`run_code` schema](../../../docs/tool-catalog.m ```markdown ## Writing code for run_code -Pass `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: +`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. Inside the program: - 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. - 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. diff --git a/packages/core/tools/README.zh.md b/packages/core/tools/README.zh.md index 92ea290e5f..6e26a802c2 100644 --- a/packages/core/tools/README.zh.md +++ b/packages/core/tools/README.zh.md @@ -155,7 +155,7 @@ Code Mode 会公开生成的 [`run_code` schema](../../../docs/tool-catalog.md#d ```markdown ## Writing code for run_code -Pass `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: +`run_code` takes two required arguments: `code` — the body of an async TypeScript function (erasable syntax only — no `enum` or namespaces; type annotations are advisory, the code runs type-stripped) — and `description`, a short summary of what the program does. Inside the program: - 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. - 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. diff --git a/packages/core/tools/src/code-mode.ts b/packages/core/tools/src/code-mode.ts index ab23021885..49c2c7a551 100644 --- a/packages/core/tools/src/code-mode.ts +++ b/packages/core/tools/src/code-mode.ts @@ -45,10 +45,11 @@ interface RunCodeFlavor { */ const TYPESCRIPT_FLAVOR: RunCodeFlavor = { 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.', + 'Execute a TypeScript program against the available tools. Takes two required ' + + 'arguments: `code`, the BODY of an async function (erasable syntax only; top-level ' + + '`await` and `return` work), and `description`, a short summary of what the program ' + + 'does. Call tools as `await tools.name(args)` per the declarations in the system ' + + 'prompt. Only what you print or return comes back — curate it.', codeDescription: 'The program: the body of an async TypeScript function.', } @@ -59,8 +60,9 @@ const TYPESCRIPT_FLAVOR: RunCodeFlavor = { */ const PYTHON_FLAVOR: RunCodeFlavor = { description: - 'Execute a Python program against the available tools. Write the BODY of an ' - + 'async function (top-level `await` and `return` work) and call tools as ' + 'Execute a Python program against the available tools. Takes two required ' + + 'arguments: `code`, the BODY of an async function (top-level `await` and `return` ' + + 'work), and `description`, a short summary of what the program does. Call tools as ' + '`await tools.name(args)` per the declarations in the system prompt. Answer ' + 'with `print(...)` and/or `return ` — only that comes back, so curate it.', codeDescription: 'The program: the body of an async Python function.', diff --git a/packages/core/tools/src/py-types.ts b/packages/core/tools/src/py-types.ts index 4898ec80e1..6141df14ed 100644 --- a/packages/core/tools/src/py-types.ts +++ b/packages/core/tools/src/py-types.ts @@ -733,7 +733,7 @@ export function jsonSchemaToPy(schema: unknown): string { /** The fixed model-facing usage contract rendered above the declarations. */ const SDK_INSTRUCTIONS = `## Writing code for run_code -Pass \`run_code\` the body of an async Python function (top-level \`await\` and \`return\` both work). At run time exactly two of the names declared below are bound: \`tools\` and \`ToolCallError\`. Everything else is a STATIC STUB describing argument and return types — in particular the \`TypedDict\` classes do NOT exist at run time, so build arguments as plain \`dict\`/\`list\` JSON values: \`await tools.name({"field": 1})\`, never \`FooArgs(field=1)\`, which raises \`NameError\`. Inside the program: +\`run_code\` takes two required arguments: \`code\` — the body of an async Python function (top-level \`await\` and \`return\` both work) — and \`description\`, a short summary of what the program does. At run time exactly two of the names declared below are bound: \`tools\` and \`ToolCallError\`. Everything else is a STATIC STUB describing argument and return types — in particular the \`TypedDict\` classes do NOT exist at run time, so build arguments as plain \`dict\`/\`list\` JSON values: \`await tools.name({"field": 1})\`, never \`FooArgs(field=1)\`, which raises \`NameError\`. Inside the program: - Call tools as \`await tools.name(args)\` — subscript access for exotic, reserved, or underscore-leading names: \`await tools["my-tool"](args)\`. Every call resolves to the tool's typed canonical JSON value (each method's return type below). Tool arguments must be lossless JSON. - A FAILED tool call raises \`ToolCallError\`, whose \`toolName\` identifies the failed tool and whose message is human-readable — wrap in \`try/except\` to handle and continue. diff --git a/packages/core/tools/src/ts-types.ts b/packages/core/tools/src/ts-types.ts index 9b0d096a22..bbd5cd7666 100644 --- a/packages/core/tools/src/ts-types.ts +++ b/packages/core/tools/src/ts-types.ts @@ -249,7 +249,7 @@ export function jsonSchemaToTs(schema: unknown, indent = 0): string { /** The fixed model-facing usage contract rendered above the declarations (see the Code Mode Agent Note's "What the model sees"). */ const SDK_INSTRUCTIONS = `## Writing code for run_code -Pass \`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: +\`run_code\` takes two required arguments: \`code\` — the body of an async TypeScript function (erasable syntax only — no \`enum\` or namespaces; type annotations are advisory, the code runs type-stripped) — and \`description\`, a short summary of what the program does. Inside the program: - 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. - 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. diff --git a/packages/core/tools/tests/code-mode.spec.ts b/packages/core/tools/tests/code-mode.spec.ts index 6a88a2e9d5..58c190dd26 100644 --- a/packages/core/tools/tests/code-mode.spec.ts +++ b/packages/core/tools/tests/code-mode.spec.ts @@ -399,6 +399,10 @@ describe('mode-aware wire contribution', () => { const runCodeSchema = assembly.tools.find(tool => tool.name === RUN_CODE_NAME) expect(runCodeSchema?.description).toContain('Execute a TypeScript program') expect(runCodeSchema?.description).toContain('BODY of an') + // Both required arguments are named here, not only in the parameter + // schema: prose that describes the call as "pass the program" is what + // leads a model to emit `{code}` alone and fail INVALID_ARGS. + expect(runCodeSchema?.description).toContain('`description`') const codeParam = (runCodeSchema?.parameters as { properties: { code: { description: string } } }).properties.code expect(codeParam.description).toBe('The program: the body of an async TypeScript function.') }) @@ -410,6 +414,7 @@ describe('mode-aware wire contribution', () => { const runCodeSchema = assembly.tools.find(tool => tool.name === RUN_CODE_NAME) expect(runCodeSchema?.description).toContain('Execute a Python program') expect(runCodeSchema?.description).toContain('`return `') + expect(runCodeSchema?.description).toContain('`description`') expect(runCodeSchema?.description).not.toContain('TypeScript') const codeParam = (runCodeSchema?.parameters as { properties: { code: { description: string } } }).properties.code expect(codeParam.description).toBe('The program: the body of an async Python function.') diff --git a/packages/core/tools/tests/py-types.spec.ts b/packages/core/tools/tests/py-types.spec.ts index 3439cb6a37..0caf9daab6 100644 --- a/packages/core/tools/tests/py-types.spec.ts +++ b/packages/core/tools/tests/py-types.spec.ts @@ -166,6 +166,15 @@ describe('renderToolsSdkPy', () => { expect(text).toContain('tools: Tools') }) + it('names both required call arguments, not just the program', () => { + // The schema requires `code` AND `description`; instructions that mention + // only the program let a model emit `{code}` alone and fail INVALID_ARGS. + const text = renderToolsSdkPy([bash]) + expect(text).toContain('`code`') + expect(text).toContain('`description`') + expect(text).toContain('two required arguments') + }) + it('renders required as plain fields and optional as NotRequired, with per-field description comments', () => { const tool: ToolSdkSchema = { name: 'search', diff --git a/packages/core/tools/tests/ts-types.spec.ts b/packages/core/tools/tests/ts-types.spec.ts index 4c4954752d..8b4e145060 100644 --- a/packages/core/tools/tests/ts-types.spec.ts +++ b/packages/core/tools/tests/ts-types.spec.ts @@ -148,6 +148,15 @@ describe('renderToolsSdk', () => { expect(text).toContain('lossless JSON') }) + it('names both required call arguments, not just the program', () => { + // The schema requires `code` AND `description`; instructions that mention + // only the program let a model emit `{code}` alone and fail INVALID_ARGS. + const text = renderToolsSdk([bash]) + expect(text).toContain('`code`') + expect(text).toContain('`description`') + expect(text).toContain('two required arguments') + }) + it('is deterministic: same tool set, byte-identical text regardless of input order', () => { expect(renderToolsSdk([bash, exotic])).toBe(renderToolsSdk([exotic, bash])) // Equal names sort stably (the comparator's equal arm). diff --git a/packages/examples/agent-spine-demo/README.i18n.yaml b/packages/examples/agent-spine-demo/README.i18n.yaml index 0ec8d38113..ac85098e2d 100644 --- a/packages/examples/agent-spine-demo/README.i18n.yaml +++ b/packages/examples/agent-spine-demo/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/examples/agent-spine-demo/README.md -README.md: 2d48b65e5b5cf08d62bac8b07967965fe38bdd77 -README.zh.md: 83df672b540649f58b112a08ba0792ca2cb72295 +README.md: e4e92d398d5f48609bc5af23ece2b5fe3e607b21 +README.zh.md: c5f78d53a59f3bd3e377afc2c471d187b73da32e diff --git a/packages/examples/agent-spine-demo/README.md b/packages/examples/agent-spine-demo/README.md index 2d48b65e5b..e4e92d398d 100644 --- a/packages/examples/agent-spine-demo/README.md +++ b/packages/examples/agent-spine-demo/README.md @@ -55,11 +55,11 @@ This applies the [Service Definition / Service provider / Consumer separation](. ```ts import type { Config } from '@deepseek-ai/dsh-agent-spine-demo' -// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, jobs?, toolJobs?, goals?, invariants? } +// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, includeRuntimeContext?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, jobs?, toolJobs?, goals?, invariants? } // workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults. ``` -The bundle forwards each field to the child that owns it. App packages supply any pre-created agents: headless and JSON-RPC compositions create `main`, while the ACP app creates agents on demand at `session/new`. Prompt, tool, title, skill, agent-instructions, invariant, goal, and task settings retain the schemas and defaults documented by their owning packages; `jobs.maxConcurrentJobsPerOwner` configures the local provider independently of the model-facing `toolJobs` controls. `pickSpineConfig()` copies only fields owned by this bundle, and conflicting `dshHome` values fail during composition. +The bundle forwards each field to the child that owns it. App packages supply any pre-created agents: headless and JSON-RPC compositions create `main`, while the ACP app creates agents on demand at `session/new`. `includeRuntimeContext: false` is forwarded to `dsh-system-prompt` and suppresses all dynamic context snapshots for fresh sessions without disabling their policy services. Prompt, tool, title, skill, agent-instructions, invariant, goal, and task settings retain the schemas and defaults documented by their owning packages; `jobs.maxConcurrentJobsPerOwner` configures the local provider independently of the model-facing `toolJobs` controls. `pickSpineConfig()` copies only fields owned by this bundle, and conflicting `dshHome` values fail during composition. For example, `{ invariants: { enabled: true, package_allowlist: ['^@deepseek-ai/dsh-'], package_blocklist: ['agent-loop$'] } }` keeps the package-owned companions mounted but suppresses the blocked owner. Blocklist matches override allowlist matches; see [`dsh-invariants`](../../runtime-diagnostics/invariants/README.md) for regex and lifecycle rules. diff --git a/packages/examples/agent-spine-demo/README.zh.md b/packages/examples/agent-spine-demo/README.zh.md index 83df672b54..c5f78d53a5 100644 --- a/packages/examples/agent-spine-demo/README.zh.md +++ b/packages/examples/agent-spine-demo/README.zh.md @@ -55,11 +55,11 @@ ```ts import type { Config } from '@deepseek-ai/dsh-agent-spine-demo' -// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, jobs?, toolJobs?, goals?, invariants? } +// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, includeRuntimeContext?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, jobs?, toolJobs?, goals?, invariants? } // workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults. ``` -组合包将每个字段转发给拥有它的子节点。应用包提供预创建的 agent：无头和 JSON-RPC 组合会创建 `main`，ACP 应用则在 `session/new` 按需创建 agent。提示词、工具、标题、skill、工作区上下文、不变式、目标和任务设置沿用其所属包记录的 schema 与默认值；`jobs.maxConcurrentJobsPerOwner` 配置本地 Service provider，并与面向模型的 `toolJobs` 控制工具相互独立。`pickSpineConfig()` 只复制该组合包拥有的字段，`dshHome` 值冲突会在组合时失败。 +组合包将每个字段转发给拥有它的子节点。应用包提供预创建的 agent：无头和 JSON-RPC 组合会创建 `main`，ACP 应用则在 `session/new` 按需创建 agent。`includeRuntimeContext: false` 会转发给 `dsh-system-prompt`，为新建会话抑制所有动态上下文快照，但不禁用其策略服务。提示词、工具、标题、skill、工作区上下文、不变式、目标和任务设置沿用其所属包记录的 schema 与默认值；`jobs.maxConcurrentJobsPerOwner` 配置本地 Service provider，并与面向模型的 `toolJobs` 控制工具相互独立。`pickSpineConfig()` 只复制该组合包拥有的字段，`dshHome` 值冲突会在组合时失败。 例如，`{ invariants: { enabled: true, package_allowlist: ['^@deepseek-ai/dsh-'], package_blocklist: ['agent-loop$'] } }` 会让包拥有的配套插件保持挂载，但抑制被阻止的拥有者。Blocklist 匹配优先于 allowlist 匹配；正则表达式与生命周期规则见 [`dsh-invariants`](../../runtime-diagnostics/invariants/README.md)。 diff --git a/packages/examples/agent-spine-demo/src/index.ts b/packages/examples/agent-spine-demo/src/index.ts index e00f2b498d..87098c38da 100644 --- a/packages/examples/agent-spine-demo/src/index.ts +++ b/packages/examples/agent-spine-demo/src/index.ts @@ -69,9 +69,10 @@ export interface GoalConfig { /** * Bundle config: each field forwarded verbatim to the child that owns it — * `agents` to the agent loop (an app that pre-creates no agents, like the ACP - * bridge, simply omits it), `includeHarnessIdentity`, `persona`, and `toolOrder` - * to the system-prompt plugin (the fixed opener, deployment persona, and explicit - * model-facing tool order), the `tools` object to the tool registry (its presentation `mode`), + * bridge, simply omits it), `includeHarnessIdentity`, `includeRuntimeContext`, + * `persona`, and `toolOrder` to the system-prompt plugin (the fixed opener, + * dynamic-context policy, deployment persona, and explicit model-facing tool + * order), the `tools` object to the tool registry (its presentation `mode`), * `dshHome` to bash environment and local skill discovery, `sessionTitle` to * the fallback title service, `skills` to the * skill registry/local provider/tool consumer, `workspaceContext` to the @@ -95,6 +96,8 @@ export interface Config { maxParallelToolCalls?: AgentLoopConfig['maxParallelToolCalls'] /** Whether the system prompt includes the fixed Harness identity (default true). */ includeHarnessIdentity?: SystemPromptConfig['includeHarnessIdentity'] + /** Whether model history includes dynamic runtime-context snapshots (default true). */ + includeRuntimeContext?: SystemPromptConfig['includeRuntimeContext'] /** The deployment persona (see dsh-system-prompt's `Config`). */ persona?: SystemPromptConfig['persona'] /** The explicit model-facing tool order (see dsh-system-prompt's `Config`). */ @@ -180,6 +183,7 @@ export function pickSpineConfig(config: Omit): Omit { it('can omit the bundled bash tool and Harness identity for a compatibility deployment', async () => { const ctx = await mount({ includeHarnessIdentity: false, + includeRuntimeContext: false, persona: 'You are a helpful software engineer assistant.', workspaceContext: false, skills: { enabled: false }, @@ -720,6 +721,8 @@ describe('dsh-agent-spine-demo bundle', () => { }, true) expect(ctx.tools.schemas()).toEqual([]) + ctx.systemPrompt.context({ name: 'policy', order: 0, text: 'hidden policy' }) + expect((await ctx.systemPrompt.assemble()).contexts).toEqual([]) expect(renderPrompt(await ctx.systemPrompt.assemble())) .toBe('You are a helpful software engineer assistant.') @@ -729,7 +732,9 @@ describe('dsh-agent-spine-demo bundle', () => { it('picks shared spine config without leaking entry-point fields', () => { const appConfig = { model: 'entrypoint-only', + maxParallelToolCalls: 3, includeHarnessIdentity: false, + includeRuntimeContext: false, persona: 'You are merged.', toolOrder: ['zulu'], tools: { mode: 'native' as const }, @@ -741,10 +746,13 @@ describe('dsh-agent-spine-demo bundle', () => { jobs: { maxConcurrentJobsPerOwner: 4 }, toolJobs: false as const, invariants: { enabled: false }, + goals: false as const, } expect(agentCore.pickSpineConfig(appConfig)).toEqual({ + maxParallelToolCalls: appConfig.maxParallelToolCalls, includeHarnessIdentity: appConfig.includeHarnessIdentity, + includeRuntimeContext: appConfig.includeRuntimeContext, persona: appConfig.persona, toolOrder: appConfig.toolOrder, tools: appConfig.tools, @@ -756,6 +764,7 @@ describe('dsh-agent-spine-demo bundle', () => { jobs: appConfig.jobs, toolJobs: appConfig.toolJobs, invariants: appConfig.invariants, + goals: appConfig.goals, }) expect(agentCore.pickSpineConfig({ workspaceContext: false })).toEqual({ workspaceContext: false }) }) diff --git a/packages/extensions/cordis-client-runner/src/client/api-catalog.ts b/packages/extensions/cordis-client-runner/src/client/api-catalog.ts index 35f3efef31..58bbbdee3f 100644 --- a/packages/extensions/cordis-client-runner/src/client/api-catalog.ts +++ b/packages/extensions/cordis-client-runner/src/client/api-catalog.ts @@ -73,7 +73,7 @@ export interface InheritedApiEntry { /** One named type declaration referenced by a Service or Event signature. */ export interface TypeApiEntry { - /** The exported type/interface name, e.g. `BashRunResult`. */ + /** The exported type/interface name, e.g. `ShellRunResult`. */ name: string /** The full declaration text, comments stripped. */ declaration: string @@ -411,7 +411,7 @@ export const TYPE_API: readonly TypeApiEntry[] = [ }, { name: 'AgentContext', - declaration: 'export type AgentContext = Omit & {\n readonly remote: TypeRTClientRemote & TypeRTRemoteScopeApi<\'agent\'>;\n};', + declaration: 'export type AgentContext = Omit & {\n readonly remote: TypertClientRemote & TypertRemoteScopeApi<\'agent\'>;\n};', }, { name: 'AssistantBlock', diff --git a/packages/extensions/tool-cordis/src/api-catalog.ts b/packages/extensions/tool-cordis/src/api-catalog.ts index 989b926264..5a812806da 100644 --- a/packages/extensions/tool-cordis/src/api-catalog.ts +++ b/packages/extensions/tool-cordis/src/api-catalog.ts @@ -1710,6 +1710,12 @@ export const SERVICE_API: readonly ServiceApiEntry[] = [ parameters: [{ name: 'context', description: 'the context contribution to register.' }], returns: 'the exact Cordis effect disposer.', }, + { + signature: 'suppressRuntimeContext(): () => void', + description: 'Suppress every dynamic runtime-context contribution in the calling context\'s scope without changing the services that own or enforce those facts. Multiple suppressors remain independently disposable.', + parameters: [], + returns: 'the exact Cordis effect disposer.', + }, { signature: 'tools(provider: (context: AssembleContext) => ToolProviderResult): () => void', description: 'Register a tool-schema provider in the calling context\'s scope. Global and matching scoped providers both contribute; returning the reserved TOOL_ORDER_REST name makes assembly fail.', @@ -4225,7 +4231,7 @@ export const TYPE_API: readonly TypeApiEntry[] = [ }, { name: 'SystemPrompt', - declaration: 'export class SystemPrompt extends Service {\n static Config: z;\n constructor(ctx: Context, config: Config);\n section(section: PromptSection): () => void;\n context(context: PromptContext): () => void;\n tools(provider: (context: AssembleContext) => ToolProviderResult): () => void;\n variable(name: string, provider: (context: AssembleContext) => string | undefined): () => void;\n async assemble(context: AssembleContext = {}): Promise;\n}', + declaration: 'export class SystemPrompt extends Service {\n static Config: z;\n constructor(ctx: Context, config: Config);\n section(section: PromptSection): () => void;\n context(context: PromptContext): () => void;\n suppressRuntimeContext(): () => void;\n tools(provider: (context: AssembleContext) => ToolProviderResult): () => void;\n variable(name: string, provider: (context: AssembleContext) => string | undefined): () => void;\n async assemble(context: AssembleContext = {}): Promise;\n}', }, { name: 'TableKeyOf', diff --git a/packages/preset/persona/README.i18n.yaml b/packages/preset/persona/README.i18n.yaml index c839c7dcda..0323741aed 100644 --- a/packages/preset/persona/README.i18n.yaml +++ b/packages/preset/persona/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/preset/persona/README.md -README.md: 742141e65fa8d50b89e6b74e6d21aa8c5bfe98cd -README.zh.md: 61112d1091e428d8c92aa056b7aaa272d3bbbc1d +README.md: 978d01ddbd52095bc2486d27d08fde2d4a8fa79b +README.zh.md: d36af6c028f1258d044608535c6da4eaf2fb92e8 diff --git a/packages/preset/persona/README.md b/packages/preset/persona/README.md index 742141e65f..978d01ddbd 100644 --- a/packages/preset/persona/README.md +++ b/packages/preset/persona/README.md @@ -16,8 +16,9 @@ Mounting this row outside an agent scope collides with the registry's own `deplo |---|---|---| | `text` | required | Persona prose rendered as the `deployment:persona` section | | `complete` | `false` | Restore this persona after assembly as the only system-prompt section | +| `includeRuntimeContext` | `true` | Include dynamic runtime-context snapshots for this agent scope; false suppresses every context contribution without disabling its owning services | -`text` is a template, like any prompt section: complete `{{…}}` groups resolve strictly against registered prompt variables when the prompt renders, not when it assembles. Empty text still occupies the slot, so it shadows the deployment persona away entirely and then disappears at render. With `complete: true`, assembly still resolves contexts, tools, variables, and cooperative listeners, then the prompt registry restores this exact persona as the sole section; no identity, tool guidance, or listener can append prompt text. +`text` is a template, like any prompt section: complete `{{…}}` groups resolve strictly against registered prompt variables when the prompt renders, not when it assembles. Empty text still occupies the slot, so it shadows the deployment persona away entirely and then disappears at render. With `complete: true`, assembly still resolves contexts, tools, variables, and cooperative listeners, then the prompt registry restores this exact persona as the sole section; no identity, tool guidance, or listener can append prompt text. With `includeRuntimeContext: false`, context providers are not evaluated for this scope and contexts added by assembly listeners are discarded. ## Model Experience @@ -25,7 +26,7 @@ Mounting this row outside an agent scope collides with the registry's own `deplo #### What the model sees -The `deployment:persona` section at order 0, immediately after the harness identity opener, carrying exactly this row's configured `text` with prompt variables resolved. For an agent whose preset mounts this row, it replaces whatever persona the deployment configured. In complete mode, the model sees only this rendered section as its system prompt. +The `deployment:persona` section at order 0, immediately after the harness identity opener, carrying exactly this row's configured `text` with prompt variables resolved. For an agent whose preset mounts this row, it replaces whatever persona the deployment configured. In complete mode, the model sees only this rendered section as its system prompt. Runtime context remains enabled by default. When disabled, a fresh agent receives no runtime-context snapshot from sandbox policy, approval policy, delegation, or another system-prompt context provider. #### Token effect diff --git a/packages/preset/persona/README.zh.md b/packages/preset/persona/README.zh.md index 61112d1091..d36af6c028 100644 --- a/packages/preset/persona/README.zh.md +++ b/packages/preset/persona/README.zh.md @@ -16,8 +16,9 @@ |---|---|---| | `text` | 必填 | 作为 `deployment:persona` 段落渲染的人设文本 | | `complete` | `false` | 组装后将此人设恢复为唯一的系统提示词段落 | +| `includeRuntimeContext` | `true` | 是否为此 agent 作用域包含动态 runtime-context 快照；false 会抑制所有上下文贡献，但不禁用拥有它们的服务 | -`text` 与任何提示词段落一样是模板：完整的 `{{…}}` 组在提示词**渲染**时（而非组装时）严格解析为已注册的提示词变量。空文本同样占据该槽位，因此会把部署级人设整个遮蔽掉，然后在渲染时消失。启用 `complete: true` 时，组装仍会解析上下文、工具、变量和协作式监听器，之后提示词注册表将这份确切人设恢复为唯一段落；身份、工具引导或监听器都无法追加提示词文本。 +`text` 与任何提示词段落一样是模板：完整的 `{{…}}` 组在提示词**渲染**时（而非组装时）严格解析为已注册的提示词变量。空文本同样占据该槽位，因此会把部署级人设整个遮蔽掉，然后在渲染时消失。启用 `complete: true` 时，组装仍会解析上下文、工具、变量和协作式监听器，之后提示词注册表将这份确切人设恢复为唯一段落；身份、工具引导或监听器都无法追加提示词文本。启用 `includeRuntimeContext: false` 时，此作用域的上下文提供方不会被求值，组装监听器添加的上下文也会被丢弃。 ## 模型体验 @@ -25,7 +26,7 @@ #### What the model sees -位于 order 0 的 `deployment:persona` 段落，紧随 harness 身份开场白之后，携带本行配置的 `text`，其中的提示词变量已解析。对于其 preset 挂载了本行的 agent，它会替换部署所配置的任何人设。在完整模式下，模型只会看到这个渲染后的段落作为系统提示词。 +位于 order 0 的 `deployment:persona` 段落，紧随 harness 身份开场白之后，携带本行配置的 `text`，其中的提示词变量已解析。对于其 preset 挂载了本行的 agent，它会替换部署所配置的任何人设。在完整模式下，模型只会看到这个渲染后的段落作为系统提示词。Runtime context 默认保持启用。禁用后，新建 agent 不会收到来自沙箱策略、批准策略、委派或其他 system-prompt 上下文提供方的 runtime-context 快照。 #### Token effect diff --git a/packages/preset/persona/src/index.ts b/packages/preset/persona/src/index.ts index 69a53678e9..f45b549656 100644 --- a/packages/preset/persona/src/index.ts +++ b/packages/preset/persona/src/index.ts @@ -40,12 +40,15 @@ export interface Config { text: string /** Make this persona the complete system prompt, suppressing every other section. */ complete?: boolean + /** Suppress dynamic runtime-context snapshots for this persona's agent scope. */ + includeRuntimeContext?: boolean } /** Runtime schema for the persona row. */ export const Config: z = z.object({ text: z.string().required(), complete: z.boolean().default(false), + includeRuntimeContext: z.boolean().default(true), }) /** @@ -61,4 +64,5 @@ export function apply(ctx: Context, config: Config): void { text: config.text, ...(config.complete ? { complete: true } : {}), }), 'persona.section()') + if (!(config.includeRuntimeContext ?? true)) ctx.systemPrompt.suppressRuntimeContext() } diff --git a/packages/preset/persona/tests/persona.spec.ts b/packages/preset/persona/tests/persona.spec.ts index 3fcc30d3a5..3343f7a6f9 100644 --- a/packages/preset/persona/tests/persona.spec.ts +++ b/packages/preset/persona/tests/persona.spec.ts @@ -102,4 +102,41 @@ describe('the persona row', () => { expect(assembly.sections).toEqual([{ name: PERSONA_SECTION, text: 'Only this.' }]) expect(renderPrompt(assembly)).toBe('Only this.') }) + + it('can suppress runtime context for its scope without changing the global assembly', async () => { + const ctx = await harness('deployment identity') + const key: ScopeKey = { agent: 'a1' } + const scope = createScope(ctx, key) + ctx.systemPrompt.context({ name: 'policy', order: 1, text: 'global policy' }) + + const fiber = await scope.ctx.plugin(Persona, { + text: 'Only this.', + includeRuntimeContext: false, + }) + const suppressed = await ctx.systemPrompt.assemble({ scope: key }) + expect(suppressed.contexts).toEqual([]) + const global = await ctx.systemPrompt.assemble() + expect(global.contexts).toEqual([ + { name: 'policy', text: 'global policy' }, + ]) + + await fiber.dispose() + expect((await ctx.systemPrompt.assemble({ scope: key })).contexts).toEqual([ + { name: 'policy', text: 'global policy' }, + ]) + }) + + it('keeps runtime context by default when apply bypasses schema defaults', async () => { + const ctx = await harness('deployment identity') + const key: ScopeKey = { agent: 'a1' } + ctx.systemPrompt.context({ name: 'policy', order: 1, text: 'global policy' }) + + await ctx.plugin(Object.assign((inner: Context) => { + Persona.apply(createScope(inner, key).ctx, { text: 'Scoped identity.' }) + }, { inject: ['systemPrompt'] })) + + expect((await ctx.systemPrompt.assemble({ scope: key })).contexts).toEqual([ + { name: 'policy', text: 'global policy' }, + ]) + }) })