Merge remote-tracking branch 'origin/master' into feat/web-session-log-export
This commit is contained in:
_Kerman
142 files changed
+3934
-262
No files matched your search
+2
-2
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-21-mandatory-app-attribution-headers.md
|
||||
2026-06-21-mandatory-app-attribution-headers.md: 12482ca80d19e5cd1e62b8860865db6cccea61ab
|
||||
2026-06-21-mandatory-app-attribution-headers.zh.md: 43f356887948b88d34ba41d46ed1cd2f1c89a1e2
|
||||
2026-06-21-mandatory-app-attribution-headers.md: 39050a53ec76e8c5a6cac4d8e31fa15b992c406e
|
||||
2026-06-21-mandatory-app-attribution-headers.zh.md: bfd6aa2540022f68cf9f69ccc2c12b3bc0978960
|
||||
+1
-3
@@ -32,7 +32,7 @@ The provider-neutral identity is owned by `dsh-llm` (`packages/llm/llm/src/attri
|
||||
|
||||
- product token for `User-Agent`: `deepseek-harness` (continuity with the pre-Agent Note wire value and the repo/org identity)
|
||||
- version: read from the owning package's manifest via `createRequire`, never a hand-copied constant
|
||||
- app URL: `https://github.com/deepseek-ai/deepseek-harness-sdk` - the planned public home, which must exist before release
|
||||
- app URL: `https://github.com/deepseek-ai/deepseek-harness` - the repository home
|
||||
|
||||
The default is mandatory and non-empty. White-label deployments pass their own `AppIdentity` to `attributionHeaders(identity)` - the override hook is the function parameter, with no deployment config plumbing until a consumer needs it - and omission falls back to the harness default rather than suppressing attribution. There is no per-request API for the model, user prompt, session id, cwd, user email, API key owner, or local machine identity to influence these fields.
|
||||
|
||||
@@ -77,8 +77,6 @@ The landed contract:
|
||||
|
||||
**Providers see that traffic comes from the harness.** That is the point, but it means deployments that previously blended into generic SDK traffic become identifiable. Mitigation: send only static public product data and let forks/white-label deployments pass their own `AppIdentity`.
|
||||
|
||||
**The app URL points at a repository that does not exist yet.** `deepseek-ai/deepseek-harness-sdk` is the planned public home; until it is created the URL is a dangling promise that blocks release.
|
||||
|
||||
**Header support differs by client library.** The hand-rolled adapter sets headers directly; the pi-ai-backed adapter depends on pi-ai continuing to honor `StreamOptions.headers` (merged last over provider defaults). The wire-level mock-server tests are the guard: if a pi-ai upgrade stops delivering the header, the suite goes red. This is useful pressure on the abstraction: a provider adapter that cannot set mandatory headers cannot fully implement the harness LLM contract.
|
||||
|
||||
**OpenRouter rankings do not benefit yet.** `User-Agent` is the correct baseline for provider-neutral HTTP identity, but it will not create OpenRouter app pages or rankings because OpenRouter requires `HTTP-Referer` for that product feature. That is deliberate: public app marketplace participation is a separate product decision, not a prerequisite for mandatory request attribution.
|
||||
+1
-3
@@ -32,7 +32,7 @@ OpenRouter 应用归属刻意未实现。`HTTP-Referer`、`X-OpenRouter-Title`
|
||||
|
||||
- `User-Agent` 的产品 token：`deepseek-harness`（与 Agent Note 之前的线路值及仓库/组织身份保持连续性）
|
||||
- 版本：通过 `createRequire` 从所属包的 manifest（元数据清单）读取，绝不手动复制常量
|
||||
- 应用 URL：`https://github.com/deepseek-ai/deepseek-harness-sdk`——计划中的公开主页，且必须在发布前实际存在
|
||||
- 应用 URL：`https://github.com/deepseek-ai/deepseek-harness`——仓库主页
|
||||
|
||||
默认值是强制的且非空。白标部署通过向 `attributionHeaders(identity)` 传入自己的 `AppIdentity` 来覆盖——覆盖钩子就是函数参数，在有消费方需要之前不做部署配置管道——省略时回退到 harness 默认值而非抑制归属。没有逐请求 API 允许模型、用户提示词、会话 id、cwd、用户邮箱、API key 所有者或本地机器身份影响这些字段。
|
||||
|
||||
@@ -77,8 +77,6 @@ OpenRouter 应用归属刻意未实现。`HTTP-Referer`、`X-OpenRouter-Title`
|
||||
|
||||
**提供方看到流量来自 harness。** 这正是目的，但意味着此前混在通用 SDK 流量中的部署变得可识别。缓解措施：仅发送静态公开产品数据，并允许 fork/白标部署传入自己的 `AppIdentity`。
|
||||
|
||||
**应用 URL 指向一个尚不存在的仓库。** `deepseek-ai/deepseek-harness-sdk` 是计划中的公开主页；在它创建之前，该 URL 是一个阻塞发布的悬空承诺。
|
||||
|
||||
**不同客户端库的头部支持有差异。** 手写适配器直接设置头部；基于 pi-ai 的适配器依赖 pi-ai 继续尊重 `StreamOptions.headers`（最后合并覆盖提供方默认值）。线路级 mock 服务器测试是守卫：如果 pi-ai 升级后不再投递该头部，套件会变红。这对抽象施加了有益的压力：一个无法设置强制头部的提供方适配器不能完整实现 harness 的 LLM 约定。
|
||||
|
||||
**OpenRouter 排名尚未受益。** `User-Agent` 是提供方无关的 HTTP 身份的正确基线，但它不会创建 OpenRouter 应用页面或排名，因为 OpenRouter 要求 `HTTP-Referer` 来实现该产品功能。这是有意为之：公开应用市场参与是一个独立的产品决策，不是强制请求归属的前提。
|
||||
+2
-2
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-25-web-client-session-scope-and-provide-channel.md
|
||||
2026-07-25-web-client-session-scope-and-provide-channel.md: ff411e1b387dcd83b04f8c02d5dde3c70a815932
|
||||
2026-07-25-web-client-session-scope-and-provide-channel.zh.md: 593e194654bb3552ad3cac4534847fba11cf3694
|
||||
2026-07-25-web-client-session-scope-and-provide-channel.md: f371b93ccf6cb3ba10cbdb73baa670cbf889f393
|
||||
2026-07-25-web-client-session-scope-and-provide-channel.zh.md: 78440131c4e09c9a458009fbe5a2a34a707a481c
|
||||
+2
-2
@@ -93,7 +93,7 @@ Slot scope is the closed set `root | session-maybe | session`:
|
||||
- `session-maybe` follows the current session with ADOPTION identity (the only behavior — there is no hold-identity-forever mode): an incarnation born session-less keeps its React instance across the arrival of the FIRST session (the blank shell adopts it — no remount, the DOM survives), and from then on behaves exactly like a strict session entry — switching to a different session remounts, and dropping back to no-session remounts into a fresh blank incarnation that will adopt again. Component-local per-session state therefore clears by construction; state that must survive a switch belongs in session-bound sources (machine, store, hooks). With no session, `sessionId`, the results of `useSession`/`useInput`, and `inputActions` may all be absent. The unkeyed root `SessionMaybeProvider` drives these updates by subscribing to the runtime's atomic `currentProvide` projection — selection moves and provider-roster changes publish through the same source, so a roster change under a stable current id republishes the mounted bundle instead of stranding entries on an obsolete hook/prop schema — while `SessionMaybeProvideInfo` uses the static key map to retain the complete hook/prop shape even with no session; the per-entry adoption bookkeeping (incarnation-counter key) lives in the renderer's `SessionMaybeEntry`.
|
||||
- `session` guarantees that `sessionId`, every hook source, and every prop exist; each strict entry's error boundary is keyed by `sessionId`, so switching sessions recreates that entry and its session store.
|
||||
|
||||
`conversation` is the resident `session-maybe` shell: `ConversationRoot`, HeroShell, the Workspace picker, the root-owned scrollport and composer stack, and the overlay chain's fallback frame retain their React instances across the no-session → blank-session switch. Two strict entries fill fixed regions without reparenting that tree: `conversation.session.header` carries breadcrumb/tabs/actions above the scrollport, while `conversation.session` carries the view ring and draft mirror inside it; both share the same session-scoped chat store. The composer bar (`conversation.composer.bar`) is itself `session-maybe`: with no session it renders inert (machine faces absent, `disabled` owner prop), and the same instance — textarea included — goes live when a session appears; the remaining input slots stay strict `session` and dispatch nothing until then. The blank → engaging/active transition never rebuilds the InputBar on a phase flip.
|
||||
`conversation` is the resident `session-maybe` shell: `ConversationRoot`, HeroShell, the Workspace picker, the root-owned scrollport and composer stack, and the overlay chain's fallback frame retain their React instances across the no-session → blank-session switch. Two strict entries fill fixed regions without reparenting that tree: `conversation.session.header` carries breadcrumb/tabs/actions above the scrollport, while `conversation.session` carries the view ring and draft mirror inside it; both share the same session-scoped chat store. The composer bar (`conversation.composer.bar`) is itself `session-maybe`: with no session its machine faces and message actions are inert, while the whole dashed card opens the existing Workspace picker by pointer and its read-only textarea does the same through Enter or Space. The same instance — textarea included — goes live when a session appears; the remaining input slots stay strict `session` and dispatch nothing until then. The blank → engaging/active transition never rebuilds the InputBar on a phase flip.
|
||||
|
||||
- The runtime's first built-in entry: the `'session'` hook — `useSession` itself rides the same mechanism, no special-casing.
|
||||
- Concurrent discipline: the render plane reads only from the hooks compartment (uSES consistency guarantee); props-compartment callbacks are used only in event-handler space; descriptor resolution is render-safe (idempotent caching, with prune reaping residue from abandoned renders).
|
||||
@@ -132,5 +132,5 @@ Slot scope is the closed set `root | session-maybe | session`:
|
||||
- Plugins gain session context isomorphic to the host's: per-session state hangs on the actx and mounts/tears down in one piece with the scope fiber, making leaks structurally impossible; two-session isolation is structurally guaranteed by the scope filter.
|
||||
- The client object layer converges to a wire mirror: session identity, lifecycle, and capability adjudication all defer to the host entity — the input system (the next layer) always faces a session with a real Agent, and providers like slash/skill uniformly address by sessionId directly.
|
||||
- Blank-session governance takes zero dedicated mechanisms: state rides one derived bit, visibility rides the unified list projection (only the current blank shows, as `New Session`), reclamation rides lazy persistence's existing contract (evaporation on restart), and the ordinary ceiling rides same-Workspace reuse.
|
||||
- The cost: the id→ctx handoff discipline and provide's Concurrent discipline are conventions rather than type-enforced, pinned by review and tests; fully disabled input while no workspace is picked is an experience cost the product surface accepts (the price of the single state axis).
|
||||
- The cost: the id→ctx handoff discipline and provide's Concurrent discipline are conventions rather than type-enforced, pinned by review and tests. The single state axis still withholds machine faces until a Session exists; the resident card routes activation to the Workspace picker during that interval ([decision](../feature/2026-08-07-workspace-picker-composer-entry.md)).
|
||||
- Known gaps: approval/question recovery across prune (TODO); model selection returns in live-mutation shape (the host `selectModel` trio is ready-made, its client consumer not yet built).
|
||||
+2
-2
@@ -93,7 +93,7 @@ slot scope 是闭集 `root | session-maybe | session`：
|
||||
- `session-maybe` 以**收养（adoption）身份语义**跟随 current session（唯一行为——不存在「永久保持实例」模式）：空态出生的化身在**第一个** session 到来时保持 React 实例（空壳收养它——不重挂，DOM 存活）；此后行为与严格 session entry 完全一致——切到不同 session 重挂，跌回无 session 也重挂为崭新的空态化身（之后再次收养）。因此组件本地的 per-session 状态**由构造保证**随切换清零；需要活过切换的状态必须住 session 绑定的源（machine、store、hooks）。无 session 时 `sessionId`、`useSession`/`useInput` 的选择结果及 `inputActions` 均可缺省。根部无 key 的 `SessionMaybeProvider` 通过订阅 runtime 的原子 `currentProvide` 投影驱动这条更新——选择移动和提供方名册变化经同一 source 发布，current id 不变时的名册变化也会重发已挂载 bundle，而不是把 entry 困在过期的钩子/prop 形状上——`SessionMaybeProvideInfo` 靠静态键表在无 session 时仍保留完整钩子/prop 形状；逐 entry 的收养记账（化身计数 key）住在 renderer 的 `SessionMaybeEntry`。
|
||||
- `session` 保证 `sessionId`、所有钩子 source 与 props 均存在；每个严格 entry 的错误边界以 `sessionId` 为 key，切换 session 会重建该 entry 及其 session store。
|
||||
|
||||
`conversation` 是 `session-maybe` 的常驻外壳：`ConversationRoot`、HeroShell、Workspace picker、root 持有的 scrollport 与 composer stack，以及 overlay chain 的 fallback 外框，在无 session → blank session 的切换中保持 React 实例。两个严格 session entry 只填入固定区域，不改变该树的父级：`conversation.session.header` 在 scrollport 上方承载 breadcrumb／tab／action，`conversation.session` 在其内部承载 view ring 与 draft mirror；二者共享同一个 session scope chat store。composer bar（`conversation.composer.bar`）本身即为 `session-maybe`：无 session 时以惰性态渲染（machine face 缺席、`disabled` owner prop），session 出现后同一实例（含 textarea）转为 live；其余输入 slot 保持严格 `session`，在此之前不分发任何条目。blank → engaging/active 的 InputBar 不因 phase 翻转而重建。
|
||||
`conversation` 是 `session-maybe` 的常驻外壳：`ConversationRoot`、HeroShell、Workspace picker、root 持有的 scrollport 与 composer stack，以及 overlay chain 的 fallback 外框，在无 session → blank session 的切换中保持 React 实例。两个严格 session entry 只填入固定区域，不改变该树的父级：`conversation.session.header` 在 scrollport 上方承载 breadcrumb／tab／action，`conversation.session` 在其内部承载 view ring 与 draft mirror；二者共享同一个 session scope chat store。composer bar（`conversation.composer.bar`）本身即为 `session-maybe`：无 session 时，其 machine face 和消息操作保持惰性，整张虚线卡片可经指针打开现有 Workspace picker，只读 textarea 也可通过 Enter 或 Space 打开。session 出现后同一实例（含 textarea）转为 live；其余输入 slot 保持严格 `session`，在此之前不分发任何条目。blank → engaging/active 的 InputBar 不因 phase 翻转而重建。
|
||||
|
||||
- 运行时内建第一条：`'session'` 钩子——`useSession` 本身走同一机制，无特判。
|
||||
- Concurrent 纪律：渲染平面只从 hooks 格读（uSES 一致性保证）；props 格回调只在事件 handler 空间用；描述符解析 render-safe（幂等缓存、废弃渲染残留由 prune 收尸）。
|
||||
@@ -132,5 +132,5 @@ slot scope 是闭集 `root | session-maybe | session`：
|
||||
- 插件获得与 host 同构的会话上下文：逐会话状态挂 actx、随 scope fiber 一次拆装，泄漏结构性不可能；双会话隔离由 scope filter 结构性保证。
|
||||
- client 对象层收敛为 wire 镜像：会话身份、生命周期、能力判别全部以 host 实体为准——输入体系（下一层）面对的永远是「有真 Agent 的会话」，slash/skill 等提供方一律以 sessionId 直接寻址。
|
||||
- 空会话治理零专用机制：状态靠一个派生位，可见性靠统一列表投影（仅 current blank 以 `New Session` 展示），回收靠 lazy persistence 的既有约定（重启蒸发），常规上限靠同 Workspace 复用。
|
||||
- 代价：id→ctx 换乘纪律、provide 的 Concurrent 纪律都是约定而非类型强制，靠 review 与测试钉住；「未选 workspace」期间输入全禁是产品面接受的体验代价（单一状态轴换来的）。
|
||||
- 代价：id→ctx 换乘纪律、provide 的 Concurrent 纪律都是约定而非类型强制，靠 review 与测试钉住。单一状态轴仍会在 Session 存在前隐藏 machine face；这段时间内，常驻卡片会把激活操作转到 Workspace picker（[决策](../feature/2026-08-07-workspace-picker-composer-entry.md)）。
|
||||
- 已知欠账：approval/question 跨 prune 恢复（TODO）；模型选择以 live-mutation 形状回归（host `selectModel` 三件套现成，其 client 消费方尚未构建）。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-10-message-feedback-sidecar.md
|
||||
2026-08-10-message-feedback-sidecar.md: 780cbaa840fcac7bcfa799468bfd61f5b08715cb
|
||||
2026-08-10-message-feedback-sidecar.zh.md: 72ecc82717010f65d013418a45c3b38c6084aaf9
|
||||
@@ -0,0 +1,43 @@
|
||||
# Agent Note: Lifecycle-bound message feedback sidecar
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-10-message-feedback-sidecar.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The existing `/feedback` command records an immutable Session-level `feedback/record` event. That event can release a pending telemetry prefix under `FEEDBACK_ONLY`, so it is the wrong authority for an editable positive/negative rating and optional note attached to one assistant message. Message feedback needs independent update and delete semantics without entering the canonical Session log, changing a projection, reaching the model surface, or implicitly consenting to telemetry.
|
||||
|
||||
A sidecar keyed only by `SessionId` can outlive the log lifecycle it describes when an id is recreated with a different header identity. A Session-wide revision also makes unrelated message edits conflict, while plain storage-domain read/put has no cross-process compare-and-swap. Session disposal is only live-store detach, not durable deletion, and the current Session persistence seam exposes no deletion operation that could own a truthful cascade.
|
||||
|
||||
## Decision
|
||||
|
||||
`@deepseek-ai/dsh-message-feedback` owns the `ctx.messageFeedback` service and stores message feedback as one storage-domain sidecar row per Session. The sidecar is neither Session-log content nor a Session projection. It emits no `feedback/record` event and performs no telemetry handoff; the command-feedback and message-feedback contracts remain independent.
|
||||
|
||||
Every usable row is bound to the inspected Session header identity `{createdAt, cwd}`, not merely its `SessionId`. A lifecycle mismatch is treated as absence: `list` returns no items, and `put` may replace the stale row with one bound to the current identity. An id reused with a different header identity therefore cannot inherit stale feedback. A fork receives its own Session identity and no sidecar copy: even when the fork seed contains the same assistant messages, feedback remains attached to the Session in which the human recorded it.
|
||||
|
||||
`put` accepts a target only when `SessionPersistence.inspect()` observes a non-empty, append-origin `assistant/message` with that `MessageId`. Replacement-origin messages, empty usage-only assistant records, and non-assistant targets are rejected. Inspection is the cold-safe authority: it neither publishes or resumes an Agent nor commits cold-log repair merely to validate feedback. A cold `listSnapshots()` preflight classifies definite absence; inspection failure for a catalogued Session remains an infrastructure failure. A request in the narrow live-detach-to-header-materialization interval can therefore return `session-not-found`, and the caller retries after retirement materialization.
|
||||
|
||||
Before `put` commits a sidecar row, it puts the target log behind a durability barrier. A matching live Session passes through the canonical `ctx.sessions.flush` checkpoint, then both live and cold paths are physically read from sequence zero through `SessionPersistence.readFrom`. The resulting observation's header identity and target are checked again. A missing flush participant, changed identity, vanished target, or physical-read failure prevents the sidecar write, so a committed feedback item never precedes the durable assistant message it references.
|
||||
|
||||
Each message item carries its own opaque version plus Host-assigned `createdAt` and `updatedAt` timestamps. `put` compares the caller's `ifVersion` only with the addressed item, so editing one message does not invalidate another. The comparison is strict even when the desired value already matches, preventing a stale request from crossing an ABA value cycle; a conflict returns the authoritative current item so callers can reconcile without a second read. A matching-version no-op preserves the version and timestamps, while a material update preserves `createdAt`, replaces the version, and keeps `updatedAt` from moving backward. An already-absent delete is likewise successful. Versions are tokens for equality, not counters callers may order or synthesize.
|
||||
|
||||
A per-Session mutation queue encloses lifecycle inspection, sidecar read, conflict evaluation, and whole-row write. This makes one service instance's mutations serial and preserves the per-message compare-and-swap contract inside one Host process. Plugin disposal closes admission, drains accepted queue work, and then closes the storage domain. The underlying storage-domain API provides no cross-process conditional write, so the implementation claims no cross-process linearizability or lost-update protection.
|
||||
|
||||
`maxNoteBytes` is a required deployment choice and bounds the UTF-8 byte length of an optional note; the Web Host bundle sets it explicitly to `8192`. The package publishes the Host `messageFeedback.list`, `messageFeedback.put`, and `messageFeedback.delete` contract directly through `GatewayService` and `@Remote`. Client Remote aggregate mounting and UI remain separately owned and deferred; their later adapter stays a thin consumer of this Host contract.
|
||||
|
||||
The service performs no fake deletion cascade. `session/disposed` and `host/session-removed` describe detach from live ownership, not durable Session deletion, and Session persistence currently has no delete surface. Sidecar rows can therefore remain after out-of-band log removal; a different `{createdAt, cwd}` prevents such an orphan from becoming feedback for a later Session that reuses the id.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Append edits to the Session log and derive a projection.** Rejected because editable UI metadata would become canonical conversation-adjacent history, forks would replay and inherit it, deletion would require tombstones, and reusing `feedback/record` would silently couple a message rating to telemetry consent.
|
||||
|
||||
**Key feedback globally by `MessageId`, copy it on fork, or use one Session revision.** Rejected because message ids are meaningful only within a Session lifecycle, forked conversations need independent human judgments, and unrelated message mutations must not create false conflicts.
|
||||
|
||||
**Extend `KvTable` with cross-process compare-and-swap in this change.** Rejected because the shipped storage-domain backends expose no common conditional-write primitive. A process-local queue matches the supported one-Host topology; a real multi-process guarantee requires a backend-level atomic contract and is separate work.
|
||||
|
||||
**Delete feedback on Session disposal.** Rejected because disposal includes ordinary detach and rollback paths. Treating it as durable deletion would lose feedback while the Session log still exists; cleanup waits for a real Session deletion authority.
|
||||
|
||||
## Consequences
|
||||
|
||||
Message feedback is locally durable and independently editable without changing model-visible history or telemetry behavior. Concurrent callers in one Host receive per-message conflict detection and retry-safe outcomes, while deployments with multiple writers to the same storage root remain unsupported. A differing header identity treats a stale row as absent but does not reclaim it; a cloned log that retains the same `{createdAt, cwd}` is indistinguishable by this contract. The Host Remote contract is available now; client assembly and UI can remain thin consumers rather than taking ownership of persistence or concurrency semantics.
|
||||
@@ -0,0 +1,43 @@
|
||||
# Agent Note: 绑定生命周期的消息反馈伴随记录
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-10-message-feedback-sidecar.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
现有 `/feedback` 命令记录不可变的 Session 级 `feedback/record` 事件。在 `FEEDBACK_ONLY` 下，该事件可以释放待处理的遥测前缀，因此它不适合作为挂在单条 assistant 消息上的可编辑好评／差评与可选备注的权威来源。消息反馈需要独立的更新与删除语义，且不得进入权威 Session 日志、改变投影、到达模型接口，或隐式表示遥测同意。
|
||||
|
||||
只按 `SessionId` 建索引的伴随记录可能在该 id 以不同 header 身份重建后，继续存活于其所描述的日志生命周期之外。Session 级 revision 还会让无关消息的编辑彼此冲突，而普通 storage-domain 读／写不提供跨进程 compare-and-swap。Session disposal 只是从 live store 脱离，并非持久删除；当前 Session 持久化 seam 也没有可拥有真实级联的删除操作。
|
||||
|
||||
## 决策
|
||||
|
||||
`@deepseek-ai/dsh-message-feedback` 拥有 `ctx.messageFeedback` 服务，并把消息反馈存为每个 Session 一条 storage-domain 伴随记录（sidecar）。该伴随记录既不是 Session 日志内容，也不是 Session 投影。它不发出 `feedback/record` 事件，也不执行遥测交接；command-feedback 与 message-feedback 契约保持独立。
|
||||
|
||||
每条可用记录都绑定到经检查的 Session header 身份 `{createdAt, cwd}`，而不只是其 `SessionId`。生命周期不匹配按不存在处理：`list` 返回空条目，`put` 可以用绑定当前身份的新记录替换陈旧行。因此，以不同 header 身份复用的 id 不会继承陈旧反馈。fork 拥有自己的 Session 身份，且不复制伴随记录：即使 fork 种子包含相同的 assistant 消息，反馈仍只属于人类记录它的那个 Session。
|
||||
|
||||
`put` 只接受由 `SessionPersistence.inspect()` 观测到的非空、append-origin `assistant/message`，且其 `MessageId` 必须与目标相同。replacement-origin 消息、仅承载 usage 的空 assistant 记录以及非 assistant 目标都会被拒绝。检查使用 cold-safe 权威路径：它不会仅为验证反馈而发布或恢复 Agent，也不会提交 cold 日志修复。cold 路径由 `listSnapshots()` 预检明确不存在；已进入目录的 Session 若检查失败，仍按基础设施故障处理。因此，请求若恰落在 live detach 到 header materialization 的极短窗口，可能返回 `session-not-found`，调用方在 retirement materialization 后重试。
|
||||
|
||||
`put` 提交伴随记录前，会先让目标日志通过 durability barrier。身份匹配的 live Session 经过权威 `ctx.sessions.flush` checkpoint，随后 live 与 cold 路径都会通过 `SessionPersistence.readFrom` 从序列零做物理复读。之后再次校验所得观测的 header 身份与目标。缺少 flush 参与方、身份变化、目标消失或物理读取失败都会阻止伴随记录写入，因此已提交反馈绝不会先于它引用的持久 assistant 消息。
|
||||
|
||||
每个消息条目都携带自己的 opaque version，以及 Host 分配的 `createdAt` 和 `updatedAt` 时间戳。`put` 只把调用方的 `ifVersion` 与目标条目比较，因此编辑一条消息不会使另一条消息失效。即使目标值已经相同，比较仍然严格执行，从而防止陈旧请求穿过 ABA 值循环；冲突会返回权威当前条目，调用方无需二次读取即可协调。携带匹配 version 的无变化请求会保留 version 与时间戳；实质更新保留 `createdAt`、替换 version，并保证 `updatedAt` 不倒退。删除已经不存在的条目也同样成功。version 是只能做相等比较的 token，不是调用方可以排序或自行合成的计数器。
|
||||
|
||||
按 Session 划分的变更队列覆盖生命周期检查、伴随记录读取、冲突判断与整行写入。这使同一个服务实例的变更串行化，并在单个 Host 进程内保持逐消息 compare-and-swap 契约。Plugin disposal 会关闭接纳、排空已进入队列的工作，然后关闭 storage domain。底层 storage-domain API 不提供跨进程条件写，因此实现不承诺跨进程线性一致性或防止丢失更新。
|
||||
|
||||
`maxNoteBytes` 是必填的部署选择，用于限制可选备注的 UTF-8 字节长度；Web Host bundle 将其显式设为 `8192`。该包通过 `GatewayService` 与 `@Remote` 直接发布 Host `messageFeedback.list`、`messageFeedback.put` 与 `messageFeedback.delete` 契约。客户端 Remote 聚合挂载与 UI 由各自边界负责并保持延后；后续适配层只是该 Host 契约的薄消费者。
|
||||
|
||||
服务不伪造删除级联。`session/disposed` 与 `host/session-removed` 表示脱离 live ownership，而非持久删除，Session persistence 当前也没有删除接口。因此在带外移除日志后，伴随记录可能继续存在；不同的 `{createdAt, cwd}` 可阻止此类孤儿记录变成后来复用该 id 的 Session 反馈。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**把编辑追加到 Session 日志并派生投影。** 不予采纳，因为可编辑 UI 元数据会变成权威且邻近对话的历史，fork 会回放并继承它，删除需要 tombstone，而复用 `feedback/record` 会把消息评分与遥测同意静默耦合。
|
||||
|
||||
**按全局 `MessageId` 建索引、在 fork 时复制，或使用一个 Session revision。** 不予采纳，因为消息 id 仅在某个 Session 生命周期内有意义，fork 后的对话需要独立的人类判断，而且无关消息的变更不应制造虚假冲突。
|
||||
|
||||
**在本次变更中为 `KvTable` 扩展跨进程 compare-and-swap。** 不予采纳，因为出厂 storage-domain 后端没有共同的条件写原语。进程内队列符合受支持的单 Host 拓扑；真实的多进程保证需要后端级原子契约，属于独立工作。
|
||||
|
||||
**在 Session disposal 时删除反馈。** 不予采纳，因为 disposal 包含普通 detach 与 rollback 路径。把它当成持久删除会在 Session 日志仍存在时丢失反馈；清理必须等待真正的 Session 删除权威。
|
||||
|
||||
## 后果
|
||||
|
||||
消息反馈在本地持久化并可独立编辑，且不改变模型可见历史或遥测行为。同一 Host 中的并发调用方获得逐消息冲突检测与可安全重试的结果；多个写入者共享同一存储根目录的部署仍不受支持。不同的 header 身份会让陈旧记录被视为不存在，但不会将其回收；本契约无法区分保留相同 `{createdAt, cwd}` 的克隆日志。Host Remote 契约现在可用；客户端组装与 UI 可以保持为薄消费者，而不接管持久化或并发语义。
|
||||
@@ -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/2026-07-10-parallel-tool-call-execution.md
|
||||
2026-07-10-parallel-tool-call-execution.md: 830691b7596accd1c5746fc51e38a9022f09e501
|
||||
2026-07-10-parallel-tool-call-execution.zh.md: bd336919adbf9b96f64477c1e24964172cde6d92
|
||||
2026-07-10-parallel-tool-call-execution.md: 47a95e88c3d0d8bb9e3ddd820b1e63849effd914
|
||||
2026-07-10-parallel-tool-call-execution.zh.md: 024e5d4e4cdbf9d1d980bc67db5969fb34941752
|
||||
@@ -60,7 +60,7 @@ Any shared state touched during execution must be concurrency-safe. This include
|
||||
|
||||
`maxParallelToolCalls` is a positive AgentLoop deployment cap shared by every agent the factory creates. It defaults to `10`; `1` preserves serial execution. Exact fields and defaults live in the generated [configuration catalog](../../../../docs/config-catalog.md).
|
||||
|
||||
The shipped declarations are conservative. Web search, web fetch, and filesystem read opt in. Filesystem writes and edits, bash tools, subagent delegation, workflow, user interaction, todo mutation, Code Mode, and Cordis mutation tools remain exclusive. A subagent may share its parent's workspace or external resources, and the unary classifier cannot prove that sibling delegations have disjoint effects. Bash has no proven input-sensitive classifier and remains exclusive.
|
||||
The shipped declarations are conservative. Web search, web fetch, filesystem read, the session-query trace/read tools, and subagent delegation opt in — delegation because a child works in its own session and its run never mutates the parent session, with sibling workspace coordination owned by the model ([parallel subagent Agent Note](2026-08-09-parallel-subagent-delegations.md)). Filesystem writes and edits, bash tools, the session-query search tools, workflow, user interaction, todo mutation, Code Mode, and Cordis mutation tools remain exclusive. Bash has no proven input-sensitive classifier and remains exclusive.
|
||||
|
||||
Filesystem read relies on a narrow recorder exception: its synchronous observation updates may settle out of order, but write and edit re-check the observed version before mutation, so stale state only produces `FS_STALE_VERSION`.
|
||||
|
||||
|
||||
@@ -60,7 +60,7 @@ Code Mode 仍不使用此调度器，因为模型只会发出一个原生 `run_c
|
||||
|
||||
`maxParallelToolCalls` 是 AgentLoop 的正整数部署上限，由工厂创建的所有 agent（智能体）共享。默认值为 `10`；`1` 保持串行执行。字段和默认值的精确定义见生成的[配置目录](../../../../docs/config-catalog.md)。
|
||||
|
||||
当前实现中的声明保持保守。Web 搜索、Web 获取和文件系统读取选择并行。文件系统写入与编辑、bash 工具、subagent 委派、工作流、用户交互、todo 变更、Code Mode 以及 Cordis 变更工具仍按独占方式执行。subagent 可能共享父级的工作区或外部资源，而一元分类器无法证明并列委派的作用互不重叠。Bash 没有已证明的输入敏感分类器，因此仍按独占方式执行。
|
||||
当前实现中的声明保持保守。Web 搜索、Web 获取、文件系统读取、会话查询的 trace/read 工具和 subagent 委派选择并行；委派之所以并行，是因为子 agent 在自己的会话中工作，其运行绝不变更父会话，并列委派间的工作区协调由模型负责（[并行 subagent Agent Note](2026-08-09-parallel-subagent-delegations.md)）。文件系统写入与编辑、bash 工具、会话查询的 search 工具、工作流、用户交互、todo 变更、Code Mode 以及 Cordis 变更工具仍按独占方式执行。Bash 没有已证明的输入敏感分类器，因此仍按独占方式执行。
|
||||
|
||||
文件系统读取依赖一个范围很窄的记录器例外：其同步观察更新可以不按顺序结算，但写入和编辑在变更前会重新检查已观察的版本，因此陈旧状态只会导致 `FS_STALE_VERSION`。
|
||||
|
||||
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-08-07-workspace-picker-composer-entry.md
|
||||
2026-08-07-workspace-picker-composer-entry.md: dc9c26c291de6e7614ace3c787030c0032a9740d
|
||||
2026-08-07-workspace-picker-composer-entry.zh.md: 585c59b12392823a7388ab7a635a864bd4108929
|
||||
@@ -0,0 +1,29 @@
|
||||
# Agent Note: The no-Workspace composer opens the existing picker
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-07-workspace-picker-composer-entry.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The [session-scope decision](../architecture/2026-07-25-web-client-session-scope-and-provide-channel.md) keeps one resident composer before a Workspace exists, but its textarea was disabled and only the smaller Workspace chip could open the picker. The largest and most familiar starting affordance therefore rejected the user's first click even though a recovery action was available on the same surface.
|
||||
|
||||
## Decision
|
||||
|
||||
While no Workspace owns the new Session, the whole composer card activates the existing `conversation.hero.workspace` picker by pointer click — the card owns the click handler and its disabled controls let pointer events fall through, so the full capsule is one target — and the read-only resident textarea does the same by Enter or Space. `aria-haspopup="menu"` and `aria-expanded` describe the shared picker menu while it is mounted. On a fresh installation with no Workspace rows, the picker immediately hands off to the directory dialog and clears its expanded state; that dialog exposes its own accessibility semantics. A dashed l4 stroke (an SVG dash ring, since native `dashed` has a fixed pattern) with a business-blue hover marks the card as the pick affordance. The card contains `pointerdown`, so the open picker's outside-close cannot race the click's reopen — that close-then-open flickered the chip's expansion echo. Message submission, command, permission, model, and other Session-scoped controls remain locked until Workspace selection creates or reconnects a real Session.
|
||||
|
||||
Workspace selection retains the existing owner and flow. `ConversationRoot` opens the picker, `WorkspacePicker` lists or creates the Workspace, and the same textarea DOM node becomes the editable composer after the Session arrives.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep the textarea disabled and emphasize the Workspace chip.** This preserves the old control boundary but leaves the dominant composer surface inert during the first action.
|
||||
|
||||
**Place a transparent button over the textarea.** A button has direct trigger semantics, but it creates a second focusable element over the resident textarea and complicates the DOM-identity transition that preserves focus, IME, and draft behavior.
|
||||
|
||||
**Accept a draft before Workspace selection.** This would require a client-owned draft Session or another pre-Session state axis. The feature only needs a discoverable path into the existing picker.
|
||||
|
||||
## Consequences
|
||||
|
||||
The first composer click now continues the required setup flow, and keyboard users can activate the same path. The textarea accurately reports read-only state until a Session exists, while adjacent controls remain disabled. The UI introduces no new Workspace state, transport, or directory-selection flow.
|
||||
|
||||
Component coverage pins pointer and keyboard activation, the card-wide click target, the contained `pointerdown`, locked adjacent controls, picker expansion, and the same-node transition to an editable textarea. The assembled Web helper begins fresh Workspace setup through the textarea, so replayed browser scenarios exercise the shipped path.
|
||||
@@ -0,0 +1,29 @@
|
||||
# Agent Note: 未选择 Workspace 时从编辑器打开现有选择器
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-07-workspace-picker-composer-entry.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
[Session scope 决策](../architecture/2026-07-25-web-client-session-scope-and-provide-channel.md)会在 Workspace 存在前保留同一个常驻编辑器，但 textarea 处于禁用状态，只有较小的 Workspace chip 能打开选择器。用户首次点击最显眼、也最熟悉的输入区域时，界面不会响应，尽管同一界面已有继续操作的入口。
|
||||
|
||||
## 决策
|
||||
|
||||
新会话尚未归属任何 Workspace 时，整张输入卡片都可通过鼠标点击激活现有的 `conversation.hero.workspace` 选择器——点击处理器归卡片所有，其禁用控件放行指针事件，因此整个胶囊是同一个目标；只读的常驻 textarea 也可经 Enter 或 Space 激活。`aria-haspopup="menu"` 和 `aria-expanded` 在共享选择器菜单挂载时描述其展开状态。全新安装没有 Workspace 行时，选择器会立即转交目录对话框并清除自身的展开状态；该对话框使用自己的可访问性语义。虚线 l4 描边（SVG dash ring，因为原生 `dashed` 的间距不可调）配合 hover 时的 business 蓝，把卡片标记为选择入口。卡片会拦下 `pointerdown`，使已打开选择器的外点关闭无法与点击的重新打开竞态——先关后开会让 chip 的展开回显闪动。消息提交、命令、权限、模型及其他 Session 作用域控件会保持锁定，直到用户选择 Workspace 并创建或重新连接真实 Session。
|
||||
|
||||
Workspace 选择继续使用现有 owner 和流程。`ConversationRoot` 打开选择器，`WorkspacePicker` 列出或创建 Workspace；Session 到达后，同一个 textarea DOM 节点变为可编辑状态。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**保持 textarea 禁用并突出 Workspace chip。** 这样能保留原有控件边界，但首次操作时最主要的编辑器区域仍然没有响应。
|
||||
|
||||
**在 textarea 上方放置透明按钮。** 按钮具备直接的触发器语义，但它会在常驻 textarea 上方增加第二个可聚焦元素，并使保留焦点、输入法和草稿行为的 DOM identity 过渡更复杂。
|
||||
|
||||
**在选择 Workspace 前接收草稿。** 这需要由 client 拥有的草稿 Session 或另一条 Session 前状态轴。此功能只需要提供一个更容易发现的现有选择器入口。
|
||||
|
||||
## 后果
|
||||
|
||||
用户首次点击编辑器即可继续必要的设置流程，键盘用户也能激活同一路径。textarea 会如实报告只读状态，直到 Session 存在；相邻控件仍处于禁用状态。界面没有引入新的 Workspace 状态、传输或目录选择流程。
|
||||
|
||||
组件测试会固定鼠标和键盘激活、覆盖整卡的点击目标、被拦下的 `pointerdown`、相邻控件锁定、选择器展开，以及同一节点变为可编辑 textarea 的过渡。组装后的 Web helper 会通过 textarea 开始全新 Workspace 设置，因此重放浏览器场景会覆盖实际交付路径。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-08-09-parallel-subagent-delegations.md
|
||||
2026-08-09-parallel-subagent-delegations.md: 01dc2043c9a79666857c4f5aa988fc45dc7cdf56
|
||||
2026-08-09-parallel-subagent-delegations.zh.md: 2c2bd700e362000b41adbd1f3cec9ba26ff60b93
|
||||
@@ -0,0 +1,47 @@
|
||||
# Agent Note: Parallel subagent delegations
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-09-parallel-subagent-delegations.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
A model that wants fan-out batches several `subagent` calls into one assistant message — that batch is the parallel intent. The delegation tool declared no `isConcurrencySafe` classifier, so the fail-closed scheduler ([parallel tool-call Agent Note](2026-07-10-parallel-tool-call-execution.md)) treated every foreground delegation as an exclusive barrier: nine cards in the GUI, one child running, eight queued behind it for its full runtime.
|
||||
|
||||
The original conservative stance — a unary classifier cannot prove that sibling delegations have disjoint workspace effects — had stopped protecting anything. `run_in_background: true` and continuable delegations already overlap with every later call, including writes; `dsh-workflow-workerthread` already runs up to its concurrency ceiling of children through the same `ctx.subagents.start()` providers against the shared workspace. Only the foreground variant was serialized.
|
||||
|
||||
## Decision
|
||||
|
||||
`dsh-tool-subagent` declares `isConcurrencySafe: () => true` for every call form (foreground, one-shot background, continuable), so sibling delegations in one assistant step overlap under the loop's rolling pool up to `maxParallelToolCalls`, with results still committed in model order.
|
||||
|
||||
The declaration satisfies the scheduler's safety contract structurally: a child works in its own session, a run never mutates the parent session (the start-time appends — `sandbox/mode`, `approval/policy`, `subagent/descriptor` — land only in the child's own log), and the tool returns its outputs to the loop for ordered commit. The one-shot background form's one parent-owned write is registering a Task through `tasks.start` — a synchronous, commutative insertion that satisfies the scheduler note's shared-state clause rather than the stronger no-mutation property. The provider seam requires concurrent starts and continuable preparations for distinct children to isolate operation-local state, cancellation, settlement, and cleanup. The bundled providers satisfy that contract: spawn and fork keep no mutable state between starts, fork reads only the parent's completed-turn prefix, out-of-process providers allocate state per run, and the continuation manager reserves a unique child identity and lock for each preparation.
|
||||
|
||||
Coordinating sibling workspace effects is the model's responsibility, the stance the product already takes for background, continuable, and workflow children. Peer harnesses agree: Claude Code's Task tool is unconditionally concurrency-safe (cap 10), oh-my-pi's task tool defaults to its overlapping `shared` class, opencode's task tool runs unbounded under its SDK, and Codex sidesteps the question by making delegation an asynchronous spawn/wait mailbox.
|
||||
|
||||
Capacity stays where the scheduler note put it: `maxParallelToolCalls` caps one step's unsettled tool calls — and therefore concurrently running foreground children — while background and continuable calls settle at start and free their pool slot, so children they leave running are not capped by it. LLM providers own their own capacity controls.
|
||||
|
||||
## Testing
|
||||
|
||||
Package tests pin the classifier for both call forms. A gate test drives the registry directly with two children that each block until both have started, proving the half the declaration depends on: the tool body and provider start path tolerate concurrent dispatch — hidden serialization in that stack would deadlock instead of passing silently. A continuable gate holds two provider preparations at the same await, cancels one caller before publication, and proves that the cancelled child leaves no Agent or durable Session while its sibling reaches inbox acceptance and persists independently. The scheduling half, classification actually producing overlap, is owned by the classifier pin and the snapshot below.
|
||||
|
||||
The authored `subagent-parallel` snapshot pins the assembled-app transcript: one assistant message carries two subagent calls, the parent log records `tool/call, tool/call, tool/result, tool/result` (serial execution would interleave call/result pairs), and both children complete as separate sessions. Its twin delegations are deliberately identical: `dsh-llm-replay` binds child scripts by first-call order and the harvester orders children by `createdAt`, and neither is deterministic across concurrent children (the `XXX(concurrent-subagents)` marker), so only interchangeable twins replay race-free today.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep delegations exclusive.** The status quo protected nothing: background and workflow children already overlap freely with writes, so serializing the foreground variant only added latency and contradicted the model's explicit batching intent.
|
||||
|
||||
**An input-sensitive classifier.** The call's arguments are a free-text description and prompt; nothing in them distinguishes a safe delegation from an unsafe one, so a conditional classifier would be theater.
|
||||
|
||||
**A Codex-style asynchronous spawn/wait redesign.** Continuable children plus `send_message` already provide the asynchronous channel; rebuilding the foreground contract around a mailbox would discard a working synchronous result path to solve a scheduling problem one declaration fixes.
|
||||
|
||||
**A per-instance `concurrencySafe` config knob.** No consumer needs a serial deployment: `maxParallelToolCalls: 1` already restores global serial execution, and peer-harness prior art defaults delegation to concurrency-safe.
|
||||
|
||||
## Consequences
|
||||
|
||||
Sibling children can race on shared workspace or external resources; the model owns that coordination, as it already does for every other overlapping child. Concurrent children also compete for LLM provider quota; `maxParallelToolCalls` caps only unsettled calls, not children a background or continuable call left running.
|
||||
|
||||
Two one-shot background delegations in one message acquire their model-visible task ids (`subagent-<n>`) in dispatch-race order. The ids are logged, so replay stays valid, but a snapshot scenario that distinguishes its background children would inherit the same determinism constraint as twin child sessions.
|
||||
|
||||
Ordered commits may hold a fast child's result behind a slow earlier sibling — the trade the [scheduler note](2026-07-10-parallel-tool-call-execution.md) already accepted; live surfaces still show each child's own progress.
|
||||
|
||||
A concurrent-children snapshot scenario with distinct prompts still needs replay-harness support (deterministic child-script binding and harvest ordering); until then such scenarios must use interchangeable twin delegations.
|
||||
@@ -0,0 +1,47 @@
|
||||
# Agent Note: 并行 subagent 委派
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-09-parallel-subagent-delegations.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
想要扇出的模型会把多个 `subagent` 调用合并进同一条 assistant 消息：这个批次本身就是并行意图。委派工具此前没有声明 `isConcurrencySafe` 分类器，按安全侧原则设计的调度器（[并行工具调用 Agent Note](2026-07-10-parallel-tool-call-execution.md)）便把每个前台委派都当作独占屏障：GUI 里显示九张卡片，却只有一个子 agent（智能体）在运行，其余八个要在它的整个运行期间排在其后等待。
|
||||
|
||||
最初的保守立场（一元分类器无法证明同级委派的工作区效果互不相交）已经不再保护任何东西：`run_in_background: true` 和可继续委派本来就会与其后的每个调用重叠执行，包括写入；`dsh-workflow-workerthread` 也早已通过同样的 `ctx.subagents.start()` 提供方在共享工作区上并发运行子 agent，数量可达其并发上限。只有前台形态被串行化。
|
||||
|
||||
## 决策
|
||||
|
||||
`dsh-tool-subagent` 为每种调用形态（前台、一次性后台、可继续）都声明 `isConcurrencySafe: () => true`，因此同一 assistant 步骤中的同级委派会在循环的滚动池下重叠执行，上限为 `maxParallelToolCalls`，结果仍按模型顺序提交。
|
||||
|
||||
该声明在结构上满足调度器的安全约定：子 agent 在自己的会话中工作，运行绝不变更父会话（启动时追加的 `sandbox/mode`、`approval/policy`、`subagent/descriptor` 只落在子 agent 自己的日志里），工具把输出返回给循环，由循环按顺序提交。一次性后台形态对父级拥有状态的唯一写入是通过 `tasks.start` 注册一个 Task——这是一次同步、可交换的插入，满足的是调度器 Agent Note 中的共享状态条款，而非更强的「无变更」性质。提供方 seam 要求针对不同子 agent 的并发启动和可继续准备分别隔离操作局部状态、取消、结算和清理。内置提供方满足这项约定：spawn 和 fork 在各次启动之间不保留可变状态，fork 只读取父级已完成轮次的前缀，进程外提供方按每次运行分配状态，继续执行管理器则为每次准备预留唯一的子 agent 身份和锁。
|
||||
|
||||
协调同级工作区效果是模型的职责，产品对后台、可继续和工作流子 agent 已经采取同样的立场。同类 harness 的做法一致：Claude Code 的 Task 工具无条件并发安全（上限 10）；oh-my-pi 的 task 工具默认归入其可重叠的 `shared` 类别；opencode 的 task 工具在其 SDK 下不设上限地运行；Codex 则把委派做成异步 spawn/wait 信箱，绕开了这个问题。
|
||||
|
||||
容量控制仍保持在调度器 Agent Note 所定的位置：`maxParallelToolCalls` 限制单个步骤中未结算的工具调用数量——因而也限制并发运行的前台子 agent 数量——而后台和可继续调用在启动时即结算并释放池位，它们留下运行的子 agent 不受该上限约束。LLM（大语言模型）提供方负责自身的容量控制。
|
||||
|
||||
## 测试
|
||||
|
||||
包测试固定了两种调用形态的分类器。一个门控测试直接驱动注册表，其两个子 agent 各自阻塞，直到两者都已启动，以此证明该声明所依赖的那一半：工具体和提供方启动路径能容忍并发分发——这条栈中任何隐藏的串行化都会造成死锁，而不是静默通过。一个可继续门控测试让两项提供方准备停在同一个 await 上，在发布前取消其中一个调用方，并证明已取消的子 agent 不会留下 agent 或持久会话，而其同级则到达 inbox 接受状态并独立持久化。另一半（分类真正产生重叠执行）由分类器 pin 测试和下述快照负责。
|
||||
|
||||
人工编写的 `subagent-parallel` 快照固定了组装后应用的 transcript（文本记录）：一条 assistant 消息携带两个 subagent 调用，父级日志记录为 `tool/call, tool/call, tool/result, tool/result`（串行执行会让调用/结果成对交错出现），两个子 agent 各自作为独立会话完成。其中的孪生委派刻意做成完全相同：`dsh-llm-replay` 按首次调用顺序绑定子脚本，harvester 按 `createdAt` 对子 agent 排序，二者在并发子 agent 之间都不具确定性（即 `XXX(concurrent-subagents)` 标记），因此目前只有可互换的孪生委派才能无竞态地回放。
|
||||
|
||||
## 备选方案
|
||||
|
||||
**保持委派独占。** 现状没有保护任何东西：后台和工作流子 agent 本来就可以带着写入自由重叠，串行化前台形态只会增加延迟，还违背模型显式表达的批量意图。
|
||||
|
||||
**使用输入敏感的分类器。** 该调用的参数只有自由文本的描述和提示词；其中没有任何内容能区分安全委派与不安全委派，因此条件式分类器只会流于形式。
|
||||
|
||||
**按 Codex 风格重新设计为异步 spawn/wait。** 可继续子 agent 加上 `send_message` 已经提供了异步通道；围绕信箱重建前台约定，等于为了解决一条声明就能修复的调度问题，丢弃一条可用的同步结果路径。
|
||||
|
||||
**按实例提供 `concurrencySafe` 配置开关。** 没有消费方需要串行部署：`maxParallelToolCalls: 1` 已能恢复全局串行执行，同类 harness 的先例也默认委派并发安全。
|
||||
|
||||
## 影响
|
||||
|
||||
同级子 agent 可能在共享工作区或外部资源上发生竞态；这项协调由模型负责，正如模型对其他所有重叠子 agent 已经承担的那样。并发子 agent 还会争用 LLM 提供方配额；`maxParallelToolCalls` 只限制未结算的调用，不限制后台或可继续调用留下运行的子 agent。
|
||||
|
||||
同一条消息中的两个一次性后台委派按分发竞态顺序获得各自模型可见的 task id（`subagent-<n>`）。这些 id 已被记录，因此回放仍然有效；但需要区分后台子 agent 的快照场景会继承与孪生子会话相同的确定性约束。
|
||||
|
||||
有序提交可能让快速子 agent 的结果排在更早的缓慢同级之后等待，这是[调度器 Agent Note](2026-07-10-parallel-tool-call-execution.md)已经接受的取舍；实时界面仍会展示每个子 agent 各自的进度。
|
||||
|
||||
使用不同提示词的并发子 agent 快照场景仍需要回放 harness 的支持（确定性的子脚本绑定与收集排序）；在此之前，此类场景必须使用可互换的孪生委派。
|
||||
+2
-2
@@ -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/process/2026-07-13-documentation-site-projection.md
|
||||
2026-07-13-documentation-site-projection.md: d9af915754fa6a1df51a27d18d412597472aaa73
|
||||
2026-07-13-documentation-site-projection.zh.md: 7d7b4752b8f27d55aae8426a7dc001ce4340e661
|
||||
2026-07-13-documentation-site-projection.md: 309dbd96c5f1ca87d137cdc3839acfc5c5aa22f2
|
||||
2026-07-13-documentation-site-projection.zh.md: b01965fb1b46ab618b5b8bcb748bb5a8940d6e69
|
||||
@@ -18,9 +18,9 @@ Canonical Markdown remains in the repository tier that owns it. Product-facing g
|
||||
|
||||
Locale home projections retain only the canonical YAML frontmatter. The repository-facing body can keep its H1 and bilingual source links, while the VitePress home theme owns the rendered hero and features and the site navigation owns locale switching.
|
||||
|
||||
The projector parses Markdown links without reserializing the document. A link to another published source becomes a site-relative route; a link to an unpublished repository file becomes a source link under the public `deepseek-ai/deepseek-harness-sdk` home; a repository image is copied into the generated tree and referenced from there ([why](2026-08-06-doc-site-carries-its-images.md)). Missing relative targets fail projection. Unit tests pin these transformations, and `docs:check` runs the projector tests plus a production VitePress build as part of `doc-sync` and the parallel documentation gates.
|
||||
The projector parses Markdown links without reserializing the document. A link to another published source becomes a site-relative route; a link to an unpublished repository file becomes a source link under the `deepseek-ai/deepseek-harness` repository home; a repository image is copied into the generated tree and referenced from there ([why](2026-08-06-doc-site-carries-its-images.md)). Missing relative targets fail projection. Unit tests pin these transformations, and `docs:check` runs the projector tests plus a production VitePress build as part of `doc-sync` and the parallel documentation gates.
|
||||
|
||||
`verify-public-repository-links` rejects internal repository remotes from tracked files. Public source links use the public home, while work tracking stays in repository metadata and source carries a TODO only when the local boundary matters to maintainers.
|
||||
`verify-public-repository-links` rejects references to the unavailable legacy repository from tracked files. Source and edit links use the current repository home.
|
||||
|
||||
`website/AGENTS.md` is the only maintained Markdown file in the website subtree. The projector test enumerates tracked and unignored files and rejects any other website Markdown, so site-specific locale, route, API, or generated source copies cannot bypass the publication manifest.
|
||||
|
||||
|
||||
@@ -18,9 +18,9 @@ Status: implemented
|
||||
|
||||
各 locale 的首页投影只保留权威 YAML frontmatter。面向仓库的正文可以保留其 H1 和双语源文件链接，而 VitePress 首页主题负责渲染 hero 与功能区，网站导航负责切换 locale。
|
||||
|
||||
投影器解析 Markdown 链接，但不会重新序列化文档。指向另一个已发布源文件的链接会变成站内相对路由；指向未发布仓库文件的链接会变成公开 `deepseek-ai/deepseek-harness-sdk` 主页下的源文件链接；仓库图片会被拷贝进生成树并从那里引用（[原因](2026-08-06-doc-site-carries-its-images.md)）。相对目标不存在时，投影会失败。单元测试会锁定这些转换行为，`docs:check` 则运行投影器测试和 VitePress 生产构建，并将二者纳入 `doc-sync` 和并行文档门禁。
|
||||
投影器解析 Markdown 链接，但不会重新序列化文档。指向另一个已发布源文件的链接会变成站内相对路由；指向未发布仓库文件的链接会变成 `deepseek-ai/deepseek-harness` 仓库主页下的源文件链接；仓库图片会被拷贝进生成树并从那里引用（[原因](2026-08-06-doc-site-carries-its-images.md)）。相对目标不存在时，投影会失败。单元测试会锁定这些转换行为，`docs:check` 则运行投影器测试和 VitePress 生产构建，并将二者纳入 `doc-sync` 和并行文档门禁。
|
||||
|
||||
`verify-public-repository-links` 会拒绝已跟踪文件中的内部仓库远程链接。公开源文件链接使用公开主页，而工作跟踪留在仓库元数据中；只有本地边界对维护者有意义时，源文件才保留 TODO。
|
||||
`verify-public-repository-links` 会拒绝已跟踪文件中指向不可用旧仓库的引用。源文件链接和编辑链接使用当前仓库主页。
|
||||
|
||||
`website/AGENTS.md` 是网站子树中唯一维护的 Markdown 文件。投影器测试会枚举所有已跟踪文件和未被忽略的未跟踪文件，并拒绝网站中的任何其他 Markdown，因此网站专用的 locale、路由、API 或生成源文件副本无法绕过发布 manifest。
|
||||
|
||||
|
||||
Reference in New Issue
Block a user