diff --git a/.agents/notes/README.i18n.yaml b/.agents/notes/README.i18n.yaml
index fa9c0d9a21..17bbe70379 100644
--- a/.agents/notes/README.i18n.yaml
+++ b/.agents/notes/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
-README.md: 4db9f16956b9c569cf5f9b53f04cb650f6058668
-README.zh.md: 60ec5421e7f271460daebc966aa6548f6ef8a511
+README.md: 6dec68bef44350895d30058b994ddacb63c70822
+README.zh.md: a9d8e6c74c757271841b3f9ad208fe5e6350d645
diff --git a/.agents/notes/README.md b/.agents/notes/README.md
index 4db9f16956..6dec68bef4 100644
--- a/.agents/notes/README.md
+++ b/.agents/notes/README.md
@@ -37,7 +37,11 @@ The `architecture` / `process` line: **architecture** is about the source we shi
Every non-trivial change MUST add or update at least one Agent Note in the same PR. A change is non-trivial when it alters behavior, architecture, a cross-file or cross-package contract, process or tooling, testing strategy, an on-disk, wire, or configuration format, or another decision a maintainer may reasonably revisit. A proposal for substantial future work starts in `proposed/`; a decision already made starts in `implemented/`. Pick the class folder that matches the decision (see [Classification](#classification)).
-Updating the Agent Note that already owns the decision satisfies the rule; do not create a duplicate. Only a purely mechanical or local edit with no behavioral, contractual, structural, process, or rationale change is exempt. An Agent Note is never edited into a *different decision*: supersede it with a new one and cross-link. Editing an `implemented/` Agent Note to track where its existing decision lives is required, not forbidden; see [implemented/AGENTS.md](implemented/AGENTS.md).
+Updating the Agent Note that already owns the decision satisfies the rule; do not create a duplicate. Only a purely mechanical or local edit with no behavioral, contractual, structural, process, or rationale change is exempt. An Agent Note is never edited into a *different decision*: supersede it with a new one, and keep both notes cross-linked unless the old note is later fully consolidated under the rule below. Editing an `implemented/` Agent Note to track where its existing decision lives is required, not forbidden; see [implemented/AGENTS.md](implemented/AGENTS.md).
+
+An implemented Agent Note that is fully superseded may be consolidated into the current owning note and deleted. Before deletion, the owner must preserve every unique rationale, alternative, consequence, verification contract, and named coverage gap; repair every inbound link; and delete any Chinese counterpart, consistency record, and `required` entry in [the translation-pairing manifest](../../scripts/translation-pairing.manifest.json) in the same change. Partial supersession does not qualify: keep both notes cross-linked and update every fact that remains current. Consolidation must not rewrite the old file into its opposite or rely on git history as the only copy of rationale.
+
+A feature-addition note may be consolidated into the later removal note only when the feature is absent from production code, configuration, schemas, durable or wire formats, migration, and compatibility behavior; no current documentation presents it as available; and no test exercises it as supported behavior. Removal rationale and tests that verify absence may remain. The removal owner preserves the original motivation, why it no longer justified the feature, alternatives to full removal, the capability given up, conditions for reintroduction, and verification of complete absence. Obsolete implementation inventories and tests that only verified the deleted behavior are not current verification contracts. Removing one transport, default, implementation, or presentation is partial supersession, as is any surviving durable data or compatibility handling.
## The file format
diff --git a/.agents/notes/README.zh.md b/.agents/notes/README.zh.md
index 60ec5421e7..a9d8e6c74c 100644
--- a/.agents/notes/README.zh.md
+++ b/.agents/notes/README.zh.md
@@ -39,7 +39,11 @@
每个非平凡变更都必须在同一 PR(Pull Request)中新增或更新至少一份 Agent Note。如果变更修改了行为、架构、跨文件或跨包契约、流程或工具、测试策略、磁盘、协议或配置格式,或者其他维护者可能合理重新审视的决策,就属于非平凡变更。对未来重大工作的提案从 `proposed/` 开始;已经做出的决策从 `implemented/` 开始。选择与决策匹配的类别文件夹(见[分类](#classification))。
-更新已经拥有该决策的 Agent Note 即可满足规则;不要创建重复记录。只有不涉及行为、契约、结构、流程或理由变化的纯机械性或局部编辑才可豁免。Agent Note 永远不会被编辑为一个*不同的决策*:用新 Agent Note 取代旧的,并互相链接。编辑 `implemented/` Agent Note 以跟踪其现有决策的所在位置是必需的,而非禁止的;见 [implemented/AGENTS.md](implemented/AGENTS.md)。
+更新已经拥有该决策的 Agent Note 即可满足规则;不要创建重复记录。只有不涉及行为、契约、结构、流程或理由变化的纯机械性或局部编辑才可豁免。Agent Note 永远不会被编辑为一个*不同的决策*:用新 Agent Note 取代旧记录,并让两个记录保持互相链接,除非后续依据下方规则完全合并旧记录。编辑 `implemented/` Agent Note 以跟踪其现有决策的所在位置是必需的,而非禁止的;见 [implemented/AGENTS.md](implemented/AGENTS.md)。
+
+被完全取代的 implemented Agent Note 可以合并到当前持有该决策的记录中,并删除原文件。删除前,当前记录必须保存所有独有的决策依据、备选方案、影响、验证契约和明确指出的覆盖缺口;修复所有入站链接;并在同一变更中删除中文对侧文件、一致性记录,以及[翻译配对 manifest(元数据清单)](../../scripts/translation-pairing.manifest.json)中对应的 `required` 条目。仅部分被取代的记录不符合此条件:保留两个记录并让它们互相链接,同时更新所有仍然适用的事实。合并不得将旧文件改写成与其相反的决策,也不得让 git 历史成为决策依据的唯一副本。
+
+只有当一项功能已从生产代码、配置、schema、持久化格式或协议格式、迁移和兼容行为中完全消失,当前文档不再将其描述为可用,且没有测试把它作为受支持行为来执行时,新增该功能的 Agent Note 才可合并进后续的移除记录。移除决策的依据和验证该功能已不存在的测试可以保留。移除决策的持有记录必须保留最初动机、为什么该动机已不足以证明保留该功能的合理性、完全移除之外的备选方案、放弃的能力、重新引入的条件,以及证明已彻底移除的验证。过时的实现清单和只验证已删除行为的测试不属于当前验证契约。仅移除一种传输、默认值、实现或展示属于部分取代;仍有任何持久数据或兼容处理也同样如此。
diff --git a/.agents/notes/implemented/AGENTS.md b/.agents/notes/implemented/AGENTS.md
index 5fb3fde8e2..c34e1a49b8 100644
--- a/.agents/notes/implemented/AGENTS.md
+++ b/.agents/notes/implemented/AGENTS.md
@@ -8,4 +8,4 @@ Keep paths, symbols, defaults, and mechanisms current in the same change that al
### This is not a license to rewrite the *decision*
-Update factual realization in place. A reversal of the decision or its rationale requires a new Agent Note and cross-link; see the [Agent Note contract](../README.md).
+Update factual realization in place. A reversal of the decision or its rationale requires a new Agent Note and cross-link; a fully superseded old note may be deleted only through the consolidation rule in the [Agent Note contract](../README.md).
diff --git a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.md b/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.md
deleted file mode 100644
index 947d53555d..0000000000
--- a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.md
+++ /dev/null
@@ -1,23 +0,0 @@
-# Agent Note: Custom typed tool-schema DSL instead of schemastery
-
-Status: implemented
-
-English | [中文](2026-06-11-custom-schema-dsl.zh.md)
-
-## Problem
-
-Tool parameters must reach the model as standard JSON Schema while giving tool authors typed `execute(args)` without casts. Schemastery already serves plugin config, but the tool-author API needs per-property `required: true` booleans rather than JSON Schema's separate `required` array.
-
-## Decision
-
-This decision is superseded by the [unified JSON-value schema DSL](2026-07-20-unified-json-value-schema-dsl.md), which retains the small authoring surface while making parameters and typed values share one vocabulary. `ParameterSchemaSpec` keeps per-property `required: true`; `InferArgs` maps required keys to non-optional properties; `parameterSchemaSpecToJsonSchema()` compiles the implicit open object root; and `defineTool()` ties inference, compilation, and validation together. Raw JSON-Schema `ToolDefinition`s remain accepted by `ToolRegistry.register()` for MCP and other external tools.
-
-## Alternatives considered
-
-**Schemastery** (already vendored, used for plugin Config) was evaluated and rejected for this use: it targets validation / transformation against StandardSchema, not JSON Schema *generation*, so it would add indirection without producing the wire format cleanly.
-
-## Consequences
-
-- First-party tool authors get zero-cast typed args; the type gymnastics cost stays inside the core package (sanctioned by the AGENTS.md type-safety policy).
-- The owning unified note defines the current nodes, literal constraints, unions, JSON-value boundary, and object-openness rules.
-- The `InferArgs` mapping is regression-tested at the type level after an early optionality bug.
diff --git a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.zh.md b/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.zh.md
deleted file mode 100644
index 26ebfe2fb1..0000000000
--- a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.zh.md
+++ /dev/null
@@ -1,23 +0,0 @@
-# Agent Note: 使用自定义类型化工具 schema DSL 替代 schemastery
-
-Status: implemented
-
-[English](2026-06-11-custom-schema-dsl.md) | 中文
-
-## 问题
-
-工具参数必须以标准 JSON Schema 形式到达模型,同时让工具作者在 `execute(args)` 中获得类型化的参数而无需类型断言。Schemastery 已用于插件配置,但工具作者 API 需要逐属性的 `required: true` 布尔值,而非 JSON Schema 的独立 `required` 数组。
-
-## 决策
-
-该决策已由[统一 JSON 值 schema DSL](2026-07-20-unified-json-value-schema-dsl.md)取代;新设计保留小型编写接口,同时让参数与类型化值共享一套词汇。`ParameterSchemaSpec` 保留逐属性的 `required: true`;`InferArgs` 将必需键映射为非可选属性;`parameterSchemaSpecToJsonSchema()` 编译隐式开放的对象根;`defineTool()` 则将类型推导、编译与校验串联起来。原始 JSON Schema 的 `ToolDefinition` 仍是 `ToolRegistry.register()` 接受的输入,供 MCP 和其他外部工具使用。
-
-## 曾考虑的替代方案
-
-**Schemastery**(已作为 vendor 引入,用于插件 Config)经评估后被否决:它面向的是基于 StandardSchema 的校验/转换,而非 JSON Schema *生成*,因此会增加间接层却无法干净地产出协议格式(wire format)。
-
-## 后果
-
-- 第一方工具作者获得零类型断言的类型化参数;类型体操的成本留在核心包内部(符合 AGENTS.md 的类型安全策略)。
-- 当前节点、字面量约束、联合类型、JSON 值边界与对象开放性规则均由上述统一说明定义。
-- `InferArgs` 映射在类型层面有回归测试,源于早期一个可选性 bug。
diff --git a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.i18n.yaml
index 8b9aaea118..33a0bce890 100644
--- a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.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
-2026-06-14-session-persistence.md: 2683bcd68e1f52fbd2bc78660fbe17e556a9a60d
-2026-06-14-session-persistence.zh.md: 28e32925a5866fe520c0d9c3eb7b0647845d36f5
+2026-06-14-session-persistence.md: 52434930bb662b0c97e61f7c2f69b67c309b6317
+2026-06-14-session-persistence.zh.md: 143b58d32191108d7ba24b489bd4f898b1547aab
diff --git a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.md b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.md
index 2683bcd68e..52434930bb 100644
--- a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.md
+++ b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.md
@@ -6,7 +6,7 @@ English | [中文](2026-06-14-session-persistence.zh.md)
## Problem
-Sessions lived only in memory. The example `session-jsonl.ts` plugin (duplicated byte-for-byte in both examples) was write-only telemetry: it buffered `session/event` and appended JSON lines, with no read/replay path, no crash-safety (no fsync, no atomic write, a fire-and-forget dispose drain), no listing, and no format versioning. Nothing could rehydrate a past session from disk into a live agent, so durable resume ("continue yesterday's task"), durable forking, and the ACP `session/load` method ([ACP support](../feature/2026-06-14-acp-agent-client-protocol.md)) were all impossible.
+Sessions lived only in memory. The example `session-jsonl.ts` plugin (duplicated byte-for-byte in both examples) was write-only telemetry: it buffered `session/event` and appended JSON lines, with no read/replay path, no crash-safety (no fsync, no atomic write, a fire-and-forget dispose drain), no listing, and no format versioning. Nothing could rehydrate a past session from disk into a live agent, so durable resume, durable forking, and host-side session browsing were all impossible.
The [event-sourced model](2026-06-11-event-sourced-sessions.md) makes the append-only log the single source of truth and derives LLM history from it. Persistence had to stay faithful to that: persist the existing `SessionEvent` directly, with no parallel "persisted message" type that the log is converted to and from. The backend also had to be swappable — a file store now, a database store later — behind one interface.
@@ -33,4 +33,4 @@ Format versioning: the header carries a `version`; `load` rejects any non-curren
## Consequences
-Two new packages and the metadata seam in `dsh-session` (`session.header`, the `create(id?, options?)` signature). Bought: durable resume/fork, a read/replay path, crash tolerance, and the foundation the ACP `session/load` ([ACP support](../feature/2026-06-14-acp-agent-client-protocol.md)) needs — all over the existing event-sourced log, with the backend swappable behind one interface. The reusable `runPersistenceContract` suite holds every backend to the same append-only, contiguous-seq, lazy-materialization, integer-metadata, and serializability semantics. Persisting the full log also settles event fidelity: `assistant/chunk` remains verbatim. SQLite initialization either commits its complete owned schema and header identity or leaves no partial schema to strand on the next open.
+Two new packages and the metadata seam in `dsh-session` (`session.header`, the `create(id?, options?)` signature). Bought: durable resume/fork, a read/replay path, crash tolerance, and host-side session access over the existing event-sourced log, with the backend swappable behind one interface. The reusable `runPersistenceContract` suite holds every backend to the same append-only, contiguous-seq, lazy-materialization, integer-metadata, and serializability semantics. Persisting the full log also settles event fidelity: `assistant/chunk` remains verbatim. SQLite initialization either commits its complete owned schema and header identity or leaves no partial schema to strand on the next open.
diff --git a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.zh.md b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.zh.md
index 28e32925a5..143b58d321 100644
--- a/.agents/notes/implemented/architecture/2026-06-14-session-persistence.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-14-session-persistence.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-会话此前仅存在于内存中。示例插件 `session-jsonl.ts`(在两个示例中逐字节重复)是只写的遥测:它缓冲 `session/event` 并追加 JSON 行,没有读取/回放路径,没有崩溃安全性(无 fsync、无原子写入、fire-and-forget 的 dispose 排空),没有列表功能,也没有格式版本控制。没有任何机制能将磁盘上的历史会话重新注入到活跃的 agent(智能体)中,因此持久恢复(「继续昨天的任务」)、持久 fork 以及 ACP(Agent Client Protocol)的 `session/load` 方法([ACP 支持](../feature/2026-06-14-acp-agent-client-protocol.md))都无法实现。
+会话此前仅存在于内存中。示例插件 `session-jsonl.ts`(在两个示例中逐字节重复)是只写的遥测:它缓冲 `session/event` 并追加 JSON 行,没有读取/回放路径,没有崩溃安全性(无 fsync、无原子写入、fire-and-forget 的 dispose 排空),没有列表功能,也没有格式版本控制。没有任何机制能将磁盘上的历史会话重新注入到活跃的 agent(智能体)中,因此持久恢复、持久 fork 以及宿主侧的会话浏览都无法实现。
[事件溯源模型](2026-06-11-event-sourced-sessions.md)将仅追加日志作为唯一真源,并从中派生 LLM(大语言模型)历史。持久化必须忠实于这一设计:直接持久化现有的 `SessionEvent`,不引入需要来回转换的并行「持久化消息」类型。后端也必须可替换——当前用文件存储,以后用数据库存储——统一在一个接口之后。
@@ -33,4 +33,4 @@ Status: implemented
## 后果
-新增两个包(package),以及 `dsh-session` 中的元数据 seam(`session.header`,`create(id?, options?)` 签名)。收益:持久恢复/fork、读取/回放路径、崩溃容忍,以及 ACP `session/load`([ACP 支持](../feature/2026-06-14-acp-agent-client-protocol.md))所需的基础——全部基于现有的事件溯源日志,后端在一个接口之后可替换。可复用的 `runPersistenceContract` 测试套件以相同的仅追加、连续 seq、惰性物化、整数元数据与可序列化语义约束每个后端。持久化完整日志还确定了事件保真度:`assistant/chunk` 保持逐字节不变。SQLite 初始化要么提交完整的自有 schema 与 header 标识,要么不留下任何会使下次打开受阻的部分 schema。
+新增两个包(package),以及 `dsh-session` 中的元数据 seam(`session.header`,`create(id?, options?)` 签名)。收益:持久恢复/fork、读取/回放路径、崩溃容忍,以及基于现有事件溯源日志的宿主侧会话访问,后端在一个接口之后可替换。可复用的 `runPersistenceContract` 测试套件以相同的仅追加、连续 seq、惰性物化、整数元数据与可序列化语义约束每个后端。持久化完整日志还确定了事件保真度:`assistant/chunk` 保持逐字节不变。SQLite 初始化要么提交完整的自有 schema 与 header 标识,要么不留下任何会使下次打开受阻的部分 schema。
diff --git a/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.i18n.yaml
index 8c6edb72a9..3b07958faa 100644
--- a/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.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
-2026-06-18-agent-lifecycle-and-ownership-seams.md: 70ebf1c6de97cb14e27377ec1f9bac18fc0766a6
-2026-06-18-agent-lifecycle-and-ownership-seams.zh.md: 54d174aa325384f7c7492f072f10954f0b1d08c9
+2026-06-18-agent-lifecycle-and-ownership-seams.md: f190b4ba2b7f22d29f473c8a2725401ff371488e
+2026-06-18-agent-lifecycle-and-ownership-seams.zh.md: dcaa319232baa8951a4f515abc6bce5611da5576
diff --git a/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.md b/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.md
index 70ebf1c6de..f190b4ba2b 100644
--- a/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.md
+++ b/.agents/notes/implemented/architecture/2026-06-18-agent-lifecycle-and-ownership-seams.md
@@ -18,7 +18,7 @@ A new `cancel()` verb on the `Agent` interface — the single public stop primit
### 2. `AgentHandle` async disposer
-`ctx.agents.create`/`resume` (and the `AgentFactory` interface) return `AgentHandle = { agent: Agent; dispose(): Promise }`. The disposer is a **consumer capability** — a registry observer holding only the bare `Agent` cannot tear it down. The caller fiber and registered factory provider are structural co-owners: caller unload enforces structured ownership, while provider unload must stop old instances whose scoped dependency surface resolves through that provider. All three paths reach the same memoized teardown: stop the loop, await its exit and idle flushes (true quiescence, not just the `disposed` status flip), detach the agent, detach its session, and unwind its scope. Each public ID becomes reusable when its exact registry entry detaches; there is no separate reservation-release phase. Config-created agents are already owned by the `AgentLoop` fiber (the handle is discarded). ACP holds each session's disposer in its `SessionRecord` and runs it on disconnect/teardown, so a bare client disconnect leaves no registered agent and no session-store entry — even when `session/load` races teardown (the just-resumed handle is disposed before the closed-guard throw).
+`ctx.agents.create`/`resume` (and the `AgentFactory` interface) return `AgentHandle = { agent: Agent; dispose(): Promise }`. The disposer is a **consumer capability** — a registry observer holding only the bare `Agent` cannot tear it down. The caller fiber and registered factory provider are structural co-owners: caller unload enforces structured ownership, while provider unload must stop old instances whose scoped dependency surface resolves through that provider. All three paths reach the same memoized teardown: stop the loop, await its exit and idle flushes (true quiescence, not just the `disposed` status flip), detach the agent, detach its session, and unwind its scope. Each public ID becomes reusable when its exact registry entry detaches; there is no separate reservation-release phase. Config-created agents are already owned by the `AgentLoop` fiber (the handle is discarded). ACP holds each fresh session's disposer in its `SessionRecord` and runs it on disconnect or plugin teardown, so a bare client disconnect leaves no registered agent and no session-store entry. A create that loses the close race disposes its unpublished handle.
**Teardown ORDER is load-bearing for durability**, and the implementation folds the session lifecycle into the agent's SINGLE composite cordis effect (`SessionStore.prepare`/`enter`/`announce`, replacing a sibling-effect split). A fiber unload disposes sibling effects concurrently (`Promise.all`), which would race removing the session store's append publication hooks against the loop's closing `session/flush` and drop the closing `turn/end`; inside one effect the disposers run as an ordered LIFO chain (loop stopped + `await agent.done` BEFORE the session detaches), so the loop's final flush is captured on BOTH the handle's `dispose()` and a fiber unload. The contained `agent/disposed` and `session/disposed` notifications cannot reject the chain or skip later teardown.
@@ -30,7 +30,7 @@ Background-task ownership moved from a `tool-bash` plugin-local `Map }`。释放器是一种**消费方能力**——仅持有裸 `Agent` 的注册表观察者无法将其拆除。调用方 fiber 和已注册的 factory 提供方是结构上的共同所有者:调用方卸载强制结构化所有权,而提供方卸载必须停止旧实例,因为其实例作用域的依赖 surface 通过该提供方解析。三条路径都会进入同一个 memoize 的拆除过程:停止循环、等待其退出与空闲刷写完成(完全停稳,而非仅把状态翻转为 `disposed`)、分离 agent、分离其会话,然后解除其 scope。每个公开 ID 在其精确注册表条目分离时变得可复用;不存在独立的保留释放阶段。由配置创建的 agent 已归 `AgentLoop` fiber 所有(handle 被丢弃)。ACP 在其 `SessionRecord` 中保存每个会话的释放器,并在断连/拆除时运行它,因此单纯的客户端断连不会留下已注册 agent 或会话存储条目——即使 `session/load` 与拆除竞争(刚恢复的 handle 会在 closed-guard 抛出前释放)。
+`ctx.agents.create`/`resume`(以及 `AgentFactory` 接口)返回 `AgentHandle = { agent: Agent; dispose(): Promise }`。释放器是一种**消费方能力**——仅持有裸 `Agent` 的注册表观察者无法将其拆除。调用方 fiber 和已注册的 factory 提供方是结构上的共同所有者:调用方卸载强制结构化所有权,而提供方卸载必须停止旧实例,因为其实例作用域的依赖 surface 通过该提供方解析。三条路径都会进入同一个 memoize 的拆除过程:停止循环、等待其退出与空闲刷写完成(完全停稳,而非仅把状态翻转为 `disposed`)、分离 agent、分离其会话,然后解除其 scope。每个公开 ID 在其精确注册表条目分离时变得可复用;不存在独立的保留释放阶段。由配置创建的 agent 已归 `AgentLoop` fiber 所有(handle 被丢弃)。ACP 在其 `SessionRecord` 中保存每个全新会话的释放器,并在断连或插件拆除时运行它,因此单纯的客户端断连不会留下已注册 agent 或会话存储条目。在与关闭的竞态中落败的创建流程会 dispose 其尚未发布的 handle。
**拆除顺序对持久性至关重要**,实现将会话生命周期折叠进 agent 的单个复合 Cordis effect(`SessionStore.prepare`/`enter`/`announce`,取代兄弟 effect 拆分)。fiber 卸载会并发释放兄弟 effect(`Promise.all`),这会让会话存储的 append 发布钩子移除与循环关闭时的 `session/flush` 竞争,从而丢失关闭的 `turn/end`;在一个 effect 内,释放器作为有序的 LIFO 链运行(停止循环 + `await agent.done` 在会话分离之前),因此无论 handle 的 `dispose()` 还是 fiber 卸载,都会捕获循环的最终刷写。被隔离的 `agent/disposed` 和 `session/disposed` 通知无法拒绝该链或跳过后续拆除。
@@ -30,7 +30,7 @@ ACP(Agent Client Protocol)与 tool-bash 的若干限制是同一个缺失 se
以下不变式已经成立,并由测试固定:
-- ACP 断连/会话关闭后,不留下该会话的任何已注册 agent 或会话存储条目,即使 `session/load` 与拆除竞争。
+- ACP 断连或插件拆除后,任何由桥接层拥有的会话都不留下已注册 agent 或会话存储条目,包括与连接关闭竞争的创建流程。
- 已入队的提示词启动前执行 `session/cancel`,能阻止该提示词运行;后来接受的提示词仍是独立的已入队轮次。
- `tool-bash` HMR 重载不会使另一个会话能够读取或终止已有的后台任务(所有权保留在执行器上)。
- 既有的非 ACP 演示无需显式管理 handle 仍能工作;由配置创建的 agent 仍归 `AgentLoop` 插件 fiber 所有。
diff --git a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.i18n.yaml
index b289d47edd..d8dae837fe 100644
--- a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.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
-2026-06-18-shared-persistence-write-coordinator.md: ea9c4fb74f7c1bd68fb62efedd3e1657da96ea65
-2026-06-18-shared-persistence-write-coordinator.zh.md: 3b4dd7b762c2f39a908eabe23e5d734981b5767b
+2026-06-18-shared-persistence-write-coordinator.md: 4632351a6f39c44c9ba8af58d508d4665b9e9279
+2026-06-18-shared-persistence-write-coordinator.zh.md: 40a7144038ac0db4ca6cac651c0a3cef5de4afa9
diff --git a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.md b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.md
index ea9c4fb74f..4632351a6f 100644
--- a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.md
+++ b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.md
@@ -23,7 +23,7 @@ The coordinator retires a session from `session/disposed`: it waits for the cont
Five required members plus an optional lifecycle hook form the only boundary between the coordinator and storage:
- `name` — backend label for the dispose-failure `AggregateError`.
-- `loadStored(id)` — read one stored prefix by id across every storage scope (every JSONL cwd bucket; SQLite's id is globally unique). Resume/load, non-mutating inspection, live adoption, and the create-collision probe share this lookup. The coordinator asserts the returned id and rejects a stored/live cwd mismatch before repair or state publication.
+- `loadStored(id)` — read one stored prefix by id across every storage scope (every JSONL project directory; SQLite's id is globally unique). Resume/load, non-mutating inspection, live adoption, and the create-collision probe share this lookup. The coordinator asserts the returned id and rejects a stored/live cwd mismatch before repair or state publication.
- `appendBatch(meta, events, isMaterialized)` — durably append a contiguous batch, lazily materializing the session ATOMICALLY when not yet materialized (the materialize-write and the first event batch must commit together — a crash between them must not leave a materialized-but-empty session; this is why there is no separate `materialize` hook).
- `commitRepair(meta, tornMarker, closers)` — make a crash repair durable: truncate the torn tail (iff `tornMarker !== undefined`) and append `closers`. **NOT required to be atomic** — JSONL legitimately truncates-then-appends in two fsync'd steps, SQLite does DELETE+INSERT in one transaction. Used by `load` (truncate + synthetic closers) and live-adoption (truncate only, `closers = []`).
- `list()` — list all stored metadata.
diff --git a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.zh.md b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.zh.md
index 3b4dd7b762..40a7144038 100644
--- a/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-18-shared-persistence-write-coordinator.zh.md
@@ -23,7 +23,7 @@ Status: implemented
五个必需成员加一个可选的生命周期钩子,构成协调器与存储之间唯一的边界:
- `name`——后端标签,用于 dispose 失败时的 `AggregateError`。
-- `loadStored(id)`——按 id 跨所有存储范围读取一个已存储前缀(JSONL 的所有 cwd bucket;SQLite 的 id 全局唯一)。恢复/加载、不修改状态的检查、存活会话接管与创建碰撞探测共用此查找。协调器会断言返回的 id,并在修复或发布状态之前拒绝已存储记录与存活会话的 cwd 不匹配。
+- `loadStored(id)`——按 id 跨所有存储范围读取一个已存储前缀(JSONL 的所有项目目录;SQLite 的 id 全局唯一)。恢复/加载、不修改状态的检查、存活会话接管与创建碰撞探测共用此查找。协调器会断言返回的 id,并在修复或发布状态之前拒绝已存储记录与存活会话的 cwd 不匹配。
- `appendBatch(meta, events, isMaterialized)`——持久追加一个连续批次,在尚未物化时原子地惰性物化会话(物化写入与首批事件必须一起提交——崩溃不得留下一个已物化但为空的会话;这就是为什么没有单独的 `materialize` 钩子)。
- `commitRepair(meta, tornMarker, closers)`——使崩溃修复持久化:截断损坏的尾部(当且仅当 `tornMarker !== undefined`)并追加 `closers`。**不要求原子性**——JSONL 合理地分两步 fsync(先截断再追加),SQLite 在一个事务中完成 DELETE+INSERT。用于 `load`(截断 + 合成 closers)和 live-adoption(仅截断,`closers = []`)。
- `list()`——列出所有已存储的元数据。
diff --git a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.i18n.yaml
index 35c9875149..04f7dcb1a3 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.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
-2026-06-20-branded-ids.md: 93bab1d47c793cc4dd1f1d19fa3af22d1721be29
-2026-06-20-branded-ids.zh.md: 6cc8a45717b547614a873d3f13c54c3950484a22
+2026-06-20-branded-ids.md: 7c0b7ca89418e8312ec728223dac519f70edc3ed
+2026-06-20-branded-ids.zh.md: 8b41ad3c3c85690fb03b20a208f8460a1614477b
diff --git a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.md b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.md
index 93bab1d47c..7c0b7ca894 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.md
@@ -12,7 +12,7 @@ The harness brands `CallId` (`packages/llm/llm/src/brand.ts`) and the shared age
The bash **owner token** is the related sub-case: `BashExecRequest.owner?: string` and `BashExecSpec.owner: string | undefined` (`packages/bash/bash/src/types.ts`) are documented as a deliberately *opaque* isolation key, but in every live caller the value IS the owning agent's shared `Agent.id`/`SessionId` (`callerToken = (exec) => exec.agent?.id` in `packages/bash/tool-bash/src/index.ts`) wearing a different seam-local name. It is compared for access control (`owner !== callerToken(exec)`), so a mismatched-but-well-typed string here is a cross-session isolation bug the type system currently cannot catch. This is the shared id alias covered by the [unified agent/session identity decision](../simplification/2026-06-20-unify-agent-and-session-id.md).
-**Gap 2 — brand erosion at the seams of the *already-branded* IDs.** Even `CallId` and `SessionId` decay back to bare `string` at exactly the places confusion is most likely: registry/store key types and public method params. Representative sites include the session store, the agent registry (both keyed by the shared `SessionId`), `ToolPresenter`'s call-id map, ACP's session-id records and loading set, and the persistence coordinator. A brand that is dropped at a collection key buys nothing on lookups — the value of the existing brands is partly unrealized.
+**Gap 2 — brand erosion at the seams of the *already-branded* IDs.** Even `CallId` and `SessionId` decay back to bare `string` at exactly the places confusion is most likely: registry/store key types and public method params. Representative sites include the session store, the agent registry (both keyed by the shared `SessionId`), tool-presentation call-id maps, ACP's session records, and the persistence coordinator. A brand that is dropped at a collection key buys nothing on lookups — the value of the existing brands is partly unrealized.
## Decision
diff --git a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.zh.md b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.zh.md
index 6cc8a45717..8b41ad3c3c 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-branded-ids.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-branded-ids.zh.md
@@ -12,7 +12,7 @@ harness 使用 `Branded = string & { readonly [BRAND]: B }` 机制,为 `Cal
bash **owner token** 是相关的子情形:`BashExecRequest.owner?: string` 和 `BashExecSpec.owner: string | undefined`(`packages/bash/bash/src/types.ts`)被文档描述为刻意*不透明*的隔离键,但在所有实际调用方中,该值就是所属 agent(智能体)共享的 `Agent.id`/`SessionId`(`callerToken = (exec) => exec.agent?.id`,位于 `packages/bash/tool-bash/src/index.ts`),只是披着另一个 seam 本地名称。它被用于访问控制比较(`owner !== callerToken(exec)`),因此一个不匹配但类型正确的 string 在此处就是跨会话隔离 bug,而当前类型系统无法捕获。这正是[统一 agent/session 标识决策](../simplification/2026-06-20-unify-agent-and-session-id.md)覆盖的共享 id 别名。
-**缺口 2:*已经 brand* 的 ID 在 seam 处被侵蚀。** 就连 `CallId` 和 `SessionId` 也恰好在最容易混淆的地方退化为裸 `string`:注册表/store 键类型和公开方法参数。代表性位置包括会话存储、agent 注册表(二者都以共享的 `SessionId` 为键)、`ToolPresenter` 的 call-id map、ACP 的会话 id 记录和 loading set,以及持久化协调器。在集合键处丢弃 brand,会让既有 brand 在查找时毫无价值;它们的价值只实现了一部分。
+**缺口 2:*已经 brand* 的 ID 在 seam 处被侵蚀。** 就连 `CallId` 和 `SessionId` 也恰好在最容易混淆的地方退化为裸 `string`:注册表/store 键类型和公开方法参数。代表性位置包括会话存储、agent 注册表(二者都以共享的 `SessionId` 为键)、工具展示层的 call-id map、ACP 的会话记录,以及持久化协调器。在集合键处丢弃 brand,会让既有 brand 在查找时毫无价值;它们的价值只实现了一部分。
## 决策
diff --git a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.i18n.yaml
index 35bc3836ff..a27551cd40 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.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
-2026-06-20-extract-example-app-packages.md: 8e87a7164d9cc4789705b3d8fe83dc4978b4def0
-2026-06-20-extract-example-app-packages.zh.md: bb5bf6a78d749bf2fc0637a48bff4496399a787d
+2026-06-20-extract-example-app-packages.md: f2853db3f454d71572be003cfbf4f6dfd8377cdd
+2026-06-20-extract-example-app-packages.zh.md: 58d3d95996b1dacbc12178b46524374429df71ed
diff --git a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.md b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.md
index 8e87a7164d..f2853db3f4 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.md
@@ -42,7 +42,7 @@ The old `base*.yml`/`acp-tail.yml` includes already deduped the *config*, but a
- Example directories contain only their config, README, and tests: `start.ts`, the infrastructure preamble, and the shared YAML includes are gone.
- `demo:tui`, `demo:headless`, and `demo:acp` invoke the app-package bins.
- Each new package has a README and per-file 100% coverage; each app package also has a keyless real-Loader-path bin smoke that catches export-shape failures described in [postmortem 0001](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md).
-- The ACP replay transcript remains unchanged because the plugin set and load order did not change.
+- The ACP replay suite boots through the app-package bin, so protocol wiring and assembled backend behavior cross the real Loader boundary.
## Consequences
diff --git a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.zh.md b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.zh.md
index bb5bf6a78d..58d3d95996 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-extract-example-app-packages.zh.md
@@ -42,7 +42,7 @@ Status: implemented
- 示例目录只包含配置、README 和测试:`start.ts`、基础设施前导和共享 YAML include 已移除。
- `demo:tui`、`demo:headless` 和 `demo:acp` 调用应用包的 bin。
- 每个新包都有 README 和逐文件 100% 覆盖率;每个应用包还有一个 keyless 的真实 Loader 路径 bin 冒烟测试,用于捕获[事后分析 0001](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md) 中描述的导出形状故障。
-- ACP 回放 transcript(文本记录)保持不变,因为插件集合和加载顺序未改变。
+- ACP 回放套件通过应用包的 bin 启动,因此协议接线与组装后的后端行为都跨越真实的 Loader 边界。
## 后果
diff --git a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.i18n.yaml
index ec8a2059a2..d44e3ffee9 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-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
-2026-06-20-generic-long-running-tool-runtime.md: 25668a6a699576e435670b9580385073f2f036fe
-2026-06-20-generic-long-running-tool-runtime.zh.md: 3c214bb4309ea331ae2d4ba73c01df1ded8f40ef
+2026-06-20-generic-long-running-tool-runtime.md: 0b901fcf928b900bd3a32f911e6e54a6a98076e2
+2026-06-20-generic-long-running-tool-runtime.zh.md: e2860e3a91c06ec5110cd671b288e35c5d117f5d
diff --git a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.md b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.md
index 25668a6a69..0b901fcf92 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.md
@@ -69,7 +69,7 @@ A producer loaded without any control surface would let callers start work they
## Model-facing control surface
-`dsh-tool-tasks` registers three kind-independent tools with generic ACP cards:
+`dsh-tool-tasks` registers three kind-independent tools with generic UI cards:
- `task_output(task_id, wait?, timeout_ms?)` reads output and always appends `[status: ...]`. Stream tasks return only output since the previous read; final-output tasks return their result after settlement. Reads are non-blocking unless `wait: true`, whose timeout is defaulted and capped by plugin config. A wait timeout reports the still-running status and does not stop the task.
- `task_list()` returns caller-visible tasks as ` [] — `, or `(no background tasks)`.
diff --git a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.zh.md b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.zh.md
index 3c214bb430..e2860e3a91 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-generic-long-running-tool-runtime.zh.md
@@ -69,7 +69,7 @@ task id 在运行时全局可见且可预测,因此注册表会授权每次访
## 面向模型的控制接口
-`dsh-tool-tasks` 注册三个与 kind 无关的工具,并使用通用 ACP(Agent Client Protocol)卡片:
+`dsh-tool-tasks` 注册三个与 kind 无关的工具,并使用通用 UI 卡片:
- `task_output(task_id, wait?, timeout_ms?)` 读取输出,并始终追加 `[status: ...]`。流式任务只返回上次读取以来的输出;最终输出任务在结算后返回结果。除非指定 `wait: true`,否则读取不会阻塞;等待超时由插件配置提供默认值并限定上限。等待超时会报告仍在运行的状态,不会停止任务。
- `task_list()` 将调用方可见的任务返回为 ` [] — `,没有任务时返回 `(no background tasks)`。
diff --git a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.i18n.yaml
index 63ef106781..03a271a9c3 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.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
-2026-06-20-package-hierarchy.md: ab0732172934414f2361b9e4976b54e5daee298c
-2026-06-20-package-hierarchy.zh.md: 57058860de5b4cba3e8a45ecad9d611d068fb82c
+2026-06-20-package-hierarchy.md: 4e05e3487483ab8d710959c1888ec1f5c3b37432
+2026-06-20-package-hierarchy.zh.md: f57704ad082c4961aa48056af1b4b279d2f2c055
diff --git a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.md b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.md
index ab07321729..4e05e34874 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.md
@@ -4,7 +4,7 @@ Status: implemented
English | [中文](2026-06-20-package-hierarchy.zh.md)
-The later [fold-stdio-helper](../simplification/2026-07-04-fold-stdio-ui-helper.md) decision superseded the original `support/ui-stdio` placement, and the [redundant-agent removal](../simplification/2026-07-20-remove-stdio-and-echo-agents.md) subsequently removed that surface entirely. The uniform depth-two hierarchy remains the decision owned here.
+The [redundant-agent removal](../simplification/2026-07-20-remove-stdio-and-echo-agents.md) deletes the original `support/ui-stdio` surface instead of relocating it, and the [automation-only ACP decision](../simplification/2026-07-23-acp-automation-only-protocol.md) places ACP under `packages/acp/acp` instead of the human-UI group. The uniform depth-two hierarchy remains the decision owned here.
## Problem
@@ -36,8 +36,9 @@ packages/
session-persistence/
session-persistence-jsonl/
session-persistence-sqlite/
- ui/ (product integration)
+ acp/ (product automation integration)
acp/
+ ui/ (human interaction and presentation)
support/ (dev/test/example infrastructure)
invariants/
ui-stdio/
@@ -49,7 +50,7 @@ packages/
- **Same-name nesting for capability families.** A family's interface package sits at `packages///` (`llm/llm`, `bash/bash`, `session-persistence/session-persistence`), with implementations and consumers as flat siblings. There is no extra `adapters/`/`impls/` sub-tier — every package is exactly depth 2, which keeps the workspace glob a clean `packages/*/*` and lets one `@deepseek-ai/dsh-*` tsconfig wildcard resolve every package (unique dir names make first-on-disk-wins unambiguous).
- **`session` stays in `core/`; persistence is its own family.** The session log is core product API. Its storage backends form a parallel capability family (`session-persistence/`) mirroring `llm/` and `bash/`, rather than nesting under `core/session/`.
- **`agent-loop` is in `core/`.** It is the one concrete implementation of the `agent` seam, but it ships as the harness's default product loop, so it lives with the core spine. Plugins still depend on the `agent` vocabulary, never on `agent-loop`, so the loop stays swappable.
-- **`invariants` and `ui-stdio` are `support/`, not product.** `invariants` is dev-mode contract checking. `ui-stdio` was extracted from the examples for reuse and the coverage gate — it is example-coupled, so it sits in `support/` alongside `llm-replay` (the snapshot-test replay adapter). `acp` is the only `ui/` member because it is a real product surface (the ACP bridge an editor drives), structurally distinct from the readline demo helper.
+- **Product automation and human UI are separate groups.** `acp` is a product transport under `acp/`, while commands, approvals, interaction, and presentation adapters live under `ui/`. Dev-only invariants and replay infrastructure remain under `support/`.
### Deduplicating the package lists
@@ -70,7 +71,7 @@ Two doc-sync/hygiene gates keep the structure and its references honest, so the
- **A third tier (`adapters/` / `impls/` under each family)** — rejected: uniform depth 2 keeps the workspace glob a clean `packages/*/*` and lets one `@deepseek-ai/dsh-*` tsconfig wildcard resolve every package.
- **Nesting persistence under `core/session/`** — rejected: the storage backends form a parallel capability family mirroring `llm/` and `bash/`, while the session log itself stays core product API.
-- **`ui-stdio` under `ui/`** — rejected: it is example-coupled dev support, not a product surface; `acp` is the only `ui/` member because an editor actually drives it.
+- **`ui-stdio` under `ui/`** — rejected: it was example-coupled dev support, not a product surface.
## Consequences
diff --git a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.zh.md b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.zh.md
index 57058860de..f57704ad08 100644
--- a/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-20-package-hierarchy.zh.md
@@ -4,7 +4,7 @@ Status: implemented
[English](2026-06-20-package-hierarchy.md) | 中文
-后续的[折叠 stdio helper](../simplification/2026-07-04-fold-stdio-ui-helper.md)决策取代了最初的 `support/ui-stdio` 放置方式,[冗余 agent 移除](../simplification/2026-07-20-remove-stdio-and-echo-agents.md)随后又彻底移除了该接口。这里拥有的决策仍是统一的二层目录深度。
+[冗余 agent 移除](../simplification/2026-07-20-remove-stdio-and-echo-agents.md)直接删除最初的 `support/ui-stdio` 接口,而不是将其迁移;[仅面向自动化的 ACP 决策](../simplification/2026-07-23-acp-automation-only-protocol.md)把 ACP 放在 `packages/acp/acp` 下,而不是面向人类的 UI 组。这里拥有的决策仍是统一的二层目录深度。
## 问题
@@ -36,8 +36,9 @@ packages/
session-persistence/
session-persistence-jsonl/
session-persistence-sqlite/
- ui/ (product integration)
+ acp/ (product automation integration)
acp/
+ ui/ (human interaction and presentation)
support/ (dev/test/example infrastructure)
invariants/
ui-stdio/
@@ -49,7 +50,7 @@ packages/
- **能力族使用同名嵌套。** 一个族的接口包位于 `packages///`(`llm/llm`、`bash/bash`、`session-persistence/session-persistence`),实现和消费方作为扁平兄弟并列。不设额外的 `adapters/`/`impls/` 子层——每个包恰好在深度 2,这使 workspace glob 保持简洁的 `packages/*/*`,并让一条 `@deepseek-ai/dsh-*` tsconfig 通配符即可解析所有包(唯一的目录名使 first-on-disk-wins 无歧义)。
- **`session` 留在 `core/`;持久化独立成族。** 会话日志是核心产品 API。其存储后端构成一个平行的能力族(`session-persistence/`),与 `llm/` 和 `bash/` 对称,而非嵌套在 `core/session/` 下。
- **`agent-loop` 在 `core/` 中。** 它是 `agent` seam 唯一的具体实现,但作为 harness 的默认产品循环交付,因此与核心主干同处。插件仍然依赖 `agent` 的词汇,从不依赖 `agent-loop`,所以循环仍可替换。
-- **`invariants` 和 `ui-stdio` 属于 `support/`,不是产品。** `invariants` 是开发模式的契约检查。`ui-stdio` 从示例中提取出来以便复用和满足覆盖率门禁——它与示例耦合,因此与 `llm-replay`(快照测试的回放适配器)一起放在 `support/` 中。`acp` 是 `ui/` 的唯一成员,因为它是真正的产品接口(编辑器驱动的 ACP 桥接),与 readline 演示辅助工具在结构上截然不同。
+- **产品自动化与面向人类的 UI 是两个独立分组。** `acp` 是位于 `acp/` 下的产品传输层,而命令、审批、交互和展示适配器位于 `ui/` 下。仅开发用的 invariants 与回放基础设施仍留在 `support/` 中。
### 去重包列表
@@ -70,7 +71,7 @@ packages/
- **第三层(每个族下设 `adapters/`/`impls/`)**:否决。统一深度 2 使 workspace glob 保持简洁的 `packages/*/*`,并让一条 `@deepseek-ai/dsh-*` tsconfig 通配符即可解析所有包。
- **将持久化嵌套在 `core/session/` 下**:否决。存储后端构成一个平行的能力族,与 `llm/` 和 `bash/` 对称,而会话日志本身属于核心产品 API。
-- **`ui-stdio` 放在 `ui/` 下**:否决。它是与示例耦合的开发支撑,不是产品接口;`acp` 是 `ui/` 的唯一成员,因为编辑器实际驱动它。
+- **`ui-stdio` 放在 `ui/` 下**:否决。它曾是与示例耦合的开发支撑,不是产品接口。
## 后果
diff --git a/.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md b/.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md
index b8d06ebf6d..290c35461b 100644
--- a/.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md
+++ b/.agents/notes/implemented/architecture/2026-06-21-bounded-llm-request-recovery.md
@@ -42,7 +42,7 @@ The agent loop keeps `RequestError` as that exact error object and passes `LlmFa
Adapters extract structured facts before falling back to message inspection. They validate HTTP status, parse `Retry-After` seconds or dates into a positive finite millisecond delay, brand the provider request id when exposed, and distinguish their own timeout from the caller's abort. Provider-specific codes and messages may refine a mapping, but no recovery listener parses them.
-The initial shared transient-code set is intentionally small: the adapters' existing `RATE_LIMIT` and `SERVER` mappings plus explicit `TIMEOUT` and `TRANSPORT` codes for the two missing remote-failure families. Authentication, quota, invalid request, context overflow, protocol, abort, and unknown failures keep distinct stable codes and are not transient by default. Adding a code requires adapter fixtures and a documented policy decision; it does not require expanding a second failure-class enum.
+The shared transient-code set is intentionally small: adapter mappings for `RATE_LIMIT` and `SERVER`, explicit `TIMEOUT` and `TRANSPORT` codes for remote failures, and `EMPTY_RESPONSE` for a completed provider response with no content blocks. Both adapters classify the last case as an error finish; see [empty model responses are retryable](../bug-fix/2026-07-24-empty-model-response-is-retryable.md). Authentication, quota, invalid request, context overflow, protocol, abort, and unknown failures keep distinct stable codes and are not transient by default. Adding a code requires adapter fixtures and a documented policy decision; it does not require expanding a second failure-class enum.
### Put retry policy on the existing failed-step seam
@@ -62,7 +62,7 @@ interface Config {
}
```
-The defaults are two transient retries, a 500 millisecond initial delay, a 10 second delay cap, 10 percent jitter, and the four transient codes above. The count and delay bounds match the conservative edge of the inspected implementations: [OpenCode uses two request retries with 500 ms/10 s bounds](https://github.com/anomalyco/opencode/blob/9976269ab1accfc9f9dc98a4a688c516934de422/%70ackages/llm/src/route/executor.ts#L36-L39), [Pi separates three agent-level retries from provider retries and defaults provider retries to zero](https://github.com/earendil-works/pi/blob/3da591ab74ab9ab407e72ed882600b2c851fae21/%70ackages/coding-agent/docs/settings.md#L139-L147), and [Codex uses finite request/stream budgets plus a five-minute idle timeout](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/model-provider-info/src/lib.rs#L25-L33). Ten percent follows [Codex's bounded jitter](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/codex-client/src/retry.rs#L40-L47). Two retries mean at most three provider requests when no other recovery policy applies. `maxTransientRetries` is a non-negative integer, delays are positive finite numbers with `initialDelayMs <= maxDelayMs`, `jitterRatio` is in `[0, 1]`, and codes are non-empty and unique. These are Cordis config fields rather than hidden constants so deployments can choose different cost and latency budgets.
+The defaults are two transient retries, a 500 millisecond initial delay, a 10 second delay cap, 10 percent jitter, and the five transient codes above (`RATE_LIMIT`, `SERVER`, `TIMEOUT`, `TRANSPORT`, and `EMPTY_RESPONSE`). The count and delay bounds match the conservative edge of the inspected implementations: [OpenCode uses two request retries with 500 ms/10 s bounds](https://github.com/anomalyco/opencode/blob/9976269ab1accfc9f9dc98a4a688c516934de422/%70ackages/llm/src/route/executor.ts#L36-L39), [Pi separates three agent-level retries from provider retries and defaults provider retries to zero](https://github.com/earendil-works/pi/blob/3da591ab74ab9ab407e72ed882600b2c851fae21/%70ackages/coding-agent/docs/settings.md#L139-L147), and [Codex uses finite request/stream budgets plus a five-minute idle timeout](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/model-provider-info/src/lib.rs#L25-L33). Ten percent follows [Codex's bounded jitter](https://github.com/openai/codex/blob/0fb559f0f6e231a88ac02ea002d3ecd248e2b515/codex-rs/codex-client/src/retry.rs#L40-L47). Two retries mean at most three provider requests when no other recovery policy applies. `maxTransientRetries` is a non-negative integer, delays are positive finite numbers with `initialDelayMs <= maxDelayMs`, `jitterRatio` is in `[0, 1]`, and codes are non-empty and unique. These are Cordis config fields rather than hidden constants so deployments can choose different cost and latency budgets.
For an eligible failure with budget remaining, the one-based transient retry count uses bounded exponential backoff. A valid `providerRetryAfterMs` replaces exponential backoff only when it does not exceed `maxDelayMs`; a longer provider delay causes delegation instead of an earlier retry that violates the provider instruction. Local backoff multiplies by an injected random factor in `[1 - jitterRatio, 1 + jitterRatio]` and clamps the final value to `maxDelayMs`; provider delay is not jittered.
@@ -124,7 +124,7 @@ If recovery is exhausted, the final failure is stored once on `turn/end.reason`
- Pure unit tests cover transient-code selection, exponential backoff and jitter bounds, valid and over-cap `Retry-After`, exhausted budgets, deterministic timer/random seams, and abort during backoff.
- Real agent-loop tests cover failure before chunks, partial chunks then failure, thrown and in-band failures, retry to success in a new step, exhaustion to structured `turn/end.reason`, and composition with `dsh-compact-basic` context-overflow recovery.
- The partial-chunk integration test proves failed chunks remain attributed to the failed step, no assistant message or tool side effect is committed for that step, and the successful retry has distinct provenance.
-- The plugin-owned `llm/retry` event is non-surface, survives JSONL and SQLite round trips, is ignored by message derivation, and drives TUI retraction plus durable discarded-attempt markers in append-only ACP and stdio streams. Keyless snapshots cover scheduling, cancellation, success, and exhaustion.
+- The plugin-owned `llm/retry` event is non-surface, survives JSONL and SQLite round trips, is ignored by message derivation, and drives TUI retraction plus scheduled-retry rendering. Keyless snapshots cover scheduling, cancellation, success, and exhaustion; ACP automation snapshots confirm that a discarded attempt stays off the wire while the recovered reply is emitted.
- Idle-watchdog tests prove the stable signal is rearmed only while `next()` is outstanding, disarmed during consumer think time and in `finally`, and classified separately from a total-call deadline and an earlier caller abort; adapter tests prove the signal stops the underlying request rather than merely detaching it.
- Direct `ctx.llm.stream()` callers remain single-attempt and receive the same structured failure facts.
diff --git a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.i18n.yaml b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.i18n.yaml
index 2da1519019..c446db0d3f 100644
--- a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.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
-2026-06-30-event-domain-semantics.md: 7310840088d4ff77ddf83c5c16753b4d46256692
-2026-06-30-event-domain-semantics.zh.md: 61d99878ab5e04b74d55d02789c434e0919bd785
+2026-06-30-event-domain-semantics.md: 1f3452cce0235718c35d71577d7013f3e647648c
+2026-06-30-event-domain-semantics.zh.md: ec2da7786e80fb6a0df9ff338d77a50e7b3ef569
diff --git a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.md b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.md
index 7310840088..1f3452cce0 100644
--- a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.md
+++ b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.md
@@ -20,13 +20,13 @@ This vocabulary is the foundation for interception decisions, the durable `hook/
**Three domains, one job each, with a single boundary rule.**
-- **`session/*` — the durable, replayable FACT log.** Owns `SessionEventMap`; every entry is JSON-only (no live objects). One `session/event` emit per append, plus the `session/flush` parallel durability checkpoint. It is also the live transcript feed: a consumer that wants to render or react to what happened subscribes here, so live rendering and `session/load` replay share one path.
+- **`session/*` — the durable, replayable FACT log.** Owns `SessionEventMap`; every entry is JSON-only (no live objects). One `session/event` emit per append, plus the `session/flush` parallel durability checkpoint. It is also the live transcript feed: a consumer that wants to render or react to what happened subscribes here, so live rendering and replay projections share one path.
- **`agent/*` — the LIVE runtime surface.** Always carries the live `Agent`. Two shapes: INTERCEPTION waterfalls (`agent/request`, `agent/step-result`, `agent/turn-continuation`) that mutate or veto, and TRANSIENT emits (`agent/status`, `agent/error`, `agent/created`/`agent/disposed`, `agent/queued`) that notify with the `Agent` in hand. Turn and step BOUNDARIES are NOT here — they are durable session events read off `session/event`, and so are the token stream (`assistant/chunk`) and mid-turn steering (`steering/message`).
- **`tools/*` — the tool registry + execution seam.**
**The boundary rule:** a durable, replayable fact is a `SessionEvent`; a live interception or a transient/live-object signal is an `agent`/`tools` Cordis event. A turn or step boundary is a durable fact, so it lives in the session log and is read off the `session/event` feed — it is NOT mirrored as an `agent/*` emit.
-**Applying the rule to the boundary twins:** all four boundary mirrors — `agent/turn-start`, `agent/turn-end`, `agent/step-start`, `agent/step-end` — are **REMOVED**. No production consumer needs the live `Agent` at a boundary: the ACP bridge settles from `session/event` `turn/end` plus `agent/status`, and the only turn-mirror consumer (`dsh-ui-stdio`, a disposable test REPL) renders boundaries from `session/event` while retaining its live target object for the fixed `main` label. The step mirrors were removed first (they had no consumer at all); the turn mirrors followed once ui-stdio was migrated — see [the remove-boundary-mirror-events Agent Note](../simplification/2026-06-20-remove-agent-boundary-mirror-events.md), which owns that decision. Removing the emits also simplifies the loop's `closeStep`/`closeTurn` (one append each, no paired emit).
+**Applying the rule to the boundary twins:** all four boundary mirrors — `agent/turn-start`, `agent/turn-end`, `agent/step-start`, `agent/step-end` — are **REMOVED**. No production consumer needs the live `Agent` at a boundary: the ACP bridge correlates its in-flight prompt with the exact `session/event` `turn/start`/`turn/end` pair, and other transcript consumers likewise derive boundaries from the durable stream. See [the remove-boundary-mirror-events Agent Note](../simplification/2026-06-20-remove-agent-boundary-mirror-events.md), which owns that decision. Removing the emits also simplifies the loop's `closeStep`/`closeTurn` (one append each, no paired emit).
## Consequences
diff --git a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.zh.md b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.zh.md
index 61d99878ab..ec2da7786e 100644
--- a/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.zh.md
+++ b/.agents/notes/implemented/architecture/2026-06-30-event-domain-semantics.zh.md
@@ -20,13 +20,13 @@ harness 通过 Cordis 事件分类体系扩展 agent loop(智能体循环)
**三个域,各司其职,以一条边界规则统一。**
-- **`session/*`——持久的、可回放的事实日志。** 拥有 `SessionEventMap`;每条记录仅含 JSON(无活对象)。每次追加触发一次 `session/event` emit,加上 `session/flush` 并行持久性检查点。它同时也是实时 transcript(文本记录)源:想渲染或响应已发生事件的消费方在此订阅,因此实时渲染与 `session/load` 回放共享同一路径。
+- **`session/*`——持久的、可回放的事实日志。** 拥有 `SessionEventMap`;每条记录仅含 JSON(无活对象)。每次追加触发一次 `session/event` emit,加上 `session/flush` 并行持久性检查点。它同时也是实时 transcript(文本记录)源:想渲染或响应已发生事件的消费方在此订阅,因此实时渲染与回放投影共享同一路径。
- **`agent/*`——运行时实时表面。** 始终携带活的 `Agent`。两种形态:拦截 waterfall(瀑布式事件)(`agent/request`、`agent/step-result`、`agent/turn-continuation`)可变更或否决;瞬态 emit(`agent/status`、`agent/error`、`agent/created`/`agent/disposed`、`agent/queued`)在持有 `Agent` 的情况下通知。轮次和步骤边界不在此处——它们是持久的会话事件,从 `session/event` 读取;token 流(`assistant/chunk`)和中途 steering(中途引导)(`steering/message`)同理。
- **`tools/*`——工具注册表与执行 seam。**
**边界规则:** 持久的、可回放的事实是 `SessionEvent`;实时拦截或瞬态/活对象信号是 `agent`/`tools` Cordis 事件。轮次或步骤边界是持久事实,因此存在于会话日志中并从 `session/event` 源读取——不会被镜像为 `agent/*` emit。
-**将规则应用于边界镜像:** 全部四个边界镜像——`agent/turn-start`、`agent/turn-end`、`agent/step-start`、`agent/step-end`——被**移除**。没有生产消费方需要在边界处获取活的 `Agent`:ACP 桥接从 `session/event` 的 `turn/end` 加 `agent/status` 结算;唯一的轮次镜像消费方(`dsh-ui-stdio`,一个一次性测试 REPL)从 `session/event` 渲染边界,同时保留其实时目标对象用于固定的 `main` 标签。步骤镜像先被移除(它们完全没有消费方);轮次镜像在 ui-stdio 迁移后随之移除,见[移除边界镜像事件 Agent Note](../simplification/2026-06-20-remove-agent-boundary-mirror-events.md),该决策由它负责。移除 emit 也简化了循环的 `closeStep`/`closeTurn`(各只需一次 append,无需配对 emit)。
+**将规则应用于边界镜像:** 全部四个边界镜像——`agent/turn-start`、`agent/turn-end`、`agent/step-start`、`agent/step-end`——被**移除**。没有生产消费方需要在边界处获取活的 `Agent`:ACP 桥接将其进行中的提示词与精确对应的 `session/event` `turn/start`/`turn/end` 事件对关联,其他 transcript 消费方同样从持久流派生边界。见[移除边界镜像事件 Agent Note](../simplification/2026-06-20-remove-agent-boundary-mirror-events.md),该决策由它负责。移除 emit 也简化了循环的 `closeStep`/`closeTurn`(各只需一次 append,无需配对 emit)。
## 后果
diff --git a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.i18n.yaml
index 74c94e5b4f..e48cfbacb6 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.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
-2026-07-02-fs-per-session-cwd.md: aeb17e0a235720bee277f6533468dbdb6cae82dc
-2026-07-02-fs-per-session-cwd.zh.md: b9ad4b868c472c8feb9e6f65e42b1658ffdcd2e0
+2026-07-02-fs-per-session-cwd.md: d3f54e89e735016a373fa14c60123c681b3e7adf
+2026-07-02-fs-per-session-cwd.zh.md: ae732a3e4dacc3d4b800044aad60df3f3ce17cc0
diff --git a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.md b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.md
index aeb17e0a23..d3f54e89e7 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.md
@@ -6,9 +6,9 @@ English | [中文](2026-07-02-fs-per-session-cwd.zh.md)
## Problem
-The ACP bridge gives every session its own workspace: `session/new` records the editor's project directory as `SessionHeader.cwd`, and `dsh-tool-bash` defaults each bash call's `workdir` to the calling agent's `session.header.cwd` (see [the per-session cwd Agent Note work in `packages/ui/acp`](../../../../packages/ui/acp) and `resolveWorkdir` in `dsh-tool-bash`). So a bash command in session A runs in A's project, and in session B runs in B's — one server process, N workspaces.
+The ACP bridge gives every session its own workspace: `session/new` records the automation client's project directory as `SessionHeader.cwd`, and `dsh-tool-bash` defaults each bash call's `workdir` to the calling agent's `session.header.cwd` (see [the ACP package](../../../../packages/acp/acp) and `resolveWorkdir` in `dsh-tool-bash`). So a bash command in session A runs in A's project, and in session B runs in B's — one server process, N workspaces.
-Filesystem resolution used one plugin-load cwd while bash used the session project directory. Relative paths therefore disagreed whenever the editor project differed from the server launch directory; snapshots hid the bug by making those paths identical.
+Filesystem resolution used one plugin-load cwd while bash used the session project directory. Relative paths therefore disagreed whenever the automation client's project differed from the server launch directory; snapshots hid the bug by making those paths identical.
A valid absolute cwd can itself have two apparent parents: when it contains `symlink/..`, filesystem lookup follows the symlink before applying `..`, while `path.resolve()` erases both components lexically. Resolving sandbox policy lexically while launching bash from the raw cwd granted the unrelated lexical parent, denied writes in the real workspace, and let filesystem tools resolve relative paths into the wrong directory.
@@ -19,7 +19,7 @@ An ordinary symlink cwd exposes the same distinction when the requested relative
Thread the caller's session cwd into path resolution, exactly as `dsh-tool-bash` already does for `workdir`. When either the cwd or the requested path contains a parent segment, resolve the cwd to its native filesystem identity before any lexical join; ordinary cwd spellings stay stable for display when no traversal makes their identity observable. Reuse the resolved sandbox-policy root for mutations and sandboxed bash calls so one call has one workspace identity. The **caller** (the tool) supplies the cwd; the provider does not read a session or agent.
- `FileSystem.resolve` accepts `resolve(path: string, opts?: { cwd?: string; signal?: AbortSignal }): Promise`. `opts.cwd` is the base a RELATIVE `path` resolves against; an absolute `path` ignores it; omitting `opts.cwd` uses the backend's own default. `opts.signal` cancels resolution when the backend performs I/O. The options object keeps both caller-owned resolution controls together without positional growth.
-- `dsh-fs-local.resolve` uses `resolveLocalTarget(opts?.cwd ?? this.config.cwd, path)`. `config.cwd` stays the default for a caller that supplies none (non-ACP / no-session use, and the single-session stdio demo where `process.cwd()` IS the workspace).
+- `dsh-fs-local.resolve` uses `resolveLocalTarget(opts?.cwd ?? this.config.cwd, path)`. `config.cwd` stays the default for a caller that supplies no session cwd.
- `dsh-tool-fs`'s `read`/`write`/`edit` derive the session cwd through a shared `sessionCwd(exec, requestedPath)` helper (`exec.agent?.session.header.cwd`, mirroring bash's `resolveWorkdir`) and pass it to `resolve`. The helper uses native realpath semantics when a parent segment in either value could cross a symlink while retaining ordinary spellings otherwise; a sandboxed mutation reuses the complete policy's `workspaceRoot`; a non-agent / headerless caller yields `undefined`, so the backend applies its default.
## Alternatives considered
@@ -32,7 +32,7 @@ The default lives in ONE place — the provider's `config.cwd`. `sessionCwd` ret
## Consequences
-- In the ACP demo the fs tools and bash now agree on each session's workspace; an editor can open any project folder and both tool families act on it.
+- In the ACP demo the fs tools and bash agree on each session's workspace; an automation client can select any absolute project directory and both tool families act on it.
- A session cwd containing `symlink/..`, or an ordinary symlink cwd paired with a parent-traversing relative path, resolves from the same physical workspace for bash, filesystem tools, and the sandbox grant; the lexical parent receives no grant.
- No change to `FsTarget` identity: `targetKey` is still the realpath of the resolved absolute path, so observed-state keying and symlink identity are unaffected — a correct per-session cwd produces the same key bash targets.
- Backward compatible: every existing `resolve(path)` call (all in tests) keeps working; the new argument is optional.
diff --git a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.zh.md b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.zh.md
index b9ad4b868c..ae732a3e4d 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-fs-per-session-cwd.zh.md
@@ -6,9 +6,9 @@ Status: implemented
## 问题
-ACP(Agent Client Protocol)桥接层为每个会话提供独立的工作区:`session/new` 将编辑器的项目目录记录为 `SessionHeader.cwd`,`dsh-tool-bash` 将每次 bash 调用的 `workdir` 默认设为调用方 agent(智能体)的 `session.header.cwd`(见 [`packages/ui/acp`](../../../../packages/ui/acp) 中的每会话 cwd Agent Note 工作与 `dsh-tool-bash` 中的 `resolveWorkdir`)。因此会话 A 中的 bash 命令在 A 的项目目录执行,会话 B 中的在 B 的项目目录执行——一个服务器进程,N 个工作区。
+ACP(Agent Client Protocol)桥接层为每个会话提供独立的工作区:`session/new` 将自动化客户端的项目目录记录为 `SessionHeader.cwd`,`dsh-tool-bash` 将每次 bash 调用的 `workdir` 默认设为调用方 agent(智能体)的 `session.header.cwd`(见 [ACP 包](../../../../packages/acp/acp)与 `dsh-tool-bash` 中的 `resolveWorkdir`)。因此会话 A 中的 bash 命令在 A 的项目目录执行,会话 B 中的在 B 的项目目录执行——一个服务器进程,N 个工作区。
-文件系统解析使用的是插件加载时的 cwd,而 bash 使用的是会话的项目目录。因此,当编辑器项目目录与服务器启动目录不同时,相对路径的解析结果就会不一致;快照测试因为让这两个路径相同而掩盖了这个 bug。
+文件系统解析使用的是插件加载时的 cwd,而 bash 使用的是会话的项目目录。因此,当自动化客户端的项目目录与服务器启动目录不同时,相对路径的解析结果就会不一致;快照测试因为让这两个路径相同而掩盖了这个 bug。
一个有效的绝对 cwd 本身可能看起来有两个父目录:当它包含 `symlink/..` 时,文件系统查找会先跟随符号链接再应用 `..`,而 `path.resolve()` 会从词法上抹掉这两个组件。如果用词法解析沙箱策略却从原始 cwd 启动 bash,就会把权限授予无关的词法父目录、拒绝真实工作区内的写入,并让文件系统工具把相对路径解析进错误目录。
@@ -19,7 +19,7 @@ ACP(Agent Client Protocol)桥接层为每个会话提供独立的工作区
将调用方的会话 cwd 传入路径解析,与 `dsh-tool-bash` 对 `workdir` 的处理方式完全一致。当 cwd 或请求路径任一包含父目录段时,在任何词法 join 之前把 cwd 解析为原生文件系统标识;没有遍历会使标识可观察时,则保留普通 cwd 拼写以供展示。mutation 和沙箱化 bash 调用复用解析后的沙箱策略根目录,使一次调用只有一个工作区标识。**调用方**(即工具)提供 cwd;提供方不读取会话或 agent。
- `FileSystem.resolve` 接受 `resolve(path: string, opts?: { cwd?: string; signal?: AbortSignal }): Promise`。`opts.cwd` 是相对 `path` 解析时的基准目录;绝对 `path` 忽略它;省略 `opts.cwd` 则使用后端自身的默认值。后端执行 I/O 时,`opts.signal` 可以取消解析。options 对象把调用方拥有的两个解析控制项放在一起,避免位置参数继续增长。
-- `dsh-fs-local.resolve` 使用 `resolveLocalTarget(opts?.cwd ?? this.config.cwd, path)`。`config.cwd` 仍作为调用方未提供 cwd 时的默认值(非 ACP/无会话场景,以及 `process.cwd()` 本身就是工作区的单会话 stdio 演示)。
+- `dsh-fs-local.resolve` 使用 `resolveLocalTarget(opts?.cwd ?? this.config.cwd, path)`。`config.cwd` 仍作为调用方未提供会话 cwd 时的默认值。
- `dsh-tool-fs` 的 `read`/`write`/`edit` 通过共享的 `sessionCwd(exec, requestedPath)` 辅助函数(`exec.agent?.session.header.cwd`,与 bash 的 `resolveWorkdir` 对应)获取会话 cwd,并传给 `resolve`。只要任一值中的父目录段可能跨越符号链接,该辅助函数就使用原生 realpath 语义,否则保留普通拼写;沙箱化 mutation 复用完整策略的 `workspaceRoot`;非 agent/无 header 的调用方得到 `undefined`,后端因此应用其默认值。
## 曾考虑的替代方案
@@ -32,7 +32,7 @@ ACP(Agent Client Protocol)桥接层为每个会话提供独立的工作区
## 后果
-- 在 ACP 演示中,fs 工具与 bash 现在对每个会话的工作区达成一致;编辑器可以打开任意项目目录,两类工具都在该目录下操作。
+- 在 ACP 演示中,fs 工具与 bash 对每个会话的工作区达成一致;自动化客户端可以选择任意绝对项目目录,两类工具都在该目录下操作。
- 对于包含 `symlink/..` 的会话 cwd,或普通符号链接 cwd 搭配含父目录遍历的相对路径,bash、文件系统工具和沙箱授权都会从同一个物理工作区解析;词法父目录不会获得授权。
- `FsTarget` 的标识不变:`targetKey` 仍为解析后绝对路径的 realpath,因此 observed-state 键控与符号链接标识不受影响——正确的每会话 cwd 产生与 bash 目标相同的 key。
- 向后兼容:所有现有的 `resolve(path)` 调用(均在测试中)继续正常工作;新参数是可选的。
diff --git a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.i18n.yaml
index 632ba388a2..e6d8d8570a 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.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
-2026-07-02-result-time-applied-hunk-diffs.md: fe163484ad56858fabd55dacace1d56814bd93e4
-2026-07-02-result-time-applied-hunk-diffs.zh.md: e8283a40d96c031a84e2627117801b3043eb8904
+2026-07-02-result-time-applied-hunk-diffs.md: 55e1612aacd9070ede2f0079c73c30975e1bd5cf
+2026-07-02-result-time-applied-hunk-diffs.zh.md: 6fe0032a507af2915bdf79a43578082785411479
diff --git a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md
index fe163484ad..55e1612aac 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.md
@@ -6,7 +6,7 @@ English | [中文](2026-07-02-result-time-applied-hunk-diffs.zh.md)
## Problem
-The [tagged render-intent union](2026-07-02-tool-render-intent-union.md) gave `dsh-tool-fs` write/edit a `card:'diff'` at CALL time, derived purely from the tool's args: write ⇒ `{oldText:null, newText:content}` (the whole new file), edit ⇒ `{oldText:old_string, newText:new_string}` (the bare replaced snippet). An editor renders that as an inline diff, but it is a **context-free** diff — the bare `old_string`→`new_string` with no surrounding lines, and a `replace_all` that touched five scattered sites still renders as one snippet pair.
+The [tagged render-intent union](2026-07-02-tool-render-intent-union.md) gives `dsh-tool-fs` write/edit a `card:'diff'` at call time, derived purely from the tool's args: write ⇒ `{oldText:null, newText:content}` (the whole new file), edit ⇒ `{oldText:old_string, newText:new_string}` (the bare replaced snippet). A UI can render that as an inline diff, but it is a **context-free** diff — the bare `old_string`→`new_string` with no surrounding lines, and a `replace_all` that touched five scattered sites still renders as one snippet pair.
Driving `claude-agent-acp`'s own ACP bridge shows what a full editor diff looks like: after the mutation applies, it emits a SECOND `tool_call_update` whose diff is the **applied hunk with ±3 context lines** (and one hunk per changed site for `replace_all`), reconstructed from the tool's `structuredPatch`. That result-time hunk is what makes Zed show the change *in place* in the file rather than as a floating snippet. Our tools stopped at the call-time snippet; the completed result carried only the plain "updated successfully" text, no diff.
@@ -29,11 +29,11 @@ This remains the general shape ("a tool projects durable result presentation"),
Per the [capability-seam split](2026-06-13-capability-seams.md), the storage backend returns only **storage facts** and the model-facing tool owns **presentation**:
- `dsh-fs` widens `FsEditOutcome` with `{ before: string; after: string }` and `FsWriteOutcome` with `{ before: string | null; after: string }` (`before: null` ⇒ a create, or an existing-but-undiffable binary/non-UTF-8 file). The local backend already holds both texts at write time; it returns them as raw LF-normalized text, with **no diff/UI concept** entering the seam.
-- `dsh-tool-fs` returns canonical before/after mutation facts and projects contextual hunks as `meta: { diffs: FileDiff[] }`. Successful mutations always complete with a diff card because ACP result content replaces the pending card: creates or unchanged overwrites fall back to an args-derived whole-file diff, while edits use applied hunks. Failed mutations carry no diff metadata and render their error normally.
+- `dsh-tool-fs` returns canonical before/after mutation facts and projects contextual hunks as `meta: { diffs: FileDiff[] }`. Successful mutations complete with a diff view: creates or unchanged overwrites fall back to an args-derived whole-file diff, while edits use applied hunks. Failed mutations carry no diff metadata and render their error normally.
-### 3. The bridge renders a `diff` result card
+### 3. UI transports render a `diff` result view
-`ToolResultView` gains a `DiffResultView { card:'diff'; title?; diffs: FileDiff[] }`; the bridge's result-side `switch (view.card)` gets a `diff` arm emitting the `{type:'diff'}` `ToolCallContent` blocks (mirroring the call-side arm). An ACP `tool_call_update.content` REPLACES the call's content in an editor, so the result diff **supersedes** the call-time snippet (and keeps the model-facing result text from clobbering it) — the two-update sequence (call snippet, then result diff) matches `claude-agent-acp` exactly.
+`ToolResultView` includes `DiffResultView { card:'diff'; title?; diffs: FileDiff[] }`. TUI and JSON-RPC/Web consumers switch on the same tagged view and replace the pending call's context-free snippet with the applied result hunk. The [automation-only ACP bridge](../simplification/2026-07-23-acp-automation-only-protocol.md) does not carry tool presentation.
## Alternatives considered
diff --git a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.zh.md b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.zh.md
index e8283a40d9..6fe0032a50 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-result-time-applied-hunk-diffs.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-[带标签的 render-intent 联合类型](2026-07-02-tool-render-intent-union.md)为 `dsh-tool-fs` 的 write/edit 在调用时刻提供了 `card:'diff'`,纯粹从工具参数推导:write ⇒ `{oldText:null, newText:content}`(整个新文件),edit ⇒ `{oldText:old_string, newText:new_string}`(裸替换片段)。编辑器将其渲染为行内 diff,但这是一个**无上下文**的 diff:裸的 `old_string`→`new_string` 没有周围行,而一次触及五个分散位置的 `replace_all` 仍然渲染为一对片段。
+[带标签的 render-intent 联合类型](2026-07-02-tool-render-intent-union.md)为 `dsh-tool-fs` 的 write/edit 在调用时刻提供 `card:'diff'`,纯粹从工具参数推导:write ⇒ `{oldText:null, newText:content}`(整个新文件),edit ⇒ `{oldText:old_string, newText:new_string}`(裸替换片段)。UI 可以将其渲染为行内 diff,但这是一个**无上下文**的 diff:裸的 `old_string`→`new_string` 没有周围行,而一次触及五个分散位置的 `replace_all` 仍然渲染为一对片段。
在对接 `claude-agent-acp` 自身的 ACP(Agent Client Protocol) bridge 时可以看到完整编辑器 diff 的样子:变更应用后,它发出第二个 `tool_call_update`,其 diff 是**带 ±3 行上下文的 applied hunk**(`replace_all` 的每个变更位置各一个 hunk),由工具的 `structuredPatch` 重建。这个结果时刻的 hunk 正是让 Zed 在文件中*原位*显示变更(而非浮动片段)的关键。我们的工具止步于调用时刻的片段;完成后的结果只携带纯文本「updated successfully」,没有 diff。
@@ -29,11 +29,11 @@ Status: implemented
按照 [capability-seam 拆分](2026-06-13-capability-seams.md),存储后端只返回**存储事实**,面向模型的工具拥有**展示**:
- `dsh-fs` 将 `FsEditOutcome` 扩展为包含 `{ before: string; after: string }`,将 `FsWriteOutcome` 扩展为包含 `{ before: string | null; after: string }`(`before: null` 表示创建,或已存在但不可 diff 的二进制/非 UTF-8 文件)。本地后端在写入时已持有两份文本;它以原始 LF 规范化文本返回,**不让任何 diff/UI 概念进入 seam**。
-- `dsh-tool-fs` 返回规范的变更前/后事实,并将上下文 hunk 投影为 `meta: { diffs: FileDiff[] }`。成功的变更始终以 diff 卡片完成,因为 ACP 结果内容会替换待定卡片:创建或无变化的覆写回退到由参数推导的整文件 diff,而编辑使用 applied hunk。失败的变更不携带 diff 元数据,正常渲染其错误信息。
+- `dsh-tool-fs` 返回规范的变更前/后事实,并将上下文 hunk 投影为 `meta: { diffs: FileDiff[] }`。成功的变更以 diff 视图完成:创建或无变化的覆写回退到由参数推导的整文件 diff,而编辑使用 applied hunk。失败的变更不携带 diff 元数据,正常渲染其错误信息。
-### 3. Bridge 渲染 `diff` 结果卡片
+### 3. UI 传输层渲染 `diff` 结果视图
-`ToolResultView` 新增 `DiffResultView { card:'diff'; title?; diffs: FileDiff[] }`;bridge 结果侧的 `switch (view.card)` 增加 `diff` 分支,发出 `{type:'diff'}` 的 `ToolCallContent` 块(与调用侧分支对称)。ACP 的 `tool_call_update.content` 在编辑器中替换调用时的内容,因此结果 diff **取代**调用时刻的片段(并防止面向模型的结果文本覆盖它)——两次更新序列(先调用片段,再结果 diff)与 `claude-agent-acp` 完全一致。
+`ToolResultView` 包含 `DiffResultView { card:'diff'; title?; diffs: FileDiff[] }`。TUI 与 JSON-RPC/Web 消费方在同一个带标签的视图上做 switch,用 applied 结果 hunk 替换待定调用的无上下文片段。[仅面向自动化的 ACP 桥接层](../simplification/2026-07-23-acp-automation-only-protocol.md)不承载工具展示。
## 曾考虑的替代方案
diff --git a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.i18n.yaml
index c4e769f684..d2ac37af62 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.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
-2026-07-02-tool-render-intent-union.md: c7adf8f405000ec7940f82e1e1e2406d86253461
-2026-07-02-tool-render-intent-union.zh.md: 01dff0daaacc45f5c85eb2c4380dc78800ce9a81
+2026-07-02-tool-render-intent-union.md: 84423e9000526848a111591c1bb2ab92067bbe50
+2026-07-02-tool-render-intent-union.zh.md: 43873c622fc8483b4a7033d17b4b4fab1342a56b
diff --git a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.md b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.md
index c7adf8f405..84423e9000 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.md
@@ -4,6 +4,8 @@ Status: implemented
English | [中文](2026-07-02-tool-render-intent-union.zh.md)
+> The render-intent union remains current for UI transports; its ACP mapping is superseded by [ACP as an automation-only protocol](../simplification/2026-07-23-acp-automation-only-protocol.md).
+
## Problem
A tool declares how its calls render in a UI (an editor's tool-call card) through two callbacks, `presentCall`/`presentResult` on `ToolDefinition`, returning `ToolCallPresentation` / `ToolResultPresentation` with an optional `ToolTerminal` sub-shape. These grew incrementally into a **bag of optional fields**: `title`, `kind`, `rawInput`, `content`, `locations`, `terminal` on the call; `title`, `content`, `terminal` on the result; `cwd`/`output`/`exitCode`/`signal` on `ToolTerminal`. The split of responsibility is muddy:
@@ -12,7 +14,7 @@ A tool declares how its calls render in a UI (an editor's tool-call card) throug
- Which combinations are *valid* is unwritten: a `terminal` call that also sets `content` means "description above the card"; a generic call that sets `terminal` is meaningless but representable. The type permits nonsense.
- There is no way to express the one file-tool affordance an editor most wants — a **diff card** (`{path, oldText, newText}`, which Zed renders as an inline diff / new-file preview). `ToolCallPresentation.content` is the *LLM* `ContentBlock[]` vocabulary (text/image), so a tool literally cannot ask for a diff.
-The existing `FIXME(tool-presentation)` in `packages/core/tools/src/index.ts` named the fix: "redesign the type so a tool declares its render INTENT once (e.g. a tagged union over card kinds) rather than a bag of optional fields the bridge stitches together." The rejected Agent Note [Collapse tool-owned UI presentation](../../rejected/simplification/2026-06-20-generic-tool-rendering.md) deferred it explicitly: rich rendering "should return later as a tagged render-intent union after there are at least two real tools and two real consumers to validate the vocabulary." That bar is now met — two producer families (`dsh-tool-bash`, `dsh-tool-fs`) and two consumers (the ACP bridge live path + the snapshot replay path).
+The existing `FIXME(tool-presentation)` in `packages/core/tools/src/index.ts` named the fix: "redesign the type so a tool declares its render INTENT once (e.g. a tagged union over card kinds) rather than a bag of optional fields the bridge stitches together." The rejected Agent Note [Collapse tool-owned UI presentation](../../rejected/simplification/2026-06-20-generic-tool-rendering.md) deferred it explicitly: rich rendering "should return later as a tagged render-intent union after there are at least two real tools and two real consumers to validate the vocabulary." That bar is met by multiple producer families plus the TUI and host/client-runtime (Web) consumers.
## Decision
@@ -39,8 +41,8 @@ interface TerminalResultView { card: 'terminal'; title?: string; output?: string
### Why a tagged union beats the field-bag
- **Invalid states become unrepresentable.** A generic card cannot carry terminal output; a terminal card cannot carry a diff. The old bag permitted all of these.
-- **The bridge switches instead of stitching.** One arm per card kind, each producing exactly the wire shape that card needs, rather than reconciling five optional fields whose interactions are undocumented.
-- **`diff` is a first-class intent.** `dsh-tool-fs` write/edit declare `card:'diff'`; the bridge emits an ACP `{type:'diff', path, oldText, newText}` `ToolCallContent` (already in the SDK's `ToolCallContent` union, previously unused by the bridge). This is the affordance the redesign unlocks.
+- **Consumers switch instead of stitching.** One arm per card kind produces exactly the view that card needs, rather than reconciling five optional fields whose interactions are undocumented.
+- **`diff` is a first-class intent.** `dsh-tool-fs` write/edit declare `card:'diff'` with `{path, oldText, newText}`, allowing capable UIs to render an inline change without tool-name special cases.
### Producer mapping
@@ -52,17 +54,16 @@ interface TerminalResultView { card: 'terminal'; title?: string; output?: string
`TerminalResultView` carries only `output`/`exitCode`/`signal`. A UI without the terminal capability needs a fenced ` ```console ` text fallback; that derivation moves to the **bridge** (it wraps `output` in a fenced block on the no-capability path), rather than the tool double-encoding it. This keeps the bash tool's result a single structured shape and preserves the existing capability-gated behavior byte-for-byte.
+The terminal intent is display-only. The harness still executes the command through its bash service, preserving sandboxing, environment scrubbing, task ownership, and per-session cwd; a UI projects the completed call and never becomes a second execution backend.
+
### Purity preserved
`presentCall`/`presentResult` remain pure functions of `args` (+ the result for `presentResult`) — they run on live streaming AND session-log replay, so they must be replay-deterministic. Every view is derived from args alone: write's diff is new-file style (`oldText:null`) because the tool has no old content at call time; edit's diff is `old_string`→`new_string`.
-## Relative-path display titles
-
-`claude-agent-acp` relativizes a file card's title path against the session cwd (`toDisplayPath`) — `Read src/foo.ts`, not `/abs/proj/src/foo.ts` — while keeping `locations[]`/`diff.path` **raw** (the editor opens the real path). Our `presentCall` is pure/args-only and cannot see the session cwd, so this relativization happens at the **bridge**, which already threads the session cwd into tool-call rendering (the same cwd it uses to resolve a terminal card's header). The bridge relativizes the title only, by an exact structured replace of the known `locations[0].path`/`diffs[0].path` substring — generic over the file-card kinds, never special-casing tool names.
-
## Alternatives considered
- **Delete tool-owned presentation entirely** — [the rejected collapse proposal](../../rejected/simplification/2026-06-20-generic-tool-rendering.md); its own verdict deferred to exactly this union once two real tools and two real consumers existed, and that bar is now met.
+- **Let a UI execute terminal intents** — rejected because it would bypass the harness's bash policy and ownership contracts and fork command execution across backends. A terminal card describes harness-owned execution; it never authorizes client-side execution.
- **A merge-extensible union** (the `ContentBlockMap` pattern) — rejected: a new render intent needs new bridge code to render it anyway, so a plugin-added variant the bridge silently drops would be worse than the compile error the closed union raises at the bridge's `assertNever` switch.
- **Keeping the optional-field bag** — the status quo the Problem dissects: invalid states representable, undocumented field interactions, and no way to ask for a diff card at all.
@@ -78,5 +79,4 @@ A new render intent is a compile-breaking change at the bridge switch — delibe
- Supersedes the deferral in [Collapse tool-owned UI presentation](../../rejected/simplification/2026-06-20-generic-tool-rendering.md) (rejected — "wait for two real tools and two real consumers, then a tagged render-intent union"). That bar is now met; this is that union.
- Extended by [Result-time applied-hunk diffs](2026-07-02-result-time-applied-hunk-diffs.md), which adds a persisted `meta` channel so write/edit emit a result-time `DiffResultView` — the applied change (a contextual hunk with context lines / one per `replace_all` site, or a whole-file diff for a create) — on top of this union's call-time diff card.
-- Folds `ToolTerminal` into the `terminal` views described by [ACP terminal and tool-call rendering](../feature/2026-06-18-acp-terminal-and-tool-rendering.md) (the `_meta` terminal-card convention and capability gate are unchanged; only the harness-side presentation type changes).
-- The ACP SDK's `Diff` / `ToolCallContent` types back the new `diff` card.
+- Folds `ToolTerminal` into the tagged `terminal` views used by current UI transports.
diff --git a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.zh.md b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.zh.md
index 01dff0daaa..43873c622f 100644
--- a/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-02-tool-render-intent-union.zh.md
@@ -4,6 +4,8 @@ Status: implemented
[English](2026-07-02-tool-render-intent-union.md) | 中文
+> render-intent 联合类型对 UI 传输层仍然有效;其 ACP 映射已被 [ACP 作为仅面向自动化的协议](../simplification/2026-07-23-acp-automation-only-protocol.md)取代。
+
## 问题
工具通过 `ToolDefinition` 上的两个回调 `presentCall`/`presentResult` 声明其调用在 UI(编辑器的工具调用卡片)中如何渲染,返回 `ToolCallPresentation` / `ToolResultPresentation`,并带有一个可选的 `ToolTerminal` 子结构。这些类型在增量演进中变成了一个**可选字段的集合**:调用侧有 `title`、`kind`、`rawInput`、`content`、`locations`、`terminal`;结果侧有 `title`、`content`、`terminal`;`ToolTerminal` 上有 `cwd`/`output`/`exitCode`/`signal`。职责划分模糊不清:
@@ -12,7 +14,7 @@ Status: implemented
- 哪些组合是*合法的*没有文档说明:一个设置了 `content` 的 `terminal` 调用意味着「卡片上方的描述」;一个设置了 `terminal` 的 generic 调用毫无意义但类型上可表达。类型允许无意义的状态存在。
- 无法表达编辑器最需要的文件工具能力:**diff 卡片**(`{path, oldText, newText}`,Zed 将其渲染为内联 diff / 新文件预览)。`ToolCallPresentation.content` 使用的是 *LLM(大语言模型)* 的 `ContentBlock[]` 词汇(text/image),工具根本无法请求 diff 展示。
-`packages/core/tools/src/index.ts` 中已有的 `FIXME(tool-presentation)` 指出了修复方向:「重新设计类型,让工具一次性声明其渲染意图(例如按卡片种类的带标签联合类型),而非一堆由 bridge 拼接的可选字段。」被否决的 Agent Note [折叠工具拥有的 UI 呈现](../../rejected/simplification/2026-06-20-generic-tool-rendering.md)明确推迟了此事:富渲染「应当在至少有两个真实工具和两个真实消费方验证词汇之后,以带标签 render-intent 联合类型的形式回归。」该条件现已满足:两个生产者族(`dsh-tool-bash`、`dsh-tool-fs`)和两个消费方(ACP bridge 实时路径 + 快照回放路径)。
+`packages/core/tools/src/index.ts` 中已有的 `FIXME(tool-presentation)` 指出了修复方向:「重新设计类型,让工具一次性声明其渲染意图(例如按卡片种类的带标签联合类型),而非一堆由 bridge 拼接的可选字段。」被否决的 Agent Note [折叠工具拥有的 UI 呈现](../../rejected/simplification/2026-06-20-generic-tool-rendering.md)明确推迟了此事:富渲染「应当在至少有两个真实工具和两个真实消费方验证词汇之后,以带标签 render-intent 联合类型的形式回归。」该条件已由多个生产者族,加上 TUI 与宿主/客户端运行时(Web)这些消费方满足。
## 决策
@@ -39,8 +41,8 @@ interface TerminalResultView { card: 'terminal'; title?: string; output?: string
### 为什么带标签联合类型优于字段集合
- **无效状态变得不可表达。** generic 卡片不能携带终端输出;terminal 卡片不能携带 diff。旧的字段集合允许所有这些组合。
-- **bridge 分发而非拼接。** 每种卡片一个分支,各自精确产出该卡片所需的协议格式(wire format),而非调和五个交互关系未文档化的可选字段。
-- **`diff` 成为一等意图。** `dsh-tool-fs` 的 write/edit 声明 `card:'diff'`;bridge 输出 ACP `{type:'diff', path, oldText, newText}` 的 `ToolCallContent`(已存在于 SDK 的 `ToolCallContent` 联合类型中,此前 bridge 未使用)。这正是本次重设计解锁的能力。
+- **消费方分发而非拼接。** 每种卡片一个分支,精确产出该卡片所需的视图,而非调和五个交互关系未文档化的可选字段。
+- **`diff` 成为一等意图。** `dsh-tool-fs` 的 write/edit 声明带 `{path, oldText, newText}` 的 `card:'diff'`,让有能力的 UI 无需针对工具名做特殊处理即可渲染行内变更。
### 生产者映射
@@ -52,17 +54,16 @@ interface TerminalResultView { card: 'terminal'; title?: string; output?: string
`TerminalResultView` 只携带 `output`/`exitCode`/`signal`。不具备终端能力的 UI 需要一个围栏 ` ```console ` 文本回退;该推导移至 **bridge**(在无能力路径上将 `output` 包裹在围栏代码块中),而非由工具双重编码。这使 bash 工具的结果保持单一结构化形状,并逐字节保留既有的能力门控行为。
+terminal 意图只用于展示。harness 仍通过自身的 bash 服务执行命令,从而保留沙箱、环境清理、任务归属和每会话 cwd;UI 只呈现已完成的调用,绝不会成为第二个执行后端。
+
### 纯函数性保持不变
`presentCall`/`presentResult` 仍然是 `args`(`presentResult` 还有 result)的纯函数——它们在实时流式输出和会话日志回放中都会运行,因此必须具备回放确定性。每个 view 仅从 args 推导:write 的 diff 是新文件风格(`oldText:null`),因为工具在调用时没有旧内容;edit 的 diff 是 `old_string`→`new_string`。
-## 相对路径显示标题
-
-`claude-agent-acp` 将文件卡片标题中的路径相对于会话 cwd 做缩短处理(`toDisplayPath`)——显示 `Read src/foo.ts` 而非 `/abs/proj/src/foo.ts`——同时保持 `locations[]`/`diff.path` 为**原始路径**(编辑器打开真实路径)。我们的 `presentCall` 是纯函数/仅依赖 args,无法访问会话 cwd,因此这一相对化处理发生在 **bridge**,bridge 已经将会话 cwd 传入工具调用渲染逻辑(与它用于解析 terminal 卡片标题的 cwd 相同)。bridge 仅对标题做相对化,方式是对已知的 `locations[0].path`/`diffs[0].path` 子串做精确的结构化替换——对所有文件卡片类型通用,从不针对工具名做特殊处理。
-
## 曾考虑的替代方案
- **完全删除工具自有的展示**:即[被否决的 collapse 提案](../../rejected/simplification/2026-06-20-generic-tool-rendering.md);其自身的结论正是推迟到两个真实工具和两个真实消费方存在后再做此联合类型,该条件现已满足。
+- **让 UI 执行 terminal 意图**:否决。这样会绕过 harness 的 bash 策略与归属契约,并把命令执行分裂到不同后端。terminal 卡片描述的是 harness 拥有的执行,绝不授权客户端侧执行。
- **可合并扩展的联合类型**(`ContentBlockMap` 模式):否决。新的渲染意图无论如何需要新的 bridge 代码来渲染,因此一个被 bridge 静默丢弃的插件添加变体,比封闭联合类型在 bridge 的 `assertNever` switch 处引发的编译错误更糟糕。
- **保留可选字段集合**:即「问题」一节所剖析的现状:无效状态可表达、字段交互无文档、且完全无法请求 diff 卡片。
@@ -78,5 +79,4 @@ interface TerminalResultView { card: 'terminal'; title?: string; output?: string
- 取代[折叠工具拥有的 UI 呈现](../../rejected/simplification/2026-06-20-generic-tool-rendering.md)(已否决——「等两个真实工具和两个真实消费方,然后做带标签 render-intent 联合类型」)中的推迟决定。该条件现已满足;本 Agent Note 即为那个联合类型。
- 被[结果时已应用 hunk 差异](2026-07-02-result-time-applied-hunk-diffs.md)扩展:后者添加了一个持久化的 `meta` 通道,使 write/edit 在结果时输出 `DiffResultView`(应用后的变更:带上下文行的 contextual hunk / 每个 `replace_all` 位点一个,或创建时的整文件 diff),叠加在本联合类型的调用时 diff 卡片之上。
-- 将 `ToolTerminal` 折入 [ACP terminal 与工具调用渲染](../feature/2026-06-18-acp-terminal-and-tool-rendering.md)所描述的 `terminal` view(`_meta` terminal 卡片约定和能力门控不变;仅 harness 侧的展示类型改变)。
-- ACP SDK 的 `Diff` / `ToolCallContent` 类型支撑新的 `diff` 卡片。
+- 将 `ToolTerminal` 折入当前 UI 传输层使用的带标签 `terminal` 视图。
diff --git a/.agents/notes/implemented/architecture/2026-07-05-windows-fs-permissions.md b/.agents/notes/implemented/architecture/2026-07-05-windows-fs-permissions.md
deleted file mode 100644
index 932b6ddbf4..0000000000
--- a/.agents/notes/implemented/architecture/2026-07-05-windows-fs-permissions.md
+++ /dev/null
@@ -1,31 +0,0 @@
-# Agent Note: Windows write-permission semantics — inherited DACLs, not mode bits
-
-Status: implemented
-
-The replacement-file decision in this record is superseded by [Windows DACL preservation](../bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md).
-
-## Problem
-
-`writeFileAtomic` in `@deepseek-ai/dsh-fs-local` protects write-in-progress content with POSIX mode bits: the staging directory is created `0o700`, the temp file is opened `0o600`, and new files default to `0o600`. On POSIX this keeps temporary content owner-only regardless of the parent directory's permissions.
-
-Windows has no working equivalent behind the same API. Node's `chmod` there drives only the read-only attribute (every mode this package passes carries owner-write, so the calls are benign no-ops), and `stat().mode` reports synthetic `0o666`/`0o444` bits. The real security state is the file's DACL: a newly created file or directory inherits from its parent, while replacement needs the explicit handling owned by the superseding Agent Note.
-
-## Decision
-
-New Windows files use directory inheritance rather than synthetic mode bits: the staging directory is created inside the target's parent directory (`dirname(absolutePath)`), so it and the temp file inherit the destination directory's DACL. Replacement files follow the stricter [DACL preservation contract](../bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md).
-
-Tests assert mode bits on POSIX only. Native Windows coverage pins the package-owned replacement behavior; new-file inheritance remains an operating-system contract rather than a machine-specific ACL allowlist.
-
-## Alternatives considered
-
-**Explicit owner-only DACLs for new files.** Rejected because they would break inheritance and surprise users whose project directories are deliberately shared. Replacement writes copy the target's existing DACL rather than inventing an owner-only policy.
-
-**Test-side ACL verification.** A `Get-Acl` SID allowlist or `icacls` would verify Windows inheritance and the machine's `%TEMP%` ACL rather than package behavior; `icacls` also localizes well-known account names, making parsing locale-fragile.
-
-**Skip `chmod` on Windows.** Platform-guarding benign no-op calls adds branches without changing behavior.
-
-## Consequences
-
-POSIX keeps owner-only temp content regardless of the parent directory. A new Windows target inside a broadly accessible directory inherits that accessibility by design; a replacement retains the target's narrower DACL when one exists.
-
-Mode preservation across a replace degenerates to a no-op on Windows: a writable file probes as `0o666`, and replaying that through `chmod` leaves the read-only attribute clear. A read-only target cannot be replaced there because publication fails before the synthetic mode would matter.
diff --git a/.agents/notes/implemented/architecture/2026-07-05-windows-jsonl-durable-publish.md b/.agents/notes/implemented/architecture/2026-07-05-windows-jsonl-durable-publish.md
index 60c3b9b627..23e5f630d3 100644
--- a/.agents/notes/implemented/architecture/2026-07-05-windows-jsonl-durable-publish.md
+++ b/.agents/notes/implemented/architecture/2026-07-05-windows-jsonl-durable-publish.md
@@ -12,9 +12,9 @@ Windows has atomic namespace operations, but Node does not expose a POSIX-equiva
The JSONL backend forks inside `materialize()` before any namespace mutation. Shared code computes the session directory, final log path, and encoded header plus initial event batch; POSIX and Windows then run separate publication protocols.
-POSIX keeps the existing protocol: create the root and cwd bucket with parent directory fsyncs, write and fsync a temp file, publish with `link()` so an existing final log is never overwritten, fsync the bucket directory, then remove the redundant temp hard link.
+POSIX keeps the existing protocol: create the root, project directory, and session directory with parent directory fsyncs, write and fsync a temp file, publish with `link()` so an existing final log is never overwritten, fsync the session directory, then remove the redundant temp hard link.
-Windows creates missing directories through a durable staging publish: create a random sibling directory, then publish it to the final directory name with `MoveFileExW(..., MOVEFILE_WRITE_THROUGH)` without `MOVEFILE_REPLACE_EXISTING` or `MOVEFILE_COPY_ALLOWED`. File materialization writes and fsyncs the temp log, then publishes that temp file to the final path with the same write-through `MoveFileExW` call and no replacement. `koffi` is the minimal Win32 bridge for this API surface; its install script is allowed in `pnpm-workspace.yaml` because the package ships the native loader and prebuilt platform modules.
+Windows creates missing directories through a durable staging publish: create a random sibling directory under the constant `.dsh-mkdir-` prefix, independent of the target basename, then publish it to the final directory name with `MoveFileExW(..., MOVEFILE_WRITE_THROUGH)` without `MOVEFILE_REPLACE_EXISTING` or `MOVEFILE_COPY_ALLOWED`. File materialization writes and fsyncs the temp log, then publishes that temp file to the final path with the same write-through `MoveFileExW` call and no replacement. `koffi` is the minimal Win32 bridge for this API surface; its install script is allowed in `pnpm-workspace.yaml` because the package ships the native loader and prebuilt platform modules.
## Alternatives considered
@@ -28,6 +28,6 @@ Windows creates missing directories through a durable staging publish: create a
The backend keeps one external contract across platforms: first append either publishes a complete log at the final name or fails without overwriting an existing log. The platform split is an implementation detail; `SessionPersistence` APIs and the logical JSONL record format do not change. The later [Zstandard encoding decision](2026-07-19-zstandard-jsonl-session-logs.md) applies before either platform publishes the opaque bytes.
-Windows tests exercise the real Win32 publish path on native Windows. Power-loss behavior remains an API-contract property rather than something unit tests can prove; the testable invariants are that directory fsync is not called on Windows materialization, final-path collisions fail, temp logs are fsync'd before publication, and the resulting log loads normally.
+Windows tests exercise the real Win32 publish path on native Windows. Power-loss behavior remains an API-contract property rather than something unit tests can prove; the testable invariants are that directory fsync is not called on Windows materialization, final-path collisions fail, maximum-length target components remain materializable, temp logs are fsync'd before publication, and the resulting log loads normally.
Append and repair still use ordinary file-handle fsyncs on both platforms. A failed append closes its append-only handle, reopens the log read/write, truncates it to the pre-append size, and fsyncs the rollback because Windows rejects `ftruncate` on append-only handles.
diff --git a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.i18n.yaml
index b76ed1a9ae..1ab141230b 100644
--- a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.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
-2026-07-10-single-file-executable-sdk-runtime-distribution.md: 43ba5708d1216c37a7ad7e2904df7d2a6baf016d
-2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md: 3b33ff870d745584d2988bb6a7eb1a31e56ec3da
+2026-07-10-single-file-executable-sdk-runtime-distribution.md: 39cfb2999dea7767a18702ad7d160c9e88d7bf20
+2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md: e1a21c40647e1418d4afd02c0bc6b44ef0d4a8cf
diff --git a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md
index 43ba5708d1..39cfb2999d 100644
--- a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md
+++ b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md
@@ -25,7 +25,7 @@ Terminology reminder: pkg's `/snapshot` VFS has nothing to do with this repo's t
### The serving surface is a plugin: the two packages ui/jsonrpc + examples/jsonrpc-demo
-The deterministic protocol implementation (`server.ts` / `transport.ts`) lands as two packages on the existing `ui/acp` + `examples/acp-demo` pattern — the serving surface is itself a plugin:
+The deterministic protocol implementation (`server.ts` / `transport.ts`) lands as two packages on the existing `acp/acp` + `examples/acp-demo` pattern — the serving surface is itself a plugin:
- [`packages/ui/jsonrpc`](../../../../packages/ui/jsonrpc/README.md) (`@deepseek-ai/dsh-jsonrpc`): the pure protocol plugin; on apply it mounts `HarnessSdkServer` plus a line-delimited JSON-RPC transport on the process stdio, with disposal through `ctx.effect()`. Whether to serve is decided by `cordis.yml`; a yml that does not mount it is a legitimate process that does not serve. Protocol-level exit belongs to the plugin (after answering the `shutdown` request it disposes its own fiber, then `exit(0)`; an HMR-style unload only stops the service without exiting the process).
- [`packages/examples/jsonrpc-demo`](../../../../packages/examples/jsonrpc-demo/README.md) (`@deepseek-ai/dsh-jsonrpc-demo`): a thin app bin — `installFailLoud` + `loadEnv` + config discovery + `boot()` from [`dsh-app-boot`](../../../../packages/ui/app-boot/src/index.ts), done once boot completes; the server is brought up by the `dsh-jsonrpc` entry in the yml. Its only dependency is app-boot. Process-level exit belongs to the bin (stdin EOF/SIGTERM → dispose then 0, SIGINT → 130).
diff --git a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md
index 3b33ff870d..e1a21c4064 100644
--- a/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md
@@ -25,7 +25,7 @@ exe 使用 [@yao-pkg/pkg](https://github.com/yao-pkg/pkg)(vercel/pkg 归档后
### 对外服务接口也是插件:ui/jsonrpc + examples/jsonrpc-demo 两包
-确定性协议实现(`server.ts` / `transport.ts`)按 `ui/acp` + `examples/acp-demo` 的既有模式落为两包——对外服务接口本身也是插件:
+确定性协议实现(`server.ts` / `transport.ts`)按 `acp/acp` + `examples/acp-demo` 的既有模式落为两包——对外服务接口本身也是插件:
- [`packages/ui/jsonrpc`](../../../../packages/ui/jsonrpc/README.md)(`@deepseek-ai/dsh-jsonrpc`):纯协议插件;执行 `apply` 时,在进程 stdio 上挂载 `HarnessSdkServer` 与按行传输的 JSON-RPC 层,资源释放走 `ctx.effect()`。是否提供服务由 `cordis.yml` 决定;未挂载该插件的配置会启动一个不提供此服务的合法进程。协议级退出归插件所有(应答 `shutdown` 请求后 dispose 自身 fiber,再调用 `exit(0)`;HMR 式卸载只停止服务,不退出进程)。
- [`packages/examples/jsonrpc-demo`](../../../../packages/examples/jsonrpc-demo/README.md)(`@deepseek-ai/dsh-jsonrpc-demo`):轻量应用入口——`installFailLoud` + `loadEnv` + 配置发现 + [`dsh-app-boot`](../../../../packages/ui/app-boot/src/index.ts) 的 `boot()`;`boot()` 完成后入口即完成,服务器由 `cordis.yml` 中的 `dsh-jsonrpc` 条目启动。它只依赖 `app-boot`。进程级退出归 `bin` 所有(stdin EOF/SIGTERM → dispose 后返回 0,SIGINT → 130)。
diff --git a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.i18n.yaml
index b003d4de31..cbaedf3ab9 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.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
-2026-07-12-agent-scope-runtime-design.md: cf42bbacbfb9d2fcbd72aafba5ea02ddf83e1ce5
-2026-07-12-agent-scope-runtime-design.zh.md: d0d599664290dc7dc4debe1ccf8f419a40912ba1
+2026-07-12-agent-scope-runtime-design.md: 232fc02d66411b5ee8a21943795a3be4713bf238
+2026-07-12-agent-scope-runtime-design.zh.md: 39d558f8cde0183a3590d268aca36ea85e5f5c63
diff --git a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.md b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.md
index cf42bbacbf..232fc02d66 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.md
+++ b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.md
@@ -288,9 +288,9 @@ An ACP provider crosses a real process and wire boundary, so it retains validati
Start resolves only after `initialize` and `newSession` succeed. Abort, spawn failure, RPC failure, or invalid startup response reaps the process before rejection. After readiness, result maps the ACP prompt outcome and streamed output; dispose requests cancellation, closes the connection, and awaits process exit through one memoized path.
-## Workflows and ACP UI: retain only independent async facts
+## Workflows and ACP processes: retain only independent async facts
-Worker and editor bridges need more state than same-process registries because messages, process death, and rendering can settle independently. Their state is organized around those real facts rather than duplicate cancellation protocols.
+Worker and child-process bridges need more state than same-process registries because messages, process death, and cleanup can settle independently. Their state is organized around those real facts rather than duplicate cancellation protocols.
### Workflow children are pending starts or published records
@@ -306,11 +306,11 @@ The workflow result records the first accepted terminal outcome according to the
Public disposal claims its memoized promise before invoking callbacks. Worker death closes admission before processing any queued late child request, synthesizes missing lifecycle ends, and starts child/process cleanup without rewriting an outcome already claimed.
-### ACP prompt settlement does not depend on rendering success
+### ACP prompt settlement does not depend on update delivery
-The ACP UI correlates a prompt with its observed turn directly. It does not scan from a `logWatermark` or use session status as a second reconciliation oracle.
+The [automation-only ACP bridge](../simplification/2026-07-23-acp-automation-only-protocol.md) correlates one in-flight prompt with its observed user-message turn directly. It does not scan from a log watermark or use session status as a second reconciliation oracle.
-Prompt handling settles correlation in a `finally` around transcript rendering. A rendering failure can fail presentation, but it cannot skip prompt settlement or leave the session permanently in flight. Concurrent loads of the same persisted caller-supplied session ID remain excluded because that is a real persistence identity race, not a UUID collision concern.
+The session-event listener settles correlation from the matching `turn/end` even when a committed-message update cannot reach the client. Update delivery therefore cannot leave the session permanently in flight. ACP creates server-assigned fresh session ids and owns every resulting agent handle until connection teardown.
## Correctness enforcement
diff --git a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.zh.md b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.zh.md
index d0d5996642..39d558f8cd 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.zh.md
@@ -288,9 +288,9 @@ ACP 提供方跨越真实的进程和协议格式边界,因此它保留验证
Start 仅在 `initialize` 和 `newSession` 成功后才 resolve。Abort、spawn 失败、RPC 失败或无效启动响应在拒绝前回收进程。就绪后,result 映射 ACP 提示词结果和流式输出;dispose 请求取消、关闭连接并通过一条记忆化路径等待进程退出。
-## 工作流与 ACP UI:仅保留独立的异步事实
+## 工作流与 ACP 进程:仅保留独立的异步事实
-Worker 和编辑器桥接比同进程注册表需要更多状态,因为消息、进程死亡和渲染可以独立结算。它们的状态围绕这些真实事实组织,而非重复的取消协议。
+Worker 和子进程桥接比同进程注册表需要更多状态,因为消息、进程死亡和清理可以独立结算。它们的状态围绕这些真实事实组织,而非重复的取消协议。
### 工作流子级是待定 start 或已发布记录
@@ -306,11 +306,11 @@ Worker 边界仍然序列化请求和结果。宿主保留首个终端结果仲
公开 dispose 在调用回调之前取得其记忆化 promise 的所有权。Worker 死亡在处理任何排队的迟到子级请求之前关闭准入,合成缺失的生命周期结束,并启动子级/进程清理而不重写已声明的结果。
-### ACP 提示词结算不依赖渲染成功
+### ACP 提示词结算不依赖更新投递
-ACP UI 直接将提示词与其观察到的轮次关联。它不从 `logWatermark` 扫描,也不使用会话状态作为第二个调和预言机。
+[仅面向自动化的 ACP 桥接层](../simplification/2026-07-23-acp-automation-only-protocol.md)直接将一个进行中的提示词与其观察到的用户消息轮次关联。它不从日志水位线扫描,也不使用会话状态作为第二个调和预言机。
-提示词处理在 transcript(文本记录)渲染的 `finally` 中结算关联。渲染失败可以导致展示失败,但不能跳过提示词结算或让会话永久处于进行中状态。对同一持久化的调用方提供的会话 ID 的并发加载仍被排除,因为那是真实的持久化标识竞争,而非 UUID 碰撞问题。
+即使已提交消息的更新无法送达客户端,会话事件监听器也会从匹配的 `turn/end` 结算关联。因此更新投递不能让会话永久处于进行中状态。ACP 创建由服务器分配 id 的全新会话,并拥有由此产生的每个 agent 句柄,直到连接拆除。
## 正确性强制
diff --git a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.i18n.yaml
index b49506364b..46f0a1cd41 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.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
-2026-07-12-scoped-layers-store.md: b850b6bcbb22401b386b4458b6d5c65a160c85cd
-2026-07-12-scoped-layers-store.zh.md: 8bfc0a0e8ec1e3de624ff8d9e48b7517833fc025
+2026-07-12-scoped-layers-store.md: c5186d1652bca617eed62ec02937f2d055ea727c
+2026-07-12-scoped-layers-store.zh.md: 3183811be553428ebcd8f59f15989c44d458b477
diff --git a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.md b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.md
index b850b6bcbb..c5186d1652 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.md
+++ b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.md
@@ -123,4 +123,4 @@ All seven facades keep validation and diagnostics in their owning registry and c
- `dsh-scope` unit tests cover global construction, lazy scoped construction, non-creating reads, named merge order and shadowing, aggregate reclamation, factory and action failure cleanup, notification ordering and rollback, `notify: false`, effect labels, exact disposer identity, idempotent teardown, caller-owned duplicate errors, independent anonymous duplicates, live iterators, and drained-generation detachment.
- Focused tool, system-prompt, and command suites cover restrictions, reserved transport handling, known/restrictable-name agreement, guard re-entrancy and self-replacement, validation order, exact diagnostics, section shadow-before-evaluate, provider snapshot membership, variable re-entrancy and self-replacement, contained command observers, frozen and sorted views, direct execution, and lifecycle disposal.
- The scoped core-data type-equivalence check ties `ScopeLayer` documentation to its source declaration. Repository documentation, module-graph, build, hygiene, coverage, and built-artifact gates exercise the root export and package boundary.
-- Existing ACP, headless, and TUI keyless snapshots remain the regression boundary for tool schemas, prompt assembly, and human commands. The implementation does not update any expected transcript.
+- Existing ACP, headless, and TUI keyless snapshots remain the regression boundary for tool schemas and prompt assembly; TUI coverage owns human commands. The implementation does not update any expected transcript.
diff --git a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.zh.md b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.zh.md
index 8bfc0a0e8e..3183811be5 100644
--- a/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-12-scoped-layers-store.zh.md
@@ -123,4 +123,4 @@ export class AnonymousEntries {
- `dsh-scope` 单元测试覆盖全局构造、专属层延迟构造、非创建式读取、命名合并顺序与遮蔽、聚合回收、工厂与 action 失败清理、通知顺序与回滚、`notify: false`、effect 标签、原始 disposer 身份、幂等拆除、调用方提供的重名错误、相同匿名值的独立登记、活迭代器,以及表清空后的 generation 脱离。
- 工具、系统提示词和命令专项测试套件覆盖 restriction、保留传输处理、已知名称与可限制名称的一致性、guard 重入与自我替换、校验顺序、精确诊断、section 先遮蔽再求值、提供方快照成员关系、variable 重入与自我替换、隔离失败的命令观察者、冻结且有序的视图、直接执行和生命周期销毁。
- 作用域核心数据的类型等价性检查将 `ScopeLayer` 文档与其源声明绑定。仓库级的文档、模块图、构建、hygiene、覆盖率与构建产物门禁会覆盖包根导出与包边界。
-- 现有 ACP(Agent Client Protocol)、headless 和 TUI 无密钥快照继续作为工具 schema、提示词组装和人类命令的回归边界。实现不会更新任何预期 transcript(文本记录)。
+- 现有 ACP(Agent Client Protocol)、headless 和 TUI 无密钥快照继续作为工具 schema 与提示词组装的回归边界;人类命令由 TUI 覆盖。实现不会更新任何预期 transcript(文本记录)。
diff --git a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.i18n.yaml
index e83cfff95e..52ec59ccbf 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.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
-2026-07-15-llm-model-catalog-and-acp-selection.md: 6cc8afc6c7431fbf3eb29fc358b432db4f72b529
-2026-07-15-llm-model-catalog-and-acp-selection.zh.md: 1cce7a58d0ec83dc01feaf72ccb61d294a78ddd5
+2026-07-15-llm-model-catalog-and-acp-selection.md: adbd8671f0ea5cd2e0c049c32616453882329396
+2026-07-15-llm-model-catalog-and-acp-selection.zh.md: dfcb581e43149281cd18b28f1411ea98944983ad
diff --git a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.md b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.md
index 6cc8afc6c7..adbd8671f0 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.md
+++ b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.md
@@ -4,6 +4,8 @@ Status: implemented
English | [中文](2026-07-15-llm-model-catalog-and-acp-selection.zh.md)
+> The catalog decision remains current. Per-session ACP model selection is superseded by [ACP as an automation-only protocol](../simplification/2026-07-23-acp-automation-only-protocol.md).
+
## Problem
Provider-routed adapters let every request choose `provider + model`, but `LlmService` exposed only routing and streaming. A UI could not discover which providers were registered or which models an adapter was prepared to recommend. ACP clients therefore received no `model` session config option, so Zed, JetBrains, and VS Code integrations had no model list even though the request seam already supported runtime switching.
@@ -24,21 +26,17 @@ Catalog membership is advisory. It drives selectors and diagnostics but never ch
`dsh-llm-pi-ai` maps the configured provider's installed `getModels(provider)` entries into the neutral catalog. Its existing request-time catalog lookup remains authoritative and still rejects unknown models with `UNKNOWN_MODEL`. `dsh-llm-deepseek` accepts an optional `models` config containing display entries, defaulting to `deepseek-v4-flash` and `deepseek-v4-pro`. An explicit list replaces those defaults and an empty list disables discovery. The entries improve selector UX for known public or private models, while every unlisted model id continues to pass through unchanged.
-### ACP session config option
+### Per-session selection in the front door
-The ACP bridge advertises one select with `id: model` and `category: model` in `session/new` and `session/load` when the session has a complete target whose provider is registered. Each opaque option value encodes the full provider/model pair. Models are grouped by provider when multiple non-empty provider groups exist; a single group is flattened for clients that render simple selects better.
+A selection is owned by the front door that offers it (today the TUI `/model` selector), never by `LlmService` or `AgentOptions`: those are deployment-wide or creation-wide objects, and mutating them would couple concurrent sessions. Each opaque choice carries the full provider/model pair, because the same model id may appear under multiple routes.
-The session's current target is added to the displayed options when its adapter omits it. This preserves custom DeepSeek and private-endpoint models while keeping the adapter catalog advisory. A target with an unregistered provider is not advertised, and a model-less agent remains available to another `agent/request` supplier.
-
-`session/set_config_option` accepts only values from the current catalog snapshot and updates a target reference owned by that ACP session. No global `LlmService` or `AgentOptions` state changes, so concurrent sessions may select different providers and models. The existing permission select remains independent, and every response returns the complete refreshed option state.
+The ACP automation transport is not a catalog consumer. Its deployment config supplies one optional provider/model target for newly created agents, and it advertises no model selector or configuration-option interface.
### Prompt/request consistency and durability
-Agent setup installs scoped `system-prompt/assemble` and `agent/request` listeners. Prompt assembly snapshots the selected pair once per step, overwrites the assembled `provider` and `model` variables after downstream prompt listeners, and the request listener applies that same snapshot after downstream request listeners. A selection during asynchronous assembly therefore starts on the next step rather than splitting prompt text from routing. Other call-config fields remain untouched.
+`installAgentLlmTarget` (in `dsh-agent`) installs scoped `system-prompt/assemble` and `agent/request` listeners for a front-door-owned target. Prompt assembly snapshots the selected pair once per step, overwrites the assembled `provider` and `model` variables after downstream prompt listeners, and the request listener applies that same snapshot after downstream request listeners. A selection during asynchronous assembly therefore starts on the next step rather than splitting prompt text from routing. Other call-config fields remain untouched.
-The request header remains the durable source of truth. When a selected target is actually used, the existing full `request/header` snapshot records it. `session/load` initializes the ACP selection from the folded last request header before falling back to bridge config. A selection that is never used by a request is intentionally in-memory only because it never became model-visible state.
-
-ACP's experimental `providers/*` capability is not used. That draft surface configures provider base URLs, protocols, and headers, including secrets; it does not enumerate models and would give the UI authority to rewrite deployment-owned adapter configuration.
+The request header remains the durable source of truth. When a selected target is actually used, the existing full `request/header` snapshot records it, and a front door initializes its selection from the folded last request header before falling back to creation options. A selection that is never used by a request is intentionally in-memory only because it never became model-visible state.
## Alternatives considered
@@ -46,21 +44,19 @@ ACP's experimental `providers/*` capability is not used. That draft surface conf
**Make catalogs mandatory whitelists.** This conflicts with the hand-written adapter's arbitrary model pass-through and private deployments. The selected adapter already owns authoritative request validation.
-**Store selection in `AgentOptions` or `LlmService`.** Those are creation-wide or deployment-wide objects. Mutating them would couple concurrent ACP sessions and bypass the logged `agent/request` replacement path.
+**Store selection in `AgentOptions` or `LlmService`.** Those are creation-wide or deployment-wide objects. Mutating them would couple concurrent sessions and bypass the logged `agent/request` replacement path.
**Persist a new model-selection session event immediately.** An unused UI selection has not affected a model request. Recording the existing request header when the target is consumed preserves the model-visible-if-and-only-if-logged rule without adding a second source of truth.
-**Use ACP `providers/*`.** That unstable API changes endpoint and authentication configuration rather than selecting a model for one session, and its lifecycle and secret-handling semantics do not match this feature.
-
## Consequences
- Any adapter can expose a dynamic model list without leaking provider-library types into the core seam.
- Catalog consumers must treat absence as “not advertised,” never “invalid request.”
-- pi-ai-backed ACP deployments automatically inherit the installed pi-ai provider catalogs; hand-written DeepSeek deployments list known choices explicitly and retain arbitrary model support.
-- ACP clients receive a standard stable model config option, with provider-aware values and per-session isolation.
+- pi-ai adapters expose their installed provider catalogs; hand-written DeepSeek deployments list known choices explicitly and retain arbitrary model support.
+- Human-facing catalog consumers own their selection interaction. ACP uses its fixed deployment target and does not widen the protocol with model discovery.
- Request headers remain compatible with the provider-routed session shape; no new JSONL event or format version is required.
-- A catalog read can be asynchronous. ACP reads a detached snapshot before creating or resuming an agent, so discovery failure cannot leave a partially published session.
+- A catalog read can be asynchronous, and every caller receives detached values.
## Testing
-Unit coverage validates catalog detachment and malformed metadata, pi-ai and DeepSeek catalog projection, ACP provider grouping, custom-current insertion, invalid values, provider/model request routing, prompt-variable alignment, concurrent-session isolation, model-less fallback, and load restoration from the request header. The existing ACP transport suites verify that the additional config option does not change prompt, cancellation, replay, approval, or tool-rendering behavior.
+Unit coverage validates catalog detachment and malformed metadata, pi-ai and DeepSeek catalog projection, provider/model request routing, and prompt-variable alignment; per-agent isolation follows from installing the listeners on the agent-scoped context. ACP transport tests validate fixed provider/model forwarding independently of catalog discovery; the TUI suite covers selector interaction and header-based restoration.
diff --git a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.zh.md b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.zh.md
index 1cce7a58d0..dfcb581e43 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-15-llm-model-catalog-and-acp-selection.zh.md
@@ -4,6 +4,8 @@ Status: implemented
[English](2026-07-15-llm-model-catalog-and-acp-selection.md) | 中文
+> 目录决策仍然有效。ACP 会话级模型选择已由 [ACP 作为仅面向自动化的协议](../simplification/2026-07-23-acp-automation-only-protocol.md)取代。
+
## 问题
基于提供方路由的适配器允许每次请求选择 `provider + model`,但 `LlmService` 只暴露路由和流式调用。UI 无法发现已注册的提供方,也无法知道适配器愿意推荐哪些模型。因此,ACP 客户端收不到 `model` 会话配置项;即使请求接缝已经支持运行时切换,Zed、JetBrains 和 VS Code 集成仍没有模型列表。
@@ -14,9 +16,9 @@ ACP 选择还必须保留提供方维度。同一个模型 ID 可能存在于多
## 决策
-### 提供方中立的建议性发现
+### 提供方无关的建议性发现
-`LlmAdapter` 增加 `providerInfo(provider)` 与异步 `listModels(provider)` 方法。其提供方中立结果分别为 `LlmProviderInfo { id, name }` 和 `LlmModelInfo { provider, id, name, description? }`。默认实现以路由名称作为提供方名称,并且不展示模型,从而保持现有适配器行为。
+`LlmAdapter` 增加 `providerInfo(provider)` 与异步 `listModels(provider)` 方法。其提供方无关结果分别为 `LlmProviderInfo { id, name }` 和 `LlmModelInfo { provider, id, name, description? }`。默认实现以路由名称作为提供方名称,并且不展示模型,从而保持现有适配器行为。
`LlmService.listProviders()` 按注册顺序返回分离后的元数据。`LlmService.listModels(provider)` 委托给路由所有者,校验非空 ID 和名称,并在提供方不匹配或模型 ID 重复时以 `INVALID_CATALOG` 失败,最后返回分离后的值。未知提供方仍以 `NO_ADAPTER` 失败。提供方元数据在 `registerAdapter()` 期间进行原子校验,错误展示记录不会留下部分注册。
@@ -24,43 +26,37 @@ ACP 选择还必须保留提供方维度。同一个模型 ID 可能存在于多
`dsh-llm-pi-ai` 将已配置提供方的安装目录 `getModels(provider)` 映射为中立目录。其现有请求时目录查询仍是权威依据,未知模型仍以 `UNKNOWN_MODEL` 失败。`dsh-llm-deepseek` 接受可选的 `models` 配置作为展示条目,默认包含 `deepseek-v4-flash` 和 `deepseek-v4-pro`。显式列表会替换这些默认值,空列表则关闭发现。这些条目改善已知公开或私有模型的选择体验,而所有未列出的模型 ID 仍会原样透传。
-### ACP 会话配置项
+### 前门内的会话级选择
-当会话具有完整目标且目标提供方已注册时,ACP bridge 会在 `session/new` 与 `session/load` 中展示一个 `id: model`、`category: model` 的选择项。每个不透明选项值都编码完整的提供方/模型字段组合。存在多个非空提供方分组时按提供方分组;只有一个分组时将其展开,以便对简单选择器支持更好的客户端展示。
+选择由提供它的前门拥有(今天是 TUI 的 `/model` 选择器),而不由 `LlmService` 或 `AgentOptions` 拥有:它们是部署级或创建级对象,改动它们会把并发会话耦合在一起。每个不透明选项都携带完整的提供方/模型对,因为同一模型 ID 可能出现在多个路由下。
-如果适配器目录未包含会话当前目标,该目标仍会加入展示选项。这能保留自定义 DeepSeek 与私有端点模型,同时维持目录的建议性。提供方未注册的目标不会展示;缺少模型的 agent 仍可由其他 `agent/request` 提供者补齐。
+ACP 自动化传输层不是目录消费方。它通过部署配置为新创建的 agent 提供一个可选的提供方/模型目标,不展示模型选择器或配置选项接口。
-`session/set_config_option` 只接受当前目录快照中的值,并更新该 ACP 会话独占的目标引用。它不会修改全局 `LlmService` 或 `AgentOptions` 状态,因此并发会话可以选择不同的提供方和模型。现有权限选择项保持独立,每次响应都返回完整的刷新后配置项状态。
+### Prompt/请求一致性与持久化
-### Prompt/请求一致性与持久化
+`installAgentLlmTarget`(位于 `dsh-agent`)为前门拥有的目标安装 agent 作用域的 `system-prompt/assemble` 与 `agent/request` 监听器。Prompt 组装在每个 step 对所选组合做一次快照,在下游 prompt 监听器之后覆写组装出的 `provider` 与 `model` 变量;请求监听器在下游请求监听器之后应用同一快照。因此,发生在异步组装期间的选择会从下一个 step 生效,而不会让 prompt 文本与路由分裂。其他调用配置字段保持不变。
-Agent setup 会安装作用域内的 `system-prompt/assemble` 与 `agent/request` 监听器。Prompt 组装为每个 step 只快照一次选中的字段组合,在下游 prompt 监听器完成后覆盖组装结果中的 `provider` 与 `model` 变量;请求监听器则在下游请求监听器完成后应用同一个快照。因此,异步组装期间发生的选择会从下一个 step 生效,不会导致 prompt 文本与路由分裂。其他调用配置字段保持不变。
-
-请求头仍是持久化事实来源。当选中目标被实际使用时,现有的完整 `request/header` 快照会记录它。`session/load` 先从折叠后的最后请求头初始化 ACP 选择,再回退到 bridge 配置。一个从未被请求使用的选择只保留在内存中,因为它从未成为模型可见状态。
-
-本功能不使用 ACP 的实验性 `providers/*` 能力。该草案接口配置提供方 base URL、协议和 headers,其中可能包含密钥;它不枚举模型,并且会赋予 UI 改写部署所有的适配器配置的权力。
+请求头仍是持久化的事实来源。当所选目标真正被使用时,现有的完整 `request/header` 快照会记录它;前门先从折叠后的最后一个请求头初始化其选择,然后才回退到创建选项。从未被请求使用的选择有意只保留在内存中,因为它从未成为模型可见状态。
## 考虑过的替代方案
-**只返回模型字符串。** 仅模型值会丢失提供方路由;两个提供方暴露相同 ID 时立刻产生歧义。
+**只返回模型字符串。** 只有模型的值会丢失提供方路由,一旦两个提供方暴露相同 ID 就会产生歧义。
**将目录设为强制白名单。** 这与手写适配器的任意模型透传和私有部署冲突。请求的权威校验本就属于被选中的适配器。
-**将选择存入 `AgentOptions` 或 `LlmService`。** 这些对象分别面向创建过程或整个部署。修改它们会耦合并发 ACP 会话,并绕开带日志归因的 `agent/request` 替换路径。
+**把选择存进 `AgentOptions` 或 `LlmService`。** 它们是创建级或部署级对象。改动它们会把并发会话耦合在一起,并绕过有日志记录的 `agent/request` 替换路径。
-**立即写入新的模型选择会话事件。** 尚未使用的 UI 选择没有影响模型请求。目标被消费时记录现有请求头,既满足“模型可见当且仅当已记录”的规则,也不会引入第二个事实来源。
-
-**使用 ACP `providers/*`。** 该不稳定 API 用于修改端点与认证配置,而不是为单个会话选择模型;其生命周期和密钥处理语义都不适合本功能。
+**立即持久化一个新的模型选择会话事件。** 未被使用的 UI 选择尚未影响任何模型请求。在目标被消费时记录现有请求头,既保持“模型可见当且仅当有日志”的规则,又不会引入第二个事实来源。
## 结果
- 任意适配器都能暴露动态模型列表,无需把提供方库类型泄漏到核心接缝。
- 目录消费者必须把缺失理解为“未展示”,而不是“请求无效”。
-- 基于 pi-ai 的 ACP 部署会自动继承已安装的 pi-ai 提供方目录;手写 DeepSeek 部署显式列出已知选项,同时保留任意模型能力。
-- ACP 客户端会收到稳定标准的模型配置项,其中的值保留提供方信息,并按会话隔离。
-- 请求头继续使用基于提供方路由的会话结构;不需要增加 JSONL 事件或格式版本。
-- 目录读取可以是异步的。ACP 在创建或恢复 agent 前读取分离后的快照,因此发现失败不会留下部分发布的会话。
+- pi-ai 适配器会暴露其已安装的提供方目录;手写 DeepSeek 部署显式列出已知选项,同时保留对任意模型的支持。
+- 面向人类的目录消费方拥有各自的选择交互。ACP 使用固定部署目标,不会为模型发现扩大协议范围。
+- 请求头与基于提供方路由的会话形态保持兼容;不需要新的 JSONL 事件或格式版本。
+- 目录读取可以是异步的,且每个调用方都会收到分离后的值。
## 测试
-单元测试覆盖目录分离与错误元数据、pi-ai 和 DeepSeek 目录投影、ACP 提供方分组、自定义当前模型补入、无效值、提供方/模型请求路由、prompt 变量一致性、并发会话隔离、无模型回退,以及从请求头恢复选择。现有 ACP 传输测试验证新增配置项不会改变 prompt、取消、回放、审批或工具展示行为。
+单元测试覆盖目录分离与错误元数据、pi-ai 和 DeepSeek 目录投影、提供方/模型请求路由,以及 prompt 变量对齐;按 agent 的隔离来自监听器安装在 agent 作用域上下文这一事实。ACP 传输测试独立验证固定提供方/模型的转发行为;TUI 套件覆盖选择器交互与基于请求头的恢复。
diff --git a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.i18n.yaml
index 70b98f0cf1..005f23b151 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.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
-2026-07-15-lsp-capability-seam.md: 7265b04ac9b2f83764bdd13f07b2d3404c4c1708
-2026-07-15-lsp-capability-seam.zh.md: 10e8956005045d0934dd9dada5718b85a34cda3f
+2026-07-15-lsp-capability-seam.md: d96b3a9c5139c1455a51f4fff793293d7b5a11c0
+2026-07-15-lsp-capability-seam.zh.md: 54dd32e46dded5722dda910e9138879d3f99de07
diff --git a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.md b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.md
index 7265b04ac9..d96b3a9c51 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.md
+++ b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.md
@@ -18,7 +18,7 @@ Add LSP as a three-package capability seam with one read-only model tool and one
1. `@deepseek-ai/dsh-lsp` at `packages/lsp/lsp` owns `ctx.lsp`, provider registration and selection, normalized requests/results, execution control, and structured LSP errors.
2. `@deepseek-ai/dsh-lsp-local` at `packages/lsp/lsp-local` adapts configured stdio language servers to the seam. One plugin instance accepts a named server table and registers one isolated provider for each command and extension-to-language-id mapping.
-3. `@deepseek-ai/dsh-tool-lsp` at `packages/lsp/tool-lsp` owns the model-facing `lsp` schema, prompt guidance, argument validation, result limits and formatting, and ACP presentation.
+3. `@deepseek-ai/dsh-tool-lsp` at `packages/lsp/tool-lsp` owns the model-facing `lsp` schema, prompt guidance, argument validation, result limits and formatting, and transport-neutral UI presentation.
`dsh-lsp-local` is a generic host, not a language-server catalog or installer. Deployments explicitly configure commands and mappings; future presets belong in composition plugins or `cordis.yml` overlays.
@@ -100,7 +100,7 @@ The tool requires `workspaceRoot` from session `header.cwd`, with no fallback; a
Locations render as stable, file-grouped `path:line:character` entries. A `file:` URI accepted by Node `fileURLToPath()` becomes a relative path inside the workspace or an absolute path outside it; other URIs remain verbatim. `maxLocations` defaults to `100` and reports omitted items; `maxResultChars` defaults to `16_000` and bounds every complete rendered result, including its truncation metadata. Empty locations and `null` hover are successful no-result responses; missing or malformed server payloads fail with structured `LSP_MALFORMED_RESPONSE` errors.
-ACP uses `{ card: 'generic', kind: 'search', title, locations: [{ path: file_path, line }] }` with an args-derived operation/cursor `title`. Because `FileLocation` has no character, follow-along focuses the input line while the title preserves the cursor; presentation remains pure.
+The transport-neutral presenter uses `{ card: 'generic', kind: 'search', title, locations: [{ path: file_path, line }] }` with an args-derived operation/cursor `title`. Because `FileLocation` has no character, follow-along focuses the input line while the title preserves the cursor; presentation remains pure.
## Timeout ownership
@@ -174,7 +174,7 @@ The local provider trusts its configured server and claims no sandbox confinemen
## Testing
- Package tests pin the three-package dependency direction, runtime injections, and `ctx.lsp`-only communication.
-- Tool tests pin the four operations, coordinate validation, configured bounds and omission markers, prompt, and ACP presentation.
+- Tool tests pin the four operations, coordinate validation, configured bounds and omission markers, prompt, and UI presentation.
- Registry tests pin atomic reservation/release, order-independent selection, and structured unavailable, disposed, conflict, and unsupported-operation errors.
- Fake-stdio tests pin exact initialization capabilities, four protocol mappings, `Location`/`LocationLink` and hover normalization, and `findReferences` mapping to `references.includeDeclaration`.
- Synchronization tests pin UTF-16 negotiation and conversion, supported and rejected `textDocumentSync` forms, blocked and failed open writes, balanced transient open/close, close-write failure, and malformed-response rejection.
@@ -182,7 +182,7 @@ The local provider trusts its configured server and claims no sandbox confinemen
- Lifecycle tests pin startup single-flight, complete-lifecycle serialization with fresh queued source reads, cross-workspace parallelism, abortable queues, crash replacement without replay, failed-stdin teardown, and quiescent disposal.
- Host-filesystem tests pin session-cwd requirements, relative and absolute source containment through symlinks, document validation, file/non-file URI rendering, unformatted source, and no `fs/observed` event.
- A keyless pinned TypeScript real-server e2e exercises all four operations; runnable configuration uses the same explicit provider mapping.
-- Snapshots cover model-visible schema, prompt, results, omissions, and ACP rendering; a built-artifact smoke test covers framing and cleanup.
+- Snapshots cover model-visible schema, prompt, results, and omissions; a built-artifact smoke test covers framing and cleanup.
- Package and architecture docs cover configuration, security boundaries, and search/read guidance; the new `packages/lsp/` group is added to the AGENTS.md repository-layout block, the packages/README.md group table, and architecture.md in the same change.
## Consequences
diff --git a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.zh.md b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.zh.md
index 10e8956005..54dd32e46d 100644
--- a/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-15-lsp-capability-seam.zh.md
@@ -18,7 +18,7 @@ harness 已具备文本搜索与文件读取能力,但二者都无法识别程
1. `packages/lsp/lsp` 下的 `@deepseek-ai/dsh-lsp` 负责 `ctx.lsp`、提供方注册与选择、标准化请求与结果、执行控制,以及结构化 LSP 错误。
2. `packages/lsp/lsp-local` 下的 `@deepseek-ai/dsh-lsp-local` 将配置的 stdio 语言服务器适配到该服务边界。一个插件实例接收具名服务器表,并为每组命令及扩展名到语言 id 的映射注册一个隔离的提供方。
-3. `packages/lsp/tool-lsp` 下的 `@deepseek-ai/dsh-tool-lsp` 负责面向模型的 `lsp` schema、提示词指导、参数校验、结果限制与格式化,以及 ACP(Agent Client Protocol)展示。
+3. `packages/lsp/tool-lsp` 下的 `@deepseek-ai/dsh-tool-lsp` 负责面向模型的 `lsp` schema、提示词指导、参数校验、结果限制与格式化,以及与传输方式无关的 UI 展示。
`dsh-lsp-local` 是通用 host,不是语言服务器目录或安装器。部署显式配置命令与映射;未来 preset 属于组合插件或 `cordis.yml` overlay。
@@ -100,7 +100,7 @@ interface LspToolInput {
位置按文件稳定分组并渲染为 `path:line:character`。Node `fileURLToPath()` 可接受的 `file:` URI 在工作区内转换为相对路径,在工作区外转换为绝对路径;其他 URI 保持原样。`maxLocations` 默认值为 `100`,并报告省略的条目;`maxResultChars` 默认值为 `16_000`,并限制每个完整渲染结果,其中包括截断元数据。空位置与 `null` hover 是成功的无结果响应;服务器载荷缺失或格式错误时,以结构化 `LSP_MALFORMED_RESPONSE` 错误失败。
-ACP 使用 `{ card: 'generic', kind: 'search', title, locations: [{ path: file_path, line }] }`,`title` 由参数推导并标明操作与光标。由于 `FileLocation` 没有 character,跟随位置聚焦输入行,标题保留完整光标;展示保持纯函数。
+与传输方式无关的展示器使用 `{ card: 'generic', kind: 'search', title, locations: [{ path: file_path, line }] }`,`title` 由参数推导并标明操作与光标。由于 `FileLocation` 没有 character,跟随位置聚焦输入行,标题保留完整光标;展示保持纯函数。
## 超时归属
@@ -174,7 +174,7 @@ ACP 使用 `{ card: 'generic', kind: 'search', title, locations: [{ path: file_p
## 测试
- 包测试固定三个包的依赖方向、运行时注入和仅通过 `ctx.lsp` 通信的边界。
-- 工具测试固定四种操作、坐标校验、配置限制与省略标记、提示词和 ACP 展示。
+- 工具测试固定四种操作、坐标校验、配置限制与省略标记、提示词和 UI 展示。
- 注册表测试固定原子占用/释放、不受顺序影响的选择,以及结构化的不可用、已释放、冲突和不支持操作错误。
- 测试用 stdio server 固定精确的初始化能力、四种协议映射、`Location`/`LocationLink` 与 `hover` 归一化,以及 `findReferences` 到 `references.includeDeclaration` 的映射。
- 同步测试固定 UTF-16 协商与转换、受支持和被拒绝的 `textDocumentSync` 形式、打开写入阻塞与失败、配对的临时打开/关闭、关闭写入失败和错误响应拒绝。
@@ -182,7 +182,7 @@ ACP 使用 `{ card: 'generic', kind: 'search', title, locations: [{ path: file_p
- 生命周期测试固定启动 single-flight、完整生命周期串行化及排队查询读取最新源文件、跨工作区并行、可取消队列、崩溃后不重放的替换、stdin 失败后的进程拆除,以及释放后完全停稳。
- 主机文件系统测试固定 session cwd 要求、符号链接下相对与绝对源路径的规范 containment、文档校验、file/non-file URI 渲染、无格式源文本和不发送 `fs/observed`。
- 无密钥且固定版本的 TypeScript 真实服务器 e2e 覆盖四种操作;可运行配置使用同一项显式提供方映射。
-- 快照覆盖模型可见 schema、提示词、结果、省略提示和 ACP 渲染;构建产物冒烟测试覆盖分帧与清理。
+- 快照覆盖模型可见 schema、提示词、结果和省略提示;构建产物冒烟测试覆盖分帧与清理。
- 包与架构文档覆盖配置、安全边界和搜索/读取指导;同一改动中,新的 `packages/lsp/` 包组要加入 AGENTS.md 的仓库布局块、packages/README.md 的分组表和 architecture.md。
## 影响
diff --git a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.i18n.yaml
index 16852c1004..19a6b628c3 100644
--- a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.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
-2026-07-20-unified-json-value-schema-dsl.md: 09945c413ffe5924c74076648cdf3da60c3e18c9
-2026-07-20-unified-json-value-schema-dsl.zh.md: 00a7a199613ea857a7815f1c7794781f143a3896
+2026-07-20-unified-json-value-schema-dsl.md: 5de3523eab15a91ea32dc09e2e239146fadea6f1
+2026-07-20-unified-json-value-schema-dsl.zh.md: 321136c31a6aa6c0268150fcde2d97dcdbb0ac58
diff --git a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.md b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.md
index 09945c413f..5de3523eab 100644
--- a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.md
+++ b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.md
@@ -21,6 +21,7 @@ Object-rooting is a consumer rule rather than a vocabulary restriction. Subagent
## Alternatives considered
- **Keep separate parameter and structured-output schema systems:** rejected because every added output construct would require parallel inference, compilation, validation, and code-generation changes with no useful ownership boundary.
+- **Use Schemastery for tool parameters:** rejected because Schemastery targets validation and transformation through Standard Schema rather than JSON Schema generation. It would add an adapter layer without producing the model-facing wire schema or the shared output vocabulary.
- **Adopt full JSON Schema or Ajv:** rejected because the harness must fail on every construct it cannot project into its generated SDK and validators; accepting a larger language would make enforcement and model guidance dishonest.
- **Make every object implicitly open or closed:** rejected because either choice hides a consequential author decision. Only the legacy-shaped implicit parameter root and external raw schema retain an intentional default.
- **Define `oneOf` as first-match:** rejected because branch ordering would change validation semantics and allow overlapping branches to hide ambiguous values.
@@ -32,4 +33,5 @@ Object-rooting is a consumer rule rather than a vocabulary restriction. Subagent
- Explicit object openness and type-correct literal constraints make malformed declarations fail during authoring or registration rather than during a later model call.
- Bounded type inference retains useful exact types for ordinary declarations and degrades unusually deep tails to `JsonValue`; runtime schema enforcement remains exact at every depth.
- Raw tools may still register broader JSON Schema directly, but unified code generation treats unsupported schemas as unknown instead of pretending to enforce them.
+- Per-property `required: true` remains the tool-author contract, and type-level regression coverage pins required keys as non-optional after the original inference path exposed an optionality bug.
- Runtime and compile-time tests cover every root, exact-one overlap/no-match behavior, raw open defaults, explicit openness, lossy JSON values, inference, deep nesting across core and dynamic projections, JSON-invisible dynamic keys, and exotic schema arrays.
diff --git a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.zh.md b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.zh.md
index 00a7a19961..321136c31a 100644
--- a/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-20-unified-json-value-schema-dsl.zh.md
@@ -21,6 +21,7 @@ Status: implemented
## 备选方案
- **保留两套独立的参数与结构化输出 schema 系统:**不予采纳。每新增一种输出结构,都必须分别修改类型推导、编译、校验和代码生成,而这种重复并未形成有意义的职责边界。
+- **使用 Schemastery 处理工具参数:**不予采纳。Schemastery 通过 Standard Schema 面向校验与转换,而不是生成 JSON Schema。采用它会增加一层适配器,却不能产出面向模型的协议 schema 或共享的输出词汇。
- **采用完整 JSON Schema 或 Ajv:**不予采纳。harness 必须拒绝所有无法投影到生成 SDK 和校验器中的结构;如果接受更大的语言子集,强制执行能力和模型指引就会与事实不符。
- **让所有对象默认开放或默认封闭:**不予采纳。这两种选择都会隐藏一项影响重大的作者决策。只有保持旧有形态的隐式参数根对象和外部原始 schema 才有意保留默认值。
- **把 `oneOf` 定义为首个匹配分支:**不予采纳。这样一来,分支顺序会改变校验语义,重叠分支也会掩盖值的歧义。
@@ -32,4 +33,5 @@ Status: implemented
- 显式的对象开放方式和类型正确的字面量约束会让格式错误的声明在编写或注册阶段快速失败,而不是拖到后续模型调用时才失败。
- 有界类型推导会为常规声明保留有用的精确类型,并将异常深的尾部结构退化为 `JsonValue`;运行时 schema 强制执行在任意深度仍保持精确。
- 原始工具仍可直接注册范围更广的 JSON Schema,但统一代码生成会把不受支持的 schema 视为未知类型,不会假装自己能够强制执行。
+- 每个属性的 `required: true` 仍是工具作者契约;原有推导路径暴露可选性缺陷后,类型级回归覆盖会锁定必填键不得为可选。
- 运行时和编译期测试覆盖所有根类型、恰好匹配一个分支时的重叠/无匹配行为、原始 schema 的默认开放语义、显式开放方式、有损 JSON 值、类型推导、核心投影和动态投影中的深层嵌套、动态注册中 JSON 不可见的键,以及非普通 schema 数组。
diff --git a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.i18n.yaml
index 0ed8f5d7a4..eacbd89847 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.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
-2026-07-22-slot-type-chain-implementation.md: 65b4ebb475fe34d71d8d3a08878b40103b3c95bd
-2026-07-22-slot-type-chain-implementation.zh.md: 4c55171ca0118782568e17f349f83d6cf9211617
+2026-07-22-slot-type-chain-implementation.md: 617524475f3da8af5d281efcfe8f79d500f31be8
+2026-07-22-slot-type-chain-implementation.zh.md: 52edea30acea5989b3438cbcf4688df5a897f099
diff --git a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.md b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.md
index 65b4ebb475..617524475f 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.md
+++ b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.md
@@ -42,7 +42,7 @@ Parity rule: **the declaring entry holds the exclusive right to render its child
| Share | Type | Source of truth | Contents |
|---|---|---|---|
-| runtime | `PropsRuntime` | SlotMap entry for K | `OwnerOf` (render-site params) + session-scope standard `useSession`/`sessionId` + global `useSessions` |
+| runtime | `PropsRuntime` | SlotMap entry for K | `OwnerOf` (render-site params) + session-scope standard `useSession`/`sessionId` + global `useSessions`/`useWorkspaces` |
| child render | `PropsRenderSlots` | register's `children` keys | `renderSlot(key, owner)`, key statically narrowed to S; chain keys add `renderSlotChain` |
| store | `PropsStore` | store factory return type | `useStore` selector hook + `actions.*` (draft-param stripped) |
| business | `I` | inject return type | plain data + callbacks (hooks banned) |
@@ -84,7 +84,7 @@ An inject factory takes what its declarations earn it — `sessionId` for sessio
### Data-boundary discipline
-Hooks are framework-made only: `useSession`, `useSessions`, `useStore`, `renderSlot` are the four seats, implemented once with framework-guaranteed correctness; business code passes plain data and callbacks between parent and child (a component's own behavioral hooks that subscribe to nothing external remain fine). Live data has exactly three channels: what the parent knows travels as owner props at the renderSlot site; what only the component knows is local state; what must be shared across entries or survive remounts is a declared store. Derivation is a pure function over framework-hook data (`useMemo`), never a subscription of its own.
+Hooks are framework-made only: `useSession`, `useSessions`, `useWorkspaces`, `useStore`, `renderSlot` are the five seats, implemented once with framework-guaranteed correctness; business code passes plain data and callbacks between parent and child (a component's own behavioral hooks that subscribe to nothing external remain fine). Live data has exactly three channels: what the parent knows travels as owner props at the renderSlot site; what only the component knows is local state; what must be shared across entries or survive remounts is a declared store. Derivation is a pure function over framework-hook data (`useMemo`), never a subscription of its own.
### Tree context and the renderer seam
diff --git a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.zh.md b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.zh.md
index 4c55171ca0..52edea30ac 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.zh.md
@@ -42,7 +42,7 @@ ctx.slots.register({
| 份额 | 类型 | 真源 | 内容 |
|---|---|---|---|
-| 运行时 | `PropsRuntime` | K 对应的 SlotMap entry | `OwnerOf`(渲染现场传参)+ session scope 标配 `useSession`/`sessionId` + 全局 `useSessions` |
+| 运行时 | `PropsRuntime` | K 对应的 SlotMap entry | `OwnerOf`(渲染现场传参)+ session scope 标配 `useSession`/`sessionId` + 全局 `useSessions`/`useWorkspaces` |
| 子坑渲染 | `PropsRenderSlots` | register 的 `children` 键集 | `renderSlot(key, owner)`,键参静态收窄到 S;chain 键另有 `renderSlotChain` |
| store | `PropsStore` | store 工厂的返回类型 | `useStore` selector hook + `actions.*`(剥去 draft 形参) |
| 业务 | `I` | inject 的返回类型 | 普通数据+回调(禁 hook) |
@@ -84,7 +84,7 @@ inject 工厂只收其声明挣来的形参——session 坑得 `sessionId`,
### 数据界线纪律
-hook 只许框架造:`useSession`、`useSessions`、`useStore`、`renderSlot` 是仅有的四席,各实现一次、正确性由框架担保;业务代码在父子组件之间只传普通数据与回调(组件自用、不订阅任何外部数据源的行为 hook 不在此限)。活数据恰有三条通道:父知道的,作为 owner props 在 renderSlot 现场传入;只有组件自己知道的,是本地 state;需要跨 entry 共享或跨重挂载存活的,是声明的 store。派生是对框架 hook 数据做纯函数(`useMemo`),绝不自成一路订阅。
+hook 只许框架造:`useSession`、`useSessions`、`useWorkspaces`、`useStore`、`renderSlot` 是仅有的五席,各实现一次、正确性由框架担保;业务代码在父子组件之间只传普通数据与回调(组件自用、不订阅任何外部数据源的行为 hook 不在此限)。活数据恰有三条通道:父知道的,作为 owner props 在 renderSlot 现场传入;只有组件自己知道的,是本地 state;需要跨 entry 共享或跨重挂载存活的,是声明的 store。派生是对框架 hook 数据做纯函数(`useMemo`),绝不自成一路订阅。
### 树上语境与渲染器安装缝
diff --git a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.i18n.yaml
index d218c2a0b2..0b599bc639 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.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
-2026-07-22-tui-interactive-extension-service.md: 82e7c751b6e5b7500f9f7d7004fda8b905dccabb
-2026-07-22-tui-interactive-extension-service.zh.md: d7340e3f5dcf45e95b2d6e15ce3fc33726a555ae
+2026-07-22-tui-interactive-extension-service.md: 86cb39748358882d26766467d08f4f43510c1cc2
+2026-07-22-tui-interactive-extension-service.zh.md: d53f526a07b20fcff7086a1f501558d23e7eea8a
diff --git a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.md b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.md
index 82e7c751b6..86cb397483 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.md
+++ b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.md
@@ -28,7 +28,7 @@ Manager tests pin FIFO admission, cancellation, repeated close, shutdown outcome
**Expose pi-tui objects directly.** This gives plugins maximum freedom but makes private focus, rendering, and teardown state a public compatibility contract. It also cannot arbitrate independently loaded overlays.
-**Put interactive callbacks on command definitions.** Commands are shared by TUI and ACP and remain useful without a terminal. Adding terminal state to `ctx.commands` would couple discovery and dispatch to one presentation implementation.
+**Put interactive callbacks on command definitions.** Commands remain transport-neutral domain entries even though TUI is their only shipped consumer. Adding terminal state to `ctx.commands` would couple discovery and dispatch to one presentation implementation.
**Create a complete TUI slot and action framework at once.** Actions, editor replacement, transcript renderers, status regions, and completion providers have different composition and conflict rules. Shipping them behind one broad API would freeze those rules before a concrete consumer proves them.
diff --git a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.zh.md b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.zh.md
index d7340e3f5d..d53f526a07 100644
--- a/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-22-tui-interactive-extension-service.zh.md
@@ -28,7 +28,7 @@ Cordis 插件可以通过 `ctx.commands` 注册用户命令,但需要终端交
**直接暴露 pi-tui 对象。** 这会赋予插件最大的自由度,却会把私有的焦点、渲染与拆卸状态变成公开兼容性契约,也无法在独立加载的浮层之间进行仲裁。
-**在命令定义中加入交互回调。** 命令由 TUI 与 ACP 共享,即使没有终端也仍然有用。向 `ctx.commands` 添加终端状态,会让发现与分派流程耦合到某一种呈现实现。
+**在命令定义中加入交互回调。** 命令仍是传输无关的领域条目,尽管 TUI 是唯一已交付的消费方。向 `ctx.commands` 添加终端状态,会让发现与分派流程耦合到某一种呈现实现。
**一次性建立完整的 TUI slot 与 action 框架。** action、编辑器替换、transcript 渲染器、状态区域和补全提供方具有不同的组合规则与冲突规则。在具体消费方验证这些规则之前就将其纳入一个宽泛 API,会过早固化这些规则。
diff --git a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.i18n.yaml
index 54df5f07ab..87ce1f1bac 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.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
-2026-07-23-client-plugin-loading-model.md: 58651fd258a6b2929c58bb6f93b44adb6e8e1818
-2026-07-23-client-plugin-loading-model.zh.md: f60b06c7bfaa9c70170082ac4384ba2bd899676e
+2026-07-23-client-plugin-loading-model.md: 3513e026785fc366455bb32bf788a3a098275bb1
+2026-07-23-client-plugin-loading-model.zh.md: f31a1b076a5d93db44c67463730b38283d44ff7f
diff --git a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md
index 58651fd258..3513e02678 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md
+++ b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md
@@ -56,11 +56,11 @@ What happens between `dsh web` starting and the UI appearing? Three stages: the
**Host side — compose the graph.**
-1. The composing app (`apps/cli`) mounts the roster as in-memory Loader entries via `mountWebPlugins`. The roster is one flat list of the plugin packages, plus the `client-hmr` row under `--dev`. A roster package that fails to import throws loud at mount.
-2. The registry (`createHostWebPluginRegistry`) scans the mounted entries' package.json `dshClient` declarations and composes `window.__DSH_BOOT__`: `{ rev, entries: [{ id, url, rev, inject?, immediately? }] }`. The `inject` edges and the `immediately` mark come from manifests, never hand-copied. It refuses a declared plugin without a built `./client` bundle, and any malformed declaration field — load-time fail loud.
-3. The registry rescans on cordis `internal/plugin`, microtask-debounced; a rescan failure keeps serving the previous graph. Each bundle's content is hashed into its `rev` (cache busting + HMR diff anchor), and the row set into `graph.rev`. Every row is fetch-served: `/plugins//client.js?rev=…`. The graph types are a wire contract dual-held on both sides, because the webserver keeps zero workspace dependencies.
+1. The composing app (`apps/cli`) ships the roster as ordinary rows in its `cordis.yml` config tree — client plugin packages are entry rows like every host plugin, and `--dev` appends the `client-hmr` row in code (`AppCLIEntry`) before the settle/sweep so the fail-loud triple covers it. A roster row that fails to import is caught by the boot's `assertEntriesLoaded`.
+2. The `dsh-client-modules` node half (the package is dual-face: its browser half is the module table) scans loader entries' package.json `dshClient` declarations and composes `window.__DSH_BOOT__`: `{ rev, entries: [{ id, url, rev, inject?, immediately? }] }`. The `inject` edges and the `immediately` mark come from manifests, never hand-copied. It refuses a declared plugin without a built `./client` bundle, and any malformed declaration field — activation-time fail loud (a FAILED fiber the sweep reports).
+3. Scanning is incremental per package — there is no full-rescan code path. Each cordis `internal/plugin` emission marks the fiber's entry name dirty (entry-less fibers drop O(1)); a microtask flush reconciles each dirty name against live loader entries, with package metadata (including the negative "not a client package" verdict) cached per name forever and bundle re-hashing reachable only through `rebuilt(id)`. The activation pass seeds the same dirty set from current entries and flushes synchronously, so first scan and steady state share one implementation. Each bundle's content hash is its `rev` (cache busting + HMR diff anchor), the row set hashes into `graph.rev`, and every row is fetch-served: `/plugins//client.js?rev=…`. The graph types are single-sourced in the modules package's `./impl` export — the webserver knows nothing about the graph (it is a plain route-registration plugin; modules registers the bundle route and taps the index render itself).
-Why is the roster a hand-written list and not a scan? Because which plugins compose into a deployment is a composition decision, not a package property — a dshClient package existing in the repo does not mean this deployment mounts it, so discovery-by-scan cannot make that call. The roster lives in `apps/cli/web.ts` rather than cordis.yml only because `dsh web`'s host is a hand-assembled `bootHost` with no Loader config tree yet.
+Why is the roster yml rows and not a scan? Because which plugins compose into a deployment is a composition decision, not a package property — a dshClient package existing in the repo does not mean this deployment mounts it, so discovery-by-scan cannot make that call; the node half scans only what the tree actually mounted.
**Phase one — the module face.** The shell builds the module system over the graph, then prefetches every `immediately` row in parallel. Prefetch is fetch + execute, which registers factories only. A single row's prefetch failure is swallowed here: phase two's import retries the fetch and owns the loud failure, so one bad row cannot mask the others. `immediately` is a prefetch mark — not a barrier, not an identity. The package declares it, the registry carries it into the row. The infrastructure plugins (connection, runtime, ui-theme, i18n, plus hmr) declare it; UI plugins simply arrive on demand.
@@ -74,9 +74,9 @@ Why is the roster a hand-written list and not a scan? Because which plugins comp
### Hot reload: one driver plugin, self-watched bundles
-Whether hot reload is active is a composition decision: dev graphs include the `client-hmr` row (a normal plugin package) and turn on bundle watching; prod graphs do neither.
+Whether hot reload is active is a composition decision: dev compositions mount the `client-hmr` row (a normal plugin package, appended by `--dev`) whose node half brings the bundle watch and the SSE channel; prod compositions mount nothing and have neither.
-How does a rebuilt bundle become a reload signal? The webserver observes it itself — no builder tells it. The registry scan already holds every plugin's bundle path (`clientPath`), so in dev mode the registry stat-polls each scanned bundle file with `fs.watchFile`. Polling is by design: inotify does not fire on the weka network mount, the same reason the build-side watcher needs `--poll`. On a mtime/size change the registry re-hashes that row (`rebuilt(id)`); when the `rev` actually changed, it broadcasts a `rebuilt` frame on `GET /plugins/events` — a system SSE channel that sends the full graph on connect and `rebuilt` frames on change, presentation-only wire that never enters the session log. Watch set membership follows the table: rescans add watches for new rows and drop them for vanished ones, dispose drops all. The poll interval is a validated config field (default 500ms), not a constant. Rebuilding the bundles is any tsdown watch process's business — `scripts/dev-web.ts` remains as the watch-build entry point, its package list dshClient-discovered by scanning `packages/*/*/package.json` at startup — and builder and host share zero protocol. A torn read of a half-written bundle self-heals: the stats keep changing while the write completes, so the next poll tick re-hashes again and broadcasts the final rev.
+How does a rebuilt bundle become a reload signal? The hmr node half observes it itself — no builder tells it. It reads bundle paths from `ctx.clientModuleHost.clientPath(id)`, and one HMR-owned interval stat-polls every current graph row. Adding a row is ordered as synchronous stat baseline, then immediate `clientModuleHost.rebuilt(id)`: a write after the module host's graph hash but before that baseline is caught by the immediate re-hash, while a write after the baseline leaves a stat delta for the next poll. This avoids `fs.watchFile`, whose asynchronous first baseline can silently absorb a construction-time rebuild. Watch membership follows `onGraphChanged`; vanished rows drop out, and a bundle missing at poll time keeps its row dirty so reappearance forces a re-hash even with identical metadata. On a mtime/size delta or dirty row, `clientModuleHost.rebuilt(id)` is the single re-hash entry point; when the `rev` actually changed, the node half broadcasts a `rebuilt` frame on `GET /plugins/events` — a system SSE channel that sends the full graph on connect and `rebuilt` frames on change, presentation-only wire that never enters the session log. Polling is deliberate because inotify does not fire on the weka network mount, the same reason the build-side watcher needs `--poll`; the interval is a validated config field (default 500ms), and disposal clears the one timer. Rebuilding bundles is any tsdown watch process's business — `scripts/dev-web.ts` remains the watch-build entry point, its package list dshClient-discovered by scanning `packages/*/*/package.json` at startup — and builder and host share zero protocol. A torn read self-heals: stats keep changing while the write completes, so the next poll re-hashes and broadcasts the final rev.
On the browser side, the driver reloads one plugin per frame, serialized:
@@ -116,7 +116,7 @@ One governance implementation runs on both sides of the wire; the browser-specif
Costs accepted: the vendored Loader carries idle machinery in the browser (EntryTree persistence is a no-op, groups/isolation unused); every plugin edit in dev pays a bundle rebuild plus fiber remount; graph `inject` rows are informational — activation truth is service-level — so a mismatch surfaces at the settled sweep, not at graph validation; and the three not-yet-promoted libraries keep their static-import export surface until their DI conversions land.
-Roster endgame: when `dsh web` moves to config-tree boot, the roster lands in cordis.yml — client plugin packages become ordinary config-tree entry rows, `mountWebPlugins` and the `CLIENT_PACKAGES` constant disappear, and recomposing a deployment means swapping the yml/overlay. The registry needs zero changes for that move, since its `internal/plugin` subscription already discovers whatever entries the tree mounts.
+Roster endgame (landed 2026-07-25 with the config-tree boot move): the roster lives in `apps/cli/cordis.yml`, `mountWebPlugins` and the `CLIENT_PACKAGES` constant are gone, and recomposing a deployment means swapping the yml/overlay. The graph composer moved from a webserver-side registry into the `dsh-client-modules` node half (the package upgraded to dual-face per this note's promotion rule — its consumer now reaches it through cordis DI), and the transport split landed alongside: the webserver became a plain route-registration plugin, `/api/*` binding moved to the connection node half over the upgraded `api-gateway` plugin (`dsh-host-apiproxy` providing `ctx.apiProxy`), and the dev bundle watch + SSE channel moved to the hmr node half.
## Alternatives considered
diff --git a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.zh.md b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.zh.md
index f60b06c7bf..f31a1b076a 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.zh.md
@@ -56,11 +56,11 @@ vendored Loader 经其 `internal` seam 消费模块系统——唯一调用点
**host 侧——组合这张图。**
-1. 负责组合的 app(`apps/cli`)经 `mountWebPlugins` 把名册挂载为内存中的 Loader entry。名册是插件包的一张平铺清单,`--dev` 下外加 `client-hmr` 行。名册里 import 失败的包在挂载时大声抛错。
-2. 注册表(`createHostWebPluginRegistry`)扫描已挂载 entry 的 package.json `dshClient` 声明,组合出 `window.__DSH_BOOT__`:`{ rev, entries: [{ id, url, rev, inject?, immediately? }] }`。`inject` 边与 `immediately` 标记都来自 manifest,永不人肉抄写。它拒绝声明了插件却没有已构建 `./client` bundle 的包,也拒绝任何畸形的声明字段——装载期大声失败。
-3. 注册表在 cordis `internal/plugin` 上重扫,微任务去抖;重扫失败则继续供给上一张图。每个 bundle 的内容哈希进其 `rev`(缓存失效 + HMR diff 锚点),行集合哈希进 `graph.rev`。每一行都经 fetch 供给:`/plugins//client.js?rev=…`。图的类型是两侧各持一份的 wire 契约,因为 webserver 保持零 workspace 依赖。
+1. 负责组合的 app(`apps/cli`)把名册作为普通行放进它的 `cordis.yml` 配置树——client 插件包与每个 host 插件一样是 entry 行,`--dev` 由代码(`AppCLIEntry`)在 settle/sweep 之前追加 `client-hmr` 行,使 fail-loud 三件套一并覆盖它。名册行 import 失败由 boot 的 `assertEntriesLoaded` 捕获。
+2. `dsh-client-modules` 的 node 半(该包是双面的:浏览器半就是模块表)扫描 loader entry 的 package.json `dshClient` 声明,组合出 `window.__DSH_BOOT__`:`{ rev, entries: [{ id, url, rev, inject?, immediately? }] }`。`inject` 边与 `immediately` 标记都来自 manifest,永不人肉抄写。它拒绝声明了插件却没有已构建 `./client` bundle 的包,也拒绝任何畸形的声明字段——激活期大声失败(FAILED fiber,由 sweep 上报)。
+3. 扫描是单包增量——不存在全量重扫代码路径。每次 cordis `internal/plugin` 发射把该 fiber 的 entry 名标脏(无 entry 的 fiber O(1) 丢弃);微任务 flush 把每个脏名对账 live loader entries,包元数据(含「非 client 包」的否定结论)按名永久缓存,bundle 重哈希只经 `rebuilt(id)` 可达。激活趟从当前 entries 灌同一脏集合并同步 flush,初扫与稳态共享一条实现。每个 bundle 的内容哈希是其 `rev`(缓存失效 + HMR diff 锚点),行集合哈希进 `graph.rev`,每一行都经 fetch 供给:`/plugins//client.js?rev=…`。图类型单源在 modules 包的 `./client` 出口——webserver 对图一无所知(它是朴素路由注册插件;bundle 路由和 index 渲染 tap 都由 modules 自己注册)。
-为什么名册是手写清单而不是扫描?因为哪些插件组合进一次部署是组合决策,不是包属性——一个 dshClient 包存在于仓库里,不代表这次部署要挂载它,扫描发现无从替人做这个决定。名册住在 `apps/cli/web.ts` 而非 cordis.yml,只是因为 `dsh web` 的 host 还是一个手工装配的 `bootHost`,没有 Loader 配置树。
+为什么名册是 yml 行而不是扫描?因为哪些插件组合进一次部署是组合决策,不是包属性——一个 dshClient 包存在于仓库里,不代表这次部署要挂载它,扫描发现无从替人做这个决定;node 半只扫描配置树实际挂载了的东西。
**第一层——模块面。**壳在图之上建起模块系统,然后并行预取每个 `immediately` 行。预取即 fetch + 执行,只登记工厂。单行预取失败在这里被吞下:第二层 import 时会重试 fetch 并拥有那次大声失败,因此一个坏行藏不住其他行。`immediately` 是预取标记——不是屏障,不是身份。包声明它,注册表把它带进图行。基础设施插件(connection、runtime、ui-theme、i18n,外加 hmr)声明它;UI 插件则径直按需到达。
@@ -74,9 +74,9 @@ vendored Loader 经其 `internal` seam 消费模块系统——唯一调用点
### 热重载:一个驱动插件,自行监视的 bundle
-热重载是否启用是一项组合决策:dev 图包含 `client-hmr` 行(一个常规的插件包)并开启 bundle 监视;prod 图两者皆无。
+热重载是否启用是一项组合决策:dev 组合挂载 `client-hmr` 行(一个常规的插件包,由 `--dev` 追加),其 node 半带来 bundle 监视与 SSE 通道;prod 组合不挂载,两者皆无。
-重建好的 bundle 怎么变成重载信号?webserver 自己观察——没有构建器来通知它。注册表扫描本就握有每个插件的 bundle 路径(`clientPath`),因此 dev 模式下注册表用 `fs.watchFile` 对每个已扫描的 bundle 文件做 stat 轮询。轮询是刻意选择:inotify 在 weka 网络挂载上不触发,构建侧监视器需要 `--poll` 也是同一原因。mtime/size 一变,注册表就重哈希该行(`rebuilt(id)`);当 `rev` 真的变了,才在 `GET /plugins/events` 上广播 `rebuilt` 帧——这是一条系统级 SSE(Server-Sent Events)通道,连接即发全量图,变更时发 `rebuilt` 帧,仅供呈现的 wire,永不进会话日志。监视集合的成员随表走:重扫为新行添加监视、为消失的行撤下监视,dispose(资源释放)撤掉全部。轮询间隔是一个经校验的配置字段(默认 500ms),不是常量。重建 bundle 则是任意一个 tsdown watch 进程的事——`scripts/dev-web.ts` 仍作为 watch 构建入口保留,其包清单在启动时扫描 `packages/*/*/package.json` 按 dshClient 发现——构建器与 host 共享零协议。写一半的 bundle 被撕裂读取会自愈:写入完成期间 stat 持续变化,下一个轮询节拍会再次重哈希并广播最终的 rev。
+重建好的 bundle 怎么变成重载信号?hmr 的 node 半自己观察——没有构建器来通知它。它从 `ctx.clientModuleHost.clientPath(id)` 读取图上各行的 bundle 路径,由 HMR 自持的单个定时器对当前图上的每一行做 stat 轮询。新增图行时,顺序固定为先同步取得 stat 基线,再立即调用 `clientModuleHost.rebuilt(id)`:在模块 host 算出图哈希之后、取得基线之前发生的写入会被这次立即重哈希捕获;取得基线之后发生的写入则会留下 stat 差异,供下一次轮询捕获。这避开了 `fs.watchFile`:它以异步首次 stat 建立基线,可能把构造期间的重建静默吸收进基线。监视集合的成员随 `onGraphChanged` 更新;消失的行撤下监视,轮询时缺失的 bundle 则让对应行保持标脏状态,文件重现时即使元数据相同也强制重哈希。mtime/size 变化或行处于标脏状态时,`clientModuleHost.rebuilt(id)` 是重哈希的唯一入口;当 `rev` 真的变了,node 半才在 `GET /plugins/events` 上广播 `rebuilt` 帧——这是一条系统级 SSE(Server-Sent Events)通道,连接即发全量图,变更时发 `rebuilt` 帧,仅供呈现的 wire,永不进会话日志。轮询是刻意选择:inotify 在 weka 网络挂载上不触发,构建侧监视器需要 `--poll` 也是同一原因;轮询间隔是一个经校验的配置字段(默认 500ms),dispose(资源释放)会清掉那一个定时器。重建 bundle 则是任意一个 tsdown watch 进程的事——`scripts/dev-web.ts` 仍作为 watch 构建入口保留,其包清单在启动时扫描 `packages/*/*/package.json` 按 dshClient 发现——构建器与 host 共享零协议。写一半的 bundle 被撕裂读取会自愈:写入完成期间 stat 持续变化,下一个轮询节拍会再次重哈希并广播最终的 rev。
浏览器侧,驱动插件每帧重载一个插件,串行执行:
@@ -116,7 +116,7 @@ wire 两侧跑着同一份治理实现;浏览器特有的表面只是一套模
接受的代价:vendored Loader 在浏览器里背着闲置机件(EntryTree 持久化是 no-op,分组/隔离未用);开发期每次修改插件都要付一次 bundle 重建加 fiber 重挂;图中 `inject` 行仅是信息性说明——激活的真相在服务层——因此不匹配会在 settled 扫描时浮出,而不是在图校验时被拦下;三个尚未升格的库在各自的 DI 转换落地之前保持静态 import 的导出面。
-名册的终局:当 `dsh web` 迁到配置树 boot,名册落进 cordis.yml——client 插件包变成普通的配置树 entry 行,`mountWebPlugins` 与 `CLIENT_PACKAGES` 常量消失,重组一次部署等于换 yml/overlay。注册表为这次迁移零改动,因为它的 `internal/plugin` 订阅本就发现配置树挂载的任何 entry。
+名册的终局(2026-07-25 随配置树 boot 迁移落地):名册住 `apps/cli/cordis.yml`,`mountWebPlugins` 与 `CLIENT_PACKAGES` 常量已消失,重组一次部署等于换 yml/overlay。图的组合器从 webserver 侧的注册表迁进 `dsh-client-modules` 的 node 半(该包按本 note 的升级法则升格为双面——其消费方现经 cordis DI 到达),传输拆分同轮落地:webserver 变为朴素路由注册插件,`/api/*` 绑定迁到 connection 的 node 半、走升格后的 `api-gateway` 插件(`dsh-host-apiproxy` 提供 `ctx.apiProxy`),dev 的 bundle 监视与 SSE 通道迁到 hmr 的 node 半。
## Alternatives considered
diff --git a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.i18n.yaml
index 2a27e6432f..7b83a5dc52 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.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
-2026-07-23-unified-session-query-service.md: 0a466e1c36ff1796c858666b0eb36bbd0f480bb0
-2026-07-23-unified-session-query-service.zh.md: 448122b8e6951058b9f633cd56112b0391e1912e
+2026-07-23-unified-session-query-service.md: 676a42017ca42f9e649f6529f84787e7162faac0
+2026-07-23-unified-session-query-service.zh.md: d4449a415840d61cbb10f88def1062a13e556749
diff --git a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.md b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.md
index 0a466e1c36..676a42017c 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.md
+++ b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.md
@@ -16,6 +16,8 @@ The interface package already owns the shared record, filter, trace, search-requ
`SessionQuerySqlite` extends that service and is the sole concrete backend. One mounted instance therefore exposes every operation through `ctx.sessionQuery`; its inherited exact operations use the shared corpus implementation, while its SQLite-owned lifecycle observes sources, reconciles the derived FTS index, ranks matches, and owns cursor generations. The interface package has no standalone concrete plugin, search-provider registry, or second context key.
+SQLite reconciliation is one quiescent serialized state machine. It passes the caller's exact abort signal into durable snapshot listing and inspection, awaits each started backend operation itself, and checks cancellation after every await and before starting the next source or index operation. Cancellation therefore cannot release the serializer while an ignored or cooperative backend call is still cleaning up, and it cannot start a subsequent listing, inspection, reconciliation, or query after the signal is observed.
+
Backend configuration includes the inherited `readWindowMax` setting alongside its own index path, journal mode, page limits, and snippet limit. First-party apps that need session queries mount the SQLite backend and place its disposable index beside their configured persistence root.
This service topology supersedes the separate-key portion of the [exact query decision](../feature/2026-07-10-session-query-service.md) and [SQLite search decision](../feature/2026-07-10-sqlite-session-query-provider.md); their corpus, query, tokenizer, reconciliation, and safety decisions remain in force.
@@ -32,4 +34,6 @@ Consumers inject one service and can combine exact and full-text operations with
The unified object deliberately retains two internal observation strategies: exact operations read authoritative live/persisted sources per call, while full-text operations reconcile a disposable index. Sharing the context key does not make the derived index authoritative or couple exact-read availability to an FTS query.
+Queued cancellation remains prompt. Cancellation during active asynchronous source observation waits for that started operation to settle, which makes rejection a quiescence boundary and preserves single-file execution for a following search. Synchronous SQLite statements remain non-preemptible and are bracketed by signal checks.
+
Unit coverage pins inherited and abstract behavior on one key, SQLite coverage exercises both operation families on the concrete backend, and the real Loader path verifies that one exported plugin registers the combined service.
diff --git a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.zh.md b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.zh.md
index 448122b8e6..d4449a4158 100644
--- a/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.zh.md
+++ b/.agents/notes/implemented/architecture/2026-07-23-unified-session-query-service.zh.md
@@ -16,6 +16,8 @@ Status: implemented
`SessionQuerySqlite` 扩展该服务,并且是唯一的具体后端。因此,一个挂载实例便可通过 `ctx.sessionQuery` 暴露全部操作;其继承的精确操作使用共享的语料库实现,而由 SQLite 管理的生命周期负责观察数据源、对齐派生 FTS 索引、对匹配项排序并管理游标代际。接口包不提供独立的具体插件、搜索提供方注册表或第二个上下文键。
+SQLite 的对齐过程是一个具备静止性保证的串行状态机。它将调用方的原始中止信号传给持久化快照列表与检查操作,直接等待每个已经启动的后端操作,并在每次等待后以及启动下一个数据源或索引操作前检查是否已取消。因此,即使后端忽略取消或正在配合清理,串行器也不会提前释放;观察到中止信号后,也不会再启动后续的列表、检查、对齐或查询操作。
+
后端配置除了自身的索引路径、日志模式、分页限制与文本片段长度上限外,还包含继承的 `readWindowMax` 设置。需要会话查询的第一方应用挂载 SQLite 后端,并将其可丢弃索引放在已配置的持久化根目录旁。
这一服务拓扑取代了[精确查询决策](../feature/2026-07-10-session-query-service.md)和 [SQLite 搜索决策](../feature/2026-07-10-sqlite-session-query-provider.md)中关于分离上下文键的部分;其中关于语料库、查询、分词器、对齐与安全性的决策仍然有效。
@@ -32,4 +34,6 @@ Status: implemented
统一后的对象有意保留两种内部观察策略:精确操作在每次调用时读取权威的实时源或持久化源,全文操作则使可丢弃索引与数据源对齐。共用上下文键不会让派生索引成为权威来源,也不会使精确读取的可用性依赖 FTS 查询。
+排队阶段的取消仍会及时生效。在异步数据源观察已经开始后取消时,调用方会等待该操作完成清理后才收到拒绝;因此拒绝本身构成静止边界,并保证后续搜索仍按单一串行流程执行。同步 SQLite 语句无法在执行中被抢占,服务会在其前后检查中止信号。
+
单元测试在同一个键上同时固定继承实现与抽象方法的契约,SQLite 测试在具体后端上覆盖两类操作,真实 Loader 路径则验证单个导出的插件能够注册组合后的服务。
diff --git a/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.i18n.yaml
new file mode 100644
index 0000000000..7bdc4d2825
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+# pnpm run verify-translation-pairing --write
+2026-07-24-dsh-commander-argument-adapter.md: c1124f67a2c5d9fbba1e04c896a1021c370befc9
+2026-07-24-dsh-commander-argument-adapter.zh.md: be96c354a7dea53446f3c2e35f0e4967265596f5
diff --git a/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.md b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.md
new file mode 100644
index 0000000000..c1124f67a2
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.md
@@ -0,0 +1,53 @@
+# Agent Note: Parse `dsh` argv through one Commander adapter
+
+Status: implemented
+
+English | [中文](2026-07-24-dsh-commander-argument-adapter.zh.md)
+
+## Problem
+
+The `dsh` CLI entry (`apps/cli`) parsed argv in three hand-rolled idioms that did not compose and gave no `--help`/`--version`. `bin.ts` dispatched by raw inspection — `argv[0] === 'web'`, then `argv.includes('-p') || argv.includes('--prompt')`, else TUI — which is positional-blind: a prompt flag or a config path in the wrong position could misroute the mode, and `argv.includes('-p')` could not tell a real flag from an incidental token. `headless.ts` and `web.ts` each ran their own `node:util` `parseArgs` with inline host/port validation, and `dsh-app-boot` carried `parseResumeArg`, a ~30-line bespoke scanner reimplementing flag/`=`-form/value/repeat handling for `--resume`. Usage was a single hardcoded `usage: dsh -p "task"` line; there was no version flag and no rendered help.
+
+## Decision
+
+Argv is parsed once, in `apps/cli/src/args.ts`, through a Commander adapter (the same parser the SDK bins — `create-sdk`, `dsh-scripts` — already standardize on). `parseDshArgs(argv, version)` returns a discriminated `DshInvocation` union of the three real modes: `{ mode: 'tui', config?, resume? }`, `{ mode: 'headless', prompt }`, or `{ mode: 'web', host?, port?, dev }`. It does **not** model help/version/errors as data: Commander owns those, printing usage or the diagnostic and exiting at the point of failure. `exitOverride()` turns each into a thrown `CommanderError` carrying the intended code (0 for help/version, 1 for a parse or domain error), which one `try/catch` in `parseDshArgs` turns into `process.exit`.
+
+`bin.ts` calls the adapter once and switches on `mode` (closed union, `satisfies never` default), dynamic-importing only the chosen mode's module; only a valid, non-help invocation reaches the switch, so it has no help/version/error cases. Each mode module consumes already-parsed values: `runTui(config, resume)`, `runHeadless(task)`, `runWeb(host, port, dev, workspaceRoot)` — none re-reads argv. It is **one Commander program**: the default surface (no subcommand) carries option-only flags — `--config `, `-p/--prompt `, `--resume ` — and `web` is a real `program.command('web')` subcommand. The default surface takes no positional argument, which is what lets `web` be a real subcommand without a positional collision, so `dsh --help` lists `web` natively (no hand-pasted command text). The default action and the `web` action set the resolved mode, then bail via `command.error(...)` (print + exit 1) on the domain checks Commander cannot express: `--prompt` selects headless and rejects an empty task or a `--config`/`--resume` alongside it rather than silently dropping a TUI input; an empty `--resume=` id fails loud (agent-loop treats `''` as no-resume). Commander parses the default-surface options on either side of the `web` token into `program.opts()`; since `web` shares none of them, the `web` action rejects a leaked `--config`/`-p`/`--resume` (`dsh web -p x`, `dsh --config c.yml web`) rather than silently serving and dropping it. `dsh web`'s `--host`/`--port` are unvalidated pass-through overrides: the adapter assigns no default and does no validation, only `Number`-coercing the port string (the schema wants a number). The `dsh-host-webserver` schemastery `Config` (`host` a `127.0.0.1`/`0.0.0.0` literal union, `port` a natural ≤ 65535) is the single source of both the default (the shipped `apps/cli/cordis.yml` `webserver` row stands when a flag is absent) and validity — `AppCLIEntry` patches an explicit flag straight into that row, so a bad host/port fails loud at the schema on boot, not at parse. `--dev` mounts the client HMR driver and bundle watch, and `--workspace-root ` is a plain pass-through to `AppCLIEntry` (the parent directory for name-created workspaces). A repeated `--resume`, or a following flag captured as a `--resume`/`--prompt` value, is Commander's standard behavior (last-wins / next-token) and is left alone; a bad id fails loud downstream when the session cannot load. `--version` reads this app's `package.json`.
+
+`dsh` takes no positional argument. `--config ` names an alternate cordis tree to boot instead of the shipped default; it exists only so the demo/test call sites (`demo:cordis`, `demo:code-mode`, the keyless PTY smokes) can point the shipped bin at an example tree. A bare `dsh` boots the shipped tree plus the `~/.dsh/config.yaml` personal overlay; a real user never passes `--config`.
+
+CLI parsing lives entirely in `apps/cli`. `dsh-app-boot` holds the boot/env/config/personal-overlay helpers and no argv scanner.
+
+## Session resume through the boot context
+
+`dsh --resume ` is the one way to resume a persisted session, with no environment variable. `runTui` provides the parsed id on the boot context through `boot`'s `prepare(ctx)` hook — `ctx.provide(RESUME_SESSION_ID_KEY, id)` (a `dsh-app-boot` export, value `'resumeSessionId'`) — and the shipped tui-agent/cordis configs read it as a bare identifier: `resumeSessionId: !!js "typeof resumeSessionId === 'string' ? resumeSessionId : undefined"`. The expression is quoted because YAML otherwise parses the `?`/`:` as a mapping; the `typeof` guard tolerates a launcher that never provides the slot. The `/resume` in-place handoff (`process.execve`) rebuilds its re-exec argv from the parsed values as `dsh --resume= [--config ]`.
+
+## One terminal front door: `dsh`
+
+`dsh` is the only terminal entry point; the `dsh-tui-demo` package ships the TUI app bundle plugin the shipped config mounts, and no bin of its own. `demo:cordis`, `demo:code-mode`, and both the tui-agent and cordis-agent keyless PTY smokes launch through `apps/cli/src/bin.ts` with `--config `. `dsh`'s TTY guard (refuse piped stdio before booting, pointing at `dsh -p` for automation) is pinned by `apps/cli/tests/built-bin.e2e.ts`, which runs the built `lib/bin.js` under plain Node with piped stdio (`apps/cli/tests` is in the e2e vitest include). `cli-demo`, `acp-demo`, and `jsonrpc-demo` keep their own bins because each is a distinct surface (headless, ACP, JSON-RPC) `dsh` does not provide.
+
+## Package topology
+
+The argument surface stays inside `apps/cli`, the assembly tier, not a `packages/*` library: it is this one app's routing, not a reusable seam. `dsh-app-boot` shrinks to boot glue with no CLI-parsing responsibility. `commander@^15` is added to `apps/cli/package.json`, matching the SDK bins' pin.
+
+## Alternatives considered
+
+**Keep `node:util` `parseArgs` and only unify the dispatch** — rejected: `parseArgs` has no subcommand model, no rendered help, and no version flag, so `web` routing and `--help`/`--version` would stay hand-rolled. The repo already chose Commander for its other CLIs; a second parser idiom for `dsh` alone is the fragmentation this change removes.
+
+**Keep `parseResumeArg` as a shared helper and feed it Commander's residual args** — rejected: the whole point is to retire the bespoke scanner. Commander parses `--resume` (space and `=` forms, missing-value, position-independence) natively; keeping a parallel hand-written path for the one flag would preserve the duplication the change exists to end.
+
+**A bare `dsh ` positional for the alternate tree** — rejected: a root positional and a real `web` subcommand cannot coexist in one Commander program (the subcommand claims the first positional). A positional would force `web` into a reserved-first-token dispatch to a separate parser and a hand-maintained `web` line in `--help`. Only the demo/test sites ever need to name an alternate tree, so a `--config` flag serves them while leaving the default surface positional-free — `web` is then a normal subcommand in one program with native `--help`.
+
+**Make the argument surface a `packages/*` seam** — rejected: nothing outside `dsh` consumes it, and capability seams are not split preemptively. The Commander adapter is `apps/cli`'s own concern.
+
+**Keep `RESUME_SESSION_ID` as the resume bridge** — rejected: with `--resume` parsed into a value the bin already holds, threading it through an environment variable the config re-reads is indirection with no benefit, and it left the demo bin a second, env-only resume path. Providing the id on the boot context is the same channel `boot`'s `prepare` hook already uses for `tuiResumeHost`.
+
+**Keep the `dsh-tui-demo` bin** — rejected: it duplicated `dsh --config ` exactly, and keeping it forced the demo-only `RESUME_SESSION_ID` fallback to stay alive. Its plugin is what the configs actually mount; only the front-door bin was redundant, and `dsh` is the one terminal entry point.
+
+## Testing
+
+`apps/cli/tests/args.spec.ts` (new; `apps/*/tests` added to the vitest include and `apps/cli/tests` to `tsconfig.host.json`) covers the adapter at the level that matters: mode routing by shape (including `web --dev` and the host/port pass-through), the exit-code behavior for the adapter's fail-loud checks (empty resume/prompt, `--prompt` mixed with a config/`--resume`, unknown option, stray positional), and `--help`/`--version`, captured through a `process.exit` spy. Host/port validity is the webserver schema's job, exercised on boot by the web smoke, not the adapter spec. Both PTY smoke groups in `examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` now drive the real `apps/cli/src/bin.ts`: the `tui-agent` group boots an example tree through `--config`, and the `dsh CLI` group covers default boot, personal overlay, invalid config, the `--resume` config intake, the `process.execve` in-place resume handoff, and the source-path prompt. `examples/cordis-agent/tests/keyless-smoke.e2e.ts` likewise launches through `dsh`. `packages/ui/app-boot/tests/app-boot.spec.ts` drops its `parseResumeArg`/`replaceResumeArg` blocks; the TUI unit and snapshot fixtures use the `dsh --resume {session}` resume command.
+
+## Consequences
+
+`dsh` has rendered `--help`/`--version` and consistent fail-loud parse errors, and mode routing does not depend on flag position. Argv parsing lives in one place with one parser idiom shared with the SDK bins, at the cost of a `commander` dependency on `apps/cli` and Commander's parse semantics (its error strings, its `exitOverride` contract) sitting on the CLI's front door. `dsh-app-boot` owns no CLI-parsing surface; a consumer needing `--resume`-style parsing composes Commander. Session resume rides the boot context rather than an environment variable, and `dsh` is the single terminal front door — the `dsh-tui-demo` package is a plugin bundle a config mounts.
diff --git a/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.zh.md b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.zh.md
new file mode 100644
index 0000000000..be96c354a7
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-dsh-commander-argument-adapter.zh.md
@@ -0,0 +1,53 @@
+# Agent Note: 通过单个 Commander 适配器解析 `dsh` 的 argv
+
+Status: implemented
+
+[English](2026-07-24-dsh-commander-argument-adapter.md) | 中文
+
+## 问题
+
+`dsh` 的 CLI(命令行界面)入口(`apps/cli`)以三种手写方式解析 argv,这些方式无法组合,也不提供 `--help`/`--version`。`bin.ts` 通过原始检查进行分发:先判断 `argv[0] === 'web'`,再判断 `argv.includes('-p') || argv.includes('--prompt')`,否则走 TUI。这种方式对位置不敏感:位置错误的 prompt 标志或配置路径可能把模式路由错,而 `argv.includes('-p')` 无法区分真正的标志和偶然出现的 token。`headless.ts` 和 `web.ts` 各自运行自己的 `node:util` `parseArgs`,并内联校验 host/port,而 `dsh-app-boot` 携带 `parseResumeArg`——一个约 30 行的定制扫描器,为 `--resume` 重新实现了标志、`=` 形式、取值和重复的处理。用法说明只有一行硬编码的 `usage: dsh -p "task"`;既没有版本标志,也没有渲染出的帮助信息。
+
+## 决策
+
+argv 只在 `apps/cli/src/args.ts` 中解析一次,并使用 Commander 适配器(SDK bin `create-sdk`、`dsh-scripts` 已经统一采用的同一解析器)。`parseDshArgs(argv, version)` 返回仅包含三种实际模式的判别式 `DshInvocation` 联合类型:`{ mode: 'tui', config?, resume? }`、`{ mode: 'headless', prompt }` 或 `{ mode: 'web', host?, port?, dev }`。它**不会**将帮助、版本信息或错误建模为数据:这些情况由 Commander 处理,在触发处打印用法或诊断信息并退出。`exitOverride()` 会将每种情况转为抛出的 `CommanderError`,并携带预期退出码(帮助或版本为 0,解析错误或领域错误为 1);唯一一处 `try/catch` 位于 `parseDshArgs` 中,捕获错误后调用 `process.exit`。
+
+`bin.ts` 只调用适配器一次,并对 `mode` 做分支切换(封闭联合类型,默认分支为 `satisfies never`),仅动态导入所选模式对应的模块;只有合法的非帮助请求才会进入这段分支逻辑,因此其中没有帮助、版本或错误分支。每个模式模块只消费已解析好的值:`runTui(config, resume)`、`runHeadless(task)`、`runWeb(host, port, dev, workspaceRoot)`,都不会再次读取 argv。整个 CLI 由**单个 Commander 程序**实现:默认接口(不使用子命令时)只包含选项标志——`--config `、`-p/--prompt `、`--resume `——而 `web` 是通过 `program.command('web')` 定义的真正子命令。默认接口不接受位置参数,因此 `web` 可以成为真正的子命令且不会发生位置参数冲突,`dsh --help` 也会原生列出 `web`,无需手工拼接命令文本。默认命令和 `web` 子命令的处理函数会设置解析得到的模式,随后对 Commander 无法表达的领域校验调用 `command.error(...)` 立即终止(打印信息并以退出码 1 退出):`--prompt` 选择 headless 模式;如果任务为空,或调用中还包含 `--config` 或 `--resume`,它会拒绝调用,而不会静默丢弃 TUI 输入;空的 `--resume=` id 会显式失败(agent-loop 把 `''` 视为不恢复)。Commander 会将 `web` token 前后的默认接口选项都解析进 `program.opts()`;由于 `web` 不与默认接口共用任何选项,`web` 子命令的处理函数会拒绝误入的 `--config`/`-p`/`--resume`(`dsh web -p x`、`dsh --config c.yml web`),而不是静默启动服务并丢弃这些选项。`dsh web` 的 `--host`/`--port` 是未经校验、直接透传的覆盖值:适配器既不设置默认值,也不执行校验,只使用 `Number` 将端口字符串转换为数字(schema 要求该值为数字)。`dsh-host-webserver` 的 schemastery `Config`(`host` 是 `127.0.0.1`/`0.0.0.0` 字面量联合类型,`port` 是不大于 65535 的自然数)是默认值与有效性的唯一真源:未提供标志时,随产品提供的 `apps/cli/cordis.yml` 中 `webserver` 配置项保持原值;`AppCLIEntry` 将显式标志的值直接写入该配置项,因此无效的 host/port 会在启动时触发 schema 校验并显式失败,而不是在参数解析阶段失败。`--dev` 会挂载客户端 HMR(热模块替换)驱动,并启用构建产物监视,`--workspace-root ` 则是直接透传给 `AppCLIEntry` 的选项(按名称创建 workspace 时使用的父目录)。重复提供 `--resume`,或后续标志被捕获为 `--resume` 或 `--prompt` 的值,都是 Commander 的标准行为(最后一次取值生效/将下一 token 作为值),本适配器不作干预;无效 id 会在下游无法加载会话时显式失败。`--version` 读取本应用的 `package.json`。
+
+`dsh` 不接受位置参数。`--config ` 指定一份替代 Cordis 配置树,系统启动该配置树而不是随产品提供的默认配置树;该标志仅用于让演示和测试调用点(`demo:cordis`、`demo:code-mode`、无密钥 PTY 冒烟测试)通过随产品提供的 bin 启动一份示例树。直接运行 `dsh` 会启动随产品提供的配置树,并叠加 `~/.dsh/config.yaml` 个人覆盖;实际用户从不传入 `--config`。
+
+CLI 解析完全位于 `apps/cli` 中。`dsh-app-boot` 提供启动、环境变量、配置和个人覆盖辅助函数,不包含 argv 扫描器。
+
+## 通过启动上下文恢复会话
+
+`dsh --resume ` 是恢复持久化会话的唯一方式,无需环境变量。`runTui` 通过 `boot` 的 `prepare(ctx)` 钩子,在启动上下文中提供已解析的 id:`ctx.provide(RESUME_SESSION_ID_KEY, id)`(`dsh-app-boot` 的一项导出,值为 `'resumeSessionId'`);随产品提供的 tui-agent/cordis 配置将该值作为裸标识符读取:`resumeSessionId: !!js "typeof resumeSessionId === 'string' ? resumeSessionId : undefined"`。这个表达式需要加引号,否则 YAML 会把 `?` 和 `:` 解析为映射;`typeof` 守卫使从未提供该槽位的启动器也能正常运行。`/resume` 原地交接(`process.execve`)根据已解析的值将重新执行时的 argv 构造成 `dsh --resume= [--config ]`。
+
+## 唯一的终端入口:`dsh`
+
+`dsh` 是唯一的终端入口;`dsh-tui-demo` 包(package)提供 TUI 应用组合插件,随产品提供的配置会挂载该插件,而该包不提供自己的 bin。`demo:cordis`、`demo:code-mode` 以及 tui-agent 和 cordis-agent 的无密钥 PTY 冒烟测试都通过 `apps/cli/src/bin.ts` 启动,并传入 `--config `。`dsh` 的 TTY 守卫会在启动前拒绝标准输入输出接入管道的调用,并提示自动化场景使用 `dsh -p`;`apps/cli/tests/built-bin.e2e.ts` 锁定了这一行为:该测试将标准输入输出接入管道,并通过普通 Node 运行构建后的 `lib/bin.js`(e2e Vitest 的 include 包含 `apps/cli/tests`)。`cli-demo`、`acp-demo` 和 `jsonrpc-demo` 保留各自的 bin,因为它们分别提供 `dsh` 所没有的独立接口(headless、ACP(Agent Client Protocol)、JSON-RPC)。
+
+## 包拓扑
+
+参数解析留在 `apps/cli`(组装层)内,而不是 `packages/*` 库中:它是这一个应用自身的路由,而非可复用的 seam。`dsh-app-boot` 收缩为纯粹的 boot 胶水代码,不再承担 CLI 解析职责。`commander@^15` 被加入 `apps/cli/package.json`,与 SDK bin 锁定的版本一致。
+
+## 考虑过的替代方案
+
+**保留 `node:util` `parseArgs`,只统一分发。** 已否决:`parseArgs` 没有子命令模型、没有渲染出的帮助、也没有版本标志,因此 `web` 路由和 `--help`/`--version` 仍将保持手写。本仓库其他 CLI 已经选择了 Commander;单独为 `dsh` 引入第二套解析器方式,正是这次变更要消除的碎片化。
+
+**保留 `parseResumeArg` 作为共享辅助函数,并向它喂入 Commander 的残余参数。** 已否决:整件事的核心就是要退役这个定制扫描器。Commander 原生解析 `--resume`(空格和 `=` 形式、缺值、位置无关性);为这一个标志保留一条平行的手写路径,只会保留这次变更要终结的重复。
+
+**使用裸 `dsh ` 位置参数指定替代配置树。** 已否决:根级位置参数与真正的 `web` 子命令无法在同一个 Commander 程序中共存(子命令会占用第一个位置参数)。位置参数会迫使系统把位于首位的 `web` 作为保留 token 分发给另一个解析器,并手工维护一行 `web` 文本,供 `--help` 显示。只有演示和测试调用点需要指定替代配置树,因此 `--config` 标志既能满足这些调用点,又能让默认接口不包含位置参数;这样,`web` 就能在单个程序中成为普通子命令,并由原生 `--help` 展示。
+
+**把参数解析做成 `packages/*` 的 seam。** 已否决:`dsh` 之外没有任何消费方使用它,而能力 seam 不应被提前拆分。这个 Commander 适配器是 `apps/cli` 自身的事务。
+
+**保留 `RESUME_SESSION_ID` 作为恢复通道**:不予采纳。`--resume` 已被解析成 bin 当前持有的值;若再通过环境变量传递并由配置重新读取,只会引入无益的间接层,还会使演示 bin 保留第二条仅依赖环境变量的恢复路径。在启动上下文中提供 id,与 `boot` 的 `prepare` 钩子为 `tuiResumeHost` 提供值所采用的是同一通道。
+
+**保留 `dsh-tui-demo` bin**:不予采纳。它与 `dsh --config ` 的功能完全重复;保留它还会迫使演示专用的 `RESUME_SESSION_ID` 回退路径继续存在。配置实际挂载的是该包的插件;冗余的只有作为终端入口的 bin,而 `dsh` 是唯一的终端入口。
+
+## 测试
+
+`apps/cli/tests/args.spec.ts`(新增;`apps/*/tests` 加入 vitest include,`apps/cli/tests` 加入 `tsconfig.host.json`)覆盖适配器的关键行为:根据参数形态进行模式路由(包括 `web --dev` 和 host/port 透传),并通过 `process.exit` spy 捕获适配器的显式报错检查(恢复 id 或提示词为空、`--prompt` 与配置或 `--resume` 混用、未知选项、多余的位置参数)以及 `--help`/`--version` 的退出码。host/port 的有效性由 webserver schema 负责,并由 web 冒烟测试在启动时验证,不属于适配器测试的覆盖范围。`examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` 中的两组 PTY 冒烟测试现在都驱动真实的 `apps/cli/src/bin.ts`:`tui-agent` 组通过 `--config` 启动示例树,`dsh CLI` 组覆盖默认启动、个人覆盖、无效配置、配置对 `--resume` 的接收、通过 `process.execve` 原地恢复交接,以及包含源码路径的系统提示词。`examples/cordis-agent/tests/keyless-smoke.e2e.ts` 同样通过 `dsh` 启动。`packages/ui/app-boot/tests/app-boot.spec.ts` 移除其 `parseResumeArg` 和 `replaceResumeArg` 测试块;TUI 单元测试和快照 fixture(测试前置数据)使用 `dsh --resume {session}` 恢复命令。
+
+## 影响
+
+`dsh` 会渲染 `--help`/`--version`,并以一致方式显式报告解析错误;模式路由不依赖标志位置。argv 解析集中在一处,并与 SDK bin 共用一套解析器方式,代价是 `apps/cli` 依赖 `commander`,且 Commander 的解析语义(错误字符串和 `exitOverride` 契约)成为 CLI 入口的一部分。`dsh-app-boot` 不提供任何 CLI 解析接口;需要 `--resume` 式解析的消费方通过组合 Commander 来实现。会话恢复通过启动上下文完成,而不使用环境变量;`dsh` 是唯一的终端入口;`dsh-tui-demo` 包是由配置挂载的插件组合包。
diff --git a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.i18n.yaml
similarity index 63%
rename from .agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.i18n.yaml
rename to .agents/notes/implemented/architecture/2026-07-24-project-session-directories.i18n.yaml
index 3ed957d231..040701d1d6 100644
--- a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.i18n.yaml
+++ b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.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
-2026-07-20-tui-startup-slogans.md: a2a22baafddd08145cec0d03b65ee56b2f8114b1
-2026-07-20-tui-startup-slogans.zh.md: 58fa5790f315845f27b810d62658bd79428b519b
+2026-07-24-project-session-directories.md: 0aa3f513d5a1bb3e44cf33a0ae1eb791ee3a46c2
+2026-07-24-project-session-directories.zh.md: 3d8d33fa9fddad010ab319ac4e1f873b69b4e1dd
diff --git a/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.md b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.md
new file mode 100644
index 0000000000..0aa3f513d5
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.md
@@ -0,0 +1,52 @@
+# Agent Note: Project-grouped session directories
+
+Status: implemented
+
+English | [中文](2026-07-24-project-session-directories.zh.md)
+
+## Problem
+
+A persistence root may be local to one project, shared by several projects, temporary, or centralized. The hashed cwd buckets kept all deployments functional but made a shared root difficult to navigate because a developer could not recognize a project from its directory name.
+
+Each JSONL session also occupied one file directly inside the project bucket. That shape had no ownership directory for additional session artifacts such as metadata, attachments, spill files, or coordination state.
+
+## Decision
+
+The JSONL backend stores sessions under a readable project key and gives every session its own directory:
+
+```text
+/
+ ----/
+ /
+ session.jsonl.zstd
+```
+
+Raw mode uses `session.jsonl`, and sessions without a cwd use `_no-cwd`. Filesystem and drive separators become `-`, unsafe code units use `~XXXX`, and the readable name is bounded to keep the component within filesystem limits.
+
+The project key intentionally has no hash suffix. This follows the common human-readable convention used by coding agents and keeps the normalized project path as the complete directory name. The normalization is lossy: paths such as `/a/b-c` and `/a-b/c`, or long paths with the same retained prefix, share one project directory. Their distinct session ids still select separate session directories; reuse of the same session id remains a storage collision and is rejected.
+
+Case-insensitive filesystems can also make differently cased project keys refer to one physical directory. Identity validation accepts such an alternate spelling only when filesystem canonicalization resolves the discovered and expected paths to the same transcript. A different canonical path remains corruption, so case aliases do not weaken the same-id collision check on case-sensitive stores.
+
+The configured root remains a deployment choice. The layout neither selects a global root nor requires projects to share one. When a deployment does centralize storage, project paths remain recognizable; a project-local root uses the same deterministic structure.
+
+The encoded session id names an ownership directory rather than the transcript itself. `SessionPersistence.locate()` continues to return the fixed transcript path, preserving hook `transcript_path` and `DSH_SESSION_JSONL` semantics. Discovery ignores other entries inside the session directory so the backend can add session-owned artifacts without another layout change.
+
+Lazy materialization remains tied to the transcript: `create()` performs no filesystem I/O, and the first append creates the project/session directories before collision-safe transcript publication. Empty directories are not listed as sessions. The backend rejects flat `/.jsonl*` artifacts with an explicit layout error; the pre-release format provides no automatic data migration.
+
+## Alternatives considered
+
+**Keep opaque cwd hashes.** This preserved short names but defeated the requested navigation by project path when several projects share a persistence root.
+
+**Put session files directly in each project directory.** This matched Claude Code and pi's basic file organization but left no session-level ownership boundary for future artifacts.
+
+**Add a collision-resistant hash suffix.** This distinguishes paths whose normalized forms collide, but makes the directory name more than the normalized project path. The chosen convention accepts lossy project grouping in exchange for the simpler, recognizable name.
+
+**Mandate a centralized root.** Rejected because storage placement belongs to deployment configuration. Project grouping is useful when roots are shared and harmless when they are not.
+
+**Load both flat and directory layouts.** Rejected under the pre-release no-compatibility stance. One accepted layout keeps identity checks and discovery deterministic.
+
+## Consequences
+
+Shared stores can be navigated by recognizable project names, while local and custom roots keep their existing configuration freedom. Every session has a directory available for future backend-owned artifacts, and existing transcript consumers still receive a file path.
+
+Project directory names are longer than the former 12-hex cwd hashes. Very long paths show only a bounded prefix. Moving a project usually selects a different directory, but distinct cwd strings that normalize to the same name share one project directory by design.
diff --git a/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.zh.md b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.zh.md
new file mode 100644
index 0000000000..3d8d33fa9f
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-project-session-directories.zh.md
@@ -0,0 +1,52 @@
+# Agent Note: 按项目分组的会话目录
+
+Status: implemented
+
+[English](2026-07-24-project-session-directories.md) | 中文
+
+## 问题
+
+持久化根目录可以只供一个项目使用,也可以由多个项目共享,还可以是临时目录或集中式目录。对 cwd 进行哈希得到的分桶目录能适用于所有这些部署方式,但开发者无法从目录名辨认项目,因此共享根目录难以浏览。
+
+每个 JSONL 会话也直接以单个文件的形式放在项目分桶目录中。这种布局没有为元数据、附件、溢写文件或协调状态等其他会话产物提供归属目录。
+
+## 决策
+
+JSONL 后端按可读的项目键存储会话,并为每个会话提供独立目录:
+
+```text
+/
+ ----/
+ /
+ session.jsonl.zstd
+```
+
+原始模式使用 `session.jsonl`,没有 cwd 的会话使用 `_no-cwd`。文件系统路径分隔符和驱动器分隔符会转换为 `-`,不安全的代码单元使用 `~XXXX`,可读名称则限制长度,以确保目录项不超过文件系统限制。
+
+项目键有意不带哈希后缀。这遵循 coding agent(编码智能体)常用的易读约定,使规范化后的项目路径本身就是完整的目录名。规范化过程有损:`/a/b-c` 与 `/a-b/c` 等路径,或者保留前缀相同的长路径,会共用同一个项目目录。不同的会话 id 仍会选择不同的会话目录;复用相同的会话 id 仍构成存储冲突,系统会予以拒绝。
+
+在不区分大小写的文件系统上,大小写不同的项目键也可能指向同一个物理目录。只有当文件系统路径规范化将发现路径和预期路径解析为同一个 transcript(文本记录)时,身份验证才接受这种拼写变体。规范化后的路径如果不同,仍视为存储损坏,因此大小写别名不会让区分大小写的存储放宽同一 id 的冲突检查。
+
+根目录由部署配置决定。这种布局既不选择全局根目录,也不要求项目共享根目录。部署选择集中存储时,目录名仍能让项目路径易于辨认;使用项目本地根目录时,也采用同样的确定性结构。
+
+编码后的会话 id 用于命名归属目录,而不是 transcript 文件本身。`SessionPersistence.locate()` 仍返回固定的 transcript 路径,从而保持钩子 `transcript_path` 和 `DSH_SESSION_JSONL` 的语义不变。发现过程会忽略会话目录中的其他条目,因此后端以后添加会话自有产物时无需再次改变布局。
+
+延迟物化仍以 transcript 为界:`create()` 不执行文件系统 I/O,首次追加会先创建项目目录和会话目录,再以无冲突方式发布 transcript。空目录不会被列为会话。后端会显式报告布局错误并拒绝扁平的 `/.jsonl*` 产物;预发布格式不提供自动数据迁移。
+
+## 考虑过的替代方案
+
+**保留不透明的 cwd 哈希。** 这可以保持目录名简短,但当多个项目共享一个持久化根目录时,无法满足按项目路径浏览的需求。
+
+**把会话文件直接放入各项目目录。** 这与 Claude Code 和 pi 的基本文件组织一致,但没有为未来产物提供会话级归属边界。
+
+**添加防冲突的哈希后缀。** 这种方式能区分规范化形式相同的路径,但会使目录名不再只是规范化后的项目路径。所选约定接受有损的项目分组,以换取更简单、易于辨认的名称。
+
+**强制使用集中式根目录。** 不予采纳,因为存储位置属于部署配置。项目分组在根目录共享时有用,在不共享时也没有负面影响。
+
+**同时加载扁平布局和目录布局。** 按照预发布阶段不提供兼容性的原则,不予采纳。只接受一种布局,可以让身份检查和发现过程保持确定性。
+
+## 后果
+
+共享存储可以通过易于辨认的项目名进行浏览,本地根目录和自定义根目录则继续保有现有的配置自由。每个会话都有一个可供后端未来存放自有产物的目录,而现有 transcript 消费方仍会收到文件路径。
+
+项目目录名比原先由 12 个十六进制字符组成的 cwd 哈希更长。路径很长时,目录名只显示长度受限的前缀。移动项目通常会选择不同的目录,但按设计,不同的 cwd 字符串如果规范化成相同名称,就会共用同一个项目目录。
diff --git a/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.i18n.yaml b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.i18n.yaml
new file mode 100644
index 0000000000..b5945b5342
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+# pnpm run verify-translation-pairing --write
+2026-07-24-web-config-tree-boot-and-transport-layering.md: 377ebd2b3cf9ff1dff81dd3546bb262e0ebde88a
+2026-07-24-web-config-tree-boot-and-transport-layering.zh.md: 403e95fb3088d2164d50710d2a69de5526807764
diff --git a/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.md b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.md
new file mode 100644
index 0000000000..377ebd2b3c
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.md
@@ -0,0 +1,42 @@
+# Agent Note: dsh web config-tree boot and the web transport layering
+
+Status: implemented
+
+English | [中文](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md)
+
+> Scope: how `dsh web` composes (cordis.yml + pre-cordis boot classes + config sources) and how the web transport splits across packages (gateway / carrier / binding / graph / dev-reload). The [client plugin loading note](2026-07-23-client-plugin-loading-model.md) owns the browser-side loading chain this composition feeds.
+
+## Problem
+
+`dsh web` was the only hand-assembled surface left: `bootHost` mounted 32 plugins with configs pinned in code (violating no-hardcoded-tunables), the client roster was a `web.ts` constant, and TUI/headless had long been yml compositions. The transport layer misplaced responsibilities to match: the webserver self-described as a dumb carrier yet knew the `__DSH_BOOT__` graph, owned the SSE channel, and hard-coded the `/api/*` prefix; the dev bundle watch lived inside the prod registry behind a `watch?` flag with no lifecycle owner; the graph registry rescanned everything on every `internal/plugin` emission; per-request errors and fatal server errors shared one sink that always exited the process. One user-visible defect rode along: the web path never loaded `$DSH_HOME/.env`, so `DSH_HOME=… dsh web` could not find an API key living there.
+
+## Decision
+
+**Composition is one flat config tree.** `apps/cli/cordis.yml` holds every row — the host runtime (32 rows), the `api-gateway` row, the `webserver` row, and the ten `dshClient` rows (the browser roster; the modules row is simultaneously a host row). No spine bundle: every plugin is one row and every config field is yml-editable. `--dev` appends the `dsh-client-hmr` row in code before the settle sweep — prod and dev differ by exactly that row. Row order carries no load semantics; activation is service-availability driven, and the boot compensates with a fail-loud triple: `assertEntriesLoaded` (import failures), `installFailLoud` (late apply rejections), and an all-ACTIVE sweep (PENDING fibers — cordis inject waiting has no timeout).
+
+**Boot glue is a class pair.** `AppCLIEntry` (apps/cli) and `AppWebEntry` (the shell kernel) hold only what must exist independently of cordis: argv facts, the composed patch set, the parsed boot manifest, the module system instance, loading-page handles — everything else lives in plugins. `AppCLIEntry.run()` is three stages: layered env (ambient > cwd `.env` > `$DSH_HOME/.env`, closing the defect above) → patch composition → Loader include boot plus the triple. `AppWebEntry.run()` mirrors it browser-side: parse `window.__DSH_BOOT__` into a `BootManifest` (two views: npm-package rows for the module table, cordis-plugin rows for entry composition; malformed wire throws), build the module system, render the loading page, prefetch the `immediately` tier in parallel with Context/Loader setup, **await the prefetch before creating entries** (materialization is `tree.import`'s synchronous require, unprotected by fiber inject waiting; cross-package require edges such as i18n → runtime/client need every immediately-tier factory registered first — an empirically found 10–25% boot race otherwise), adopt the modules entry, create the graph rows, settle, sweep.
+
+**Config sources have one declaration place each.** yml static values are engineering defaults; the profile json (`./.dsh-tmp-profile/config.json`, read-only, never created, cwd-anchored until the `$DSH_HOME` migration) is user config mapped through a static `PROFILE_MAPPINGS` table onto target rows (`provider`/`model` → the `api-gateway` row, `persistenceRoot` → the jsonl row); CLI flags map onto the `webserver` row with a field set disjoint from the json's; env values enter through yml `!!js` expressions, never through the mapping table. Patches replace a row's config wholesale, so the entry class re-reads the yml row's static values (bypass parse) and merges overrides on top. An unmapped json key fails loud. The resolved frontend `distIndex` rides the same patch channel — an assembly fact, not user config.
+
+**The transport splits five ways.** `dsh-host-apiproxy` upgraded to the gateway plugin (`api-gateway` row): default-exports `ApiProxyService`, config `{provider, model}`, provides `ctx.apiProxy`, transport-agnostic and registers no routes — `createApiProxy` moved here from the retired runtime package. `dsh-host-webserver` shrank to a plain route-registration plugin: `HttpServerService` provides `ctx.httpServer` (`register(route) → disposer` with duplicate-pattern throw, `tapIndex` transforms applied in registration order, `port`), listens on activation, per-request failures answer 400 and log without exiting, and knows no harness concepts. The connection node half owns the binding: it injects both services and registers `toFetchHandler(ctx.apiProxy)` under the `/api` prefix — future IPC carriers swap connection's transport while the gateway stays untouched. The modules node half (`ClientModuleHostService`, providing `ctx.clientModuleHost`) owns the graph: incremental per-package scanning (no full-rescan code path — `internal/plugin` marks the fiber's entry name dirty, a flush reconciles each name against live entries, package metadata including negative verdicts is cached forever, re-hashing is reachable only through `rebuilt(id)`), the bundle route, the index tap, and `onRebuilt`/`onGraphChanged` notification. The hmr node half owns dev reload: `fs.watchFile` stat-polling driven by `onGraphChanged` membership, and the `/plugins/events` SSE route.
+
+**Package export discipline.** The modules package exposes exactly `.` (node half) and `./client` (the complete browser half: `ClientModuleSystem`, `parseBootManifest`, the adoption plugin face) — no bespoke subpaths; wire types re-export through the root for host-side consumers. The adoption handshake: the kernel writes the constructed instance to `window.__DSH_MODULES__` before cordis exists; the `./client` apply reads the slot (missing = loud throw) and provides `ctx.modules`.
+
+## Consequences
+
+- Recomposing a web deployment is a yml/patch edit; the retired pieces (`mountWebPlugins`, `CLIENT_PACKAGES`, `createHostWebPluginRegistry`, `startWebServer`, the webserver's graph/SSE/api knowledge) are deleted.
+- Headless boots the same composition through the same entry (landed in the stacked follow-up): port 0 is its only surface difference, the model face gains `ask_user_question`/workspace context/model titles per the unification ruling, and `bootHost`/`startHost` retired with the `dsh-host-runtime` package. The profile write path, the `$DSH_HOME` profile relocation, and IPC carriers remain recorded deferrals.
+- A TypeScript pitfall worth remembering: a `declare module 'cordis'` augmentation in a file with **no cordis import** is demoted to a standalone module declaration and silently shatters the program-wide `Context` merge (`ctx.on`/`ctx.effect` vanish across the program). Anchor with `import type {} from 'cordis'`.
+
+## Alternatives considered
+
+| Rejected | One-line reason |
+|---|---|
+| Dedicated `dsh-host-profile` receiver package | The profile json is consumed at patch time; the only runtime consumer of `{provider, model}` is the gateway itself — its config is the receiver |
+| Runtime `assembly` shim plugin providing an `apiHandler` service | Existed only because `createApiProxy` lived in runtime; moving it into apiproxy made the gateway self-hosting, and `toFetchHandler` is a pure function the binding side calls |
+| Full-rescan + incremental scan coexisting | Two implementations, two semantics; the single per-package path covers the activation pass too |
+| A bespoke `./impl` export on the modules package | Non-uniform export surface; the standard `./client` carries the whole browser half |
+| dev overlay / `cordis.dev.yml` | One yml; `!!js` cannot conditionalize row existence, and `--dev` appending one row is the entire difference |
+| env vars in the mapping table | The same field would gain env/json double sourcing and need an invented precedence |
+| Unbarriered create-after-prefetch (`arrive()` dedup as safety) | Disproved by a 10–25% boot race: in-flight dedup covers same-package double-fetch, not cross-package synchronous require edges |
+| json file used directly as loader patches | json keys would couple to yml row structure; profile writers would need cordis knowledge |
diff --git a/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.zh.md b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.zh.md
new file mode 100644
index 0000000000..403e95fb30
--- /dev/null
+++ b/.agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.zh.md
@@ -0,0 +1,42 @@
+# Agent Note:dsh web 的 config-tree boot 与 web 传输分层
+
+Status: implemented
+
+[English](2026-07-24-web-config-tree-boot-and-transport-layering.md) | 中文
+
+> 范围:`dsh web` 如何组合(cordis.yml + cordis 之前的 boot 类 + 配置源),以及 web 传输如何跨包分层(网关 / 载体 / 绑定 / 图 / 开发期重载)。浏览器侧装载链归 [client 插件装载 note](2026-07-23-client-plugin-loading-model.md) 所有,本组合只是它的供给方。
+
+## 问题
+
+`dsh web` 曾是仅剩的手工装配面:`bootHost` 逐个挂 32 个插件、config 钉死在代码里(违反 no-hardcoded-tunables),client roster 是 `web.ts` 常量,而 TUI/headless 早已是 yml 组合。传输层的职责错位与之配套:webserver 自称哑载体却认识 `__DSH_BOOT__` 图、拥有 SSE 通道、硬编码 `/api/*` 前缀;dev 的 bundle watch 寄居在 prod registry 里靠 `watch?` 参数开关、生命周期无主;图 registry 对每次 `internal/plugin` 全量重扫;单请求失败与致命 server 错误共用一个一律退进程的 sink。还有一个用户可见缺陷:web 路径不装 `$DSH_HOME/.env`,`DSH_HOME=… dsh web` 读不到自定义 home 下的 API key。
+
+## 决策
+
+**组合是一棵平铺 config tree。** `apps/cli/cordis.yml` 持有全部行——host runtime(32 行)、`api-gateway` 行、`webserver` 行、十个 `dshClient` 行(浏览器 roster;modules 行同时是 host 行)。不做 spine bundle:每插件一行、每个 config 字段 yml 可改。`--dev` 在 settle sweep 之前由代码追加 `dsh-client-hmr` 行——prod 与 dev 的全部差异就是这一行。行序无装载语义;激活由服务可用性驱动,boot 以 fail-loud 三件套补偿:`assertEntriesLoaded`(import 失败)、`installFailLoud`(迟到的 apply 拒绝)、all-ACTIVE sweep(PENDING fiber——cordis inject 等待没有超时)。
+
+**boot 胶水是一对 class。** `AppCLIEntry`(apps/cli)与 `AppWebEntry`(壳内核)只持有独立于 cordis 必须提前存在的东西:argv 事实、合成的 patch 集、解析出的 boot manifest、模块系统实例、loading 页句柄——其余一律进插件。`AppCLIEntry.run()` 三段:分层 env(ambient > cwd `.env` > `$DSH_HOME/.env`,顺手关掉上述缺陷)→ patch 合成 → Loader include boot 加三件套。`AppWebEntry.run()` 在浏览器侧镜像它:把 `window.__DSH_BOOT__` 解析成 `BootManifest`(双视角:npm 包行给模块表、cordis 插件行给 entry 组合;畸形 wire 大声抛)、建模块系统、渲染 loading 页、immediately 层预取与 Context/Loader 准备并行、**create entry 之前等预取齐**(物化是 `tree.import` 的同步 require,不受 fiber inject 等待保护;i18n → runtime/client 这类跨包 require 边要求 immediately 层工厂全部注册完——否则有实测 10–25% 的 boot 竞态)、收编 modules entry、逐图行 create、settle、sweep。
+
+**每个配置源有唯一声明位置。** yml 静态值是工程默认;profile json(`./.dsh-tmp-profile/config.json`,只读、绝不创建、暂锚 cwd 直至 `$DSH_HOME` 迁移)是用户配置,经静态 `PROFILE_MAPPINGS` 表映射到目标行(`provider`/`model` → `api-gateway` 行,`persistenceRoot` → jsonl 行);CLI flags 映射到 `webserver` 行、字段集与 json 不相交;env 值经 yml `!!js` 表达式进入,绝不进映射表。patch 整体替换行 config,故 entry 类旁路 parse 重读 yml 行静态值再叠加覆盖。未映射的 json 键 fail loud。解析出的前端 `distIndex` 走同一 patch 通道——装配事实,不是用户配置。
+
+**传输五分。** `dsh-host-apiproxy` 升格网关插件(`api-gateway` 行):默认导出 `ApiProxyService`,config `{provider, model}`,provide `ctx.apiProxy`,传输无关、不注册路由——`createApiProxy` 自已退役的 runtime 包迁入。`dsh-host-webserver` 缩成朴素路由注册插件:`HttpServerService` provide `ctx.httpServer`(`register(route) → disposer`、重复 pattern 即抛、`tapIndex` 按注册序应用、`port`),激活即 listen,单请求失败答 400 并记日志不退进程,不认识任何 harness 概念。connection node 半拥有绑定:inject 两个服务,把 `toFetchHandler(ctx.apiProxy)` 注册在 `/api` 前缀下——将来 IPC 载体只换 connection 的传输,网关零改动。modules node 半(`ClientModuleHostService`,provide `ctx.clientModuleHost`)拥有图:单包增量扫描(无全量重扫路径——`internal/plugin` 把 fiber 的 entry 名标脏,flush 逐名对账 live entries,包元数据含否定结论永久缓存,重哈希唯一入口 `rebuilt(id)`)、bundle 路由、index tap、`onRebuilt`/`onGraphChanged` 通知。hmr node 半拥有开发期重载:`fs.watchFile` stat 轮询、watch 集合跟随 `onGraphChanged`、`/plugins/events` SSE 路由。
+
+**包出口纪律。** modules 包只暴露 `.`(node 半)与 `./client`(完整浏览器半:`ClientModuleSystem`、`parseBootManifest`、收编插件面)——不设特设子路径;wire 类型经根出口 re-export 给 host 侧消费方。收编握手:内核在 cordis 之前把建好的实例写入 `window.__DSH_MODULES__`;`./client` 的 apply 读槽(缺槽大声抛)并 provide `ctx.modules`。
+
+## 后果
+
+- 重组一个 web 部署 = 改 yml/patch;退役件(`mountWebPlugins`、`CLIENT_PACKAGES`、`createHostWebPluginRegistry`、`startWebServer`、webserver 的图/SSE/api 知识)全部删除。
+- headless 已在 stacked 后续轮迁入同一组合同一入口:唯一面差异是 port 0,模型面按统一裁决获得 `ask_user_question`/workspace context/模型标题,`bootHost`/`startHost` 随 `dsh-host-runtime` 包退役。profile 写入路径、profile 迁 `$DSH_HOME`、IPC 载体仍为挂账项。
+- 一个值得记住的 TypeScript 坑:`declare module 'cordis'` augmentation 所在文件若**没有任何 cordis import**,会被降级成独立 module declaration,无声打散全程序的 `Context` merge(`ctx.on`/`ctx.effect` 全程序消失)。用 `import type {} from 'cordis'` 锚定。
+
+## Alternatives considered
+
+| 弃案 | 一行理由 |
+|---|---|
+| 专门的 `dsh-host-profile` 受体包 | profile json 在 patch 阶段消费完;`{provider, model}` 的唯一运行时消费方是网关自己——受体即网关 config |
+| runtime 里的 `assembly` 垫层插件(provide `apiHandler`) | 它的存在只因 `createApiProxy` 住 runtime;本体迁入 apiproxy 后网关自持插件身份,且 `toFetchHandler` 是绑定方自己调的纯函数 |
+| 全量重扫与增量扫描并存 | 两条实现两份语义;单包路径足以覆盖激活初扫 |
+| modules 包特设 `./impl` 出口 | 出口面不统一;标准 `./client` 承载完整浏览器半 |
+| dev overlay / `cordis.dev.yml` | 一套 yml;`!!js` 无法条件化行存在性,`--dev` 追加一行就是全部差异 |
+| env 进映射表 | 同一字段将出现 env/json 双源,需再发明优先级 |
+| create 不等预取(以 `arrive()` 去重为安全依据) | 被 10–25% boot 竞态证伪:在途去重只覆盖同包双拉,不覆盖跨包同步 require 边 |
+| json 直接当 loader patches 文件 | json 键名将耦合 yml 行结构,写入方要懂 cordis |
diff --git a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.i18n.yaml
index a813a94c95..dbd259b67a 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.i18n.yaml
+++ b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.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
-2026-07-19-windows-atomic-write-dacl-preservation.md: 013119508da9be426c417797cf7a0ec14e276814
-2026-07-19-windows-atomic-write-dacl-preservation.zh.md: 8ae82884c3b80409d07d3bbcfc8c273e8b227dc8
+2026-07-19-windows-atomic-write-dacl-preservation.md: be9f82174300a7d605c6a6e63878728f08cb37be
+2026-07-19-windows-atomic-write-dacl-preservation.zh.md: fc6ec5232c992f3a230ee0de89439c869b8b46f9
diff --git a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md
index 013119508d..be9f821743 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.md
@@ -6,13 +6,13 @@ English | [中文](2026-07-19-windows-atomic-write-dacl-preservation.zh.md)
## Problem
-On Windows, creating the staging directory and temp file under the target's parent and relying only on inherited DACLs is sufficient for a new file, but not for replacing an existing file whose explicit or protected DACL is narrower than its parent: content is written under the broader parent DACL, and rename carries that staging descriptor onto the replacement.
+Atomic writes protect POSIX staging directories with `0o700` and temp files with `0o600`, but Windows mode bits expose only a synthetic read-only view of the actual DACL. Creating staging under the target's parent and relying on inheritance is sufficient for a new file, but not for replacing an existing file whose explicit or protected DACL is narrower than its parent: content is written under the broader parent DACL, and rename carries that staging descriptor onto the replacement.
## Decision
-`dsh-fs-local` reads an existing target's DACL with `GetFileSecurityW`, applies it to the empty temp file with inheritance protected before writing content, and publishes the closed temp with `ReplaceFileW`. The protected staging descriptor prevents the temp directory's inherited entries from broadening access; `ReplaceFileW` preserves the original target access policy and other replacement metadata. Its ACL merge may reserialize auto-inheritance state or duplicate equivalent ACEs, so self-relative descriptor buffers are not a stable equality contract. New files have no prior descriptor to preserve and continue to inherit the destination directory's DACL.
+`dsh-fs-local` reads an existing target's DACL with `GetFileSecurityW`, applies it to the empty temp file with inheritance protected before writing content, and publishes the closed temp with `ReplaceFileW`. The protected staging descriptor prevents the temp directory's inherited entries from broadening access; `ReplaceFileW` preserves the original target access policy and other replacement metadata. Its ACL merge may reserialize auto-inheritance state or duplicate equivalent ACEs, so self-relative descriptor buffers are not a stable equality contract. New Windows files have no prior descriptor to preserve and continue to inherit the destination directory's DACL; their staging directory therefore lives beside the target. POSIX keeps the owner-only staging modes and preserves an existing target mode.
-Native Windows coverage protects a target DACL, inspects the written staging file, and compares the final replacement's ordered, de-duplicated ACE policy. Host-independent binding tests cover Win32 error translation and every native call boundary.
+Native Windows coverage protects a target DACL, inspects the written staging file, and compares the final replacement's ordered, de-duplicated ACE policy. Host-independent binding tests cover Win32 error translation and every native call boundary. Mode-bit assertions remain POSIX-only; new-file DACL inheritance is an operating-system contract rather than a machine-specific account allowlist.
## Alternatives considered
@@ -22,6 +22,10 @@ Native Windows coverage protects a target DACL, inspects the written staging fil
**Install an owner-only DACL for every write.** Rejected because it would discard deliberate project sharing. Copying the target DACL preserves the deployment's existing access policy instead of inventing one.
+**Assert inherited accounts with `Get-Acl` or `icacls`.** Rejected because such a test verifies machine policy rather than package behavior, and localized well-known account names make the output unstable across hosts.
+
+**Skip the existing `chmod` calls on Windows.** Rejected because Node maps these writable modes to benign no-ops; platform guards add branches without changing DACL behavior.
+
## Consequences
-Replacing a Windows file now requires permission to read the target DACL and set the temp DACL; failure is loud before content is written. The package carries Koffi for the narrow Win32 calls, loaded only on Windows replacement paths. New-file behavior remains directory-inherited, and POSIX mode behavior is unchanged.
+Replacing a Windows file now requires permission to read the target DACL and set the temp DACL; failure is loud before content is written. The package carries Koffi for the narrow Win32 calls, loaded only on Windows replacement paths. A new Windows file inherits broad directory access when the directory is broad by design, while POSIX temp content stays owner-only; a read-only Windows target still fails publication before synthetic mode replay could matter.
diff --git a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.zh.md b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.zh.md
index 8ae82884c3..fc6ec5232c 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.zh.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-19-windows-atomic-write-dacl-preservation.zh.md
@@ -6,13 +6,13 @@ Status: implemented
## 问题
-在 Windows 上,在目标文件的父目录下创建暂存目录和临时文件,并且只依赖继承的 DACL,足以满足新建文件的需要,但无法安全替换显式或受保护 DACL 比父目录更严格的现有文件:内容会在权限更宽松的父目录 DACL 下写入,而重命名又会把这个暂存安全描述符带到替换后的文件上。
+原子写入在 POSIX 上以 `0o700` 保护暂存目录、以 `0o600` 保护临时文件,但 Windows mode 位只呈现实际 DACL 的合成只读视图。在目标文件的父目录下创建暂存目录和临时文件,并且只依赖继承的 DACL,足以满足新建文件的需要,但无法安全替换显式或受保护 DACL 比父目录更严格的现有文件:内容会在权限更宽松的父目录 DACL 下写入,而重命名又会把这个暂存安全描述符带到替换后的文件上。
## 决策
-`dsh-fs-local` 通过 `GetFileSecurityW` 读取现有目标文件的 DACL,在写入内容前将其以禁止继承的形式应用到空临时文件,并通过 `ReplaceFileW` 发布已关闭的临时文件。受保护的暂存安全描述符可防止暂存目录中的继承条目扩大访问权限;`ReplaceFileW` 会保留原目标文件的访问策略及其他替换元数据。其 ACL 合并过程可能重新序列化自动继承状态或复制等价 ACE,因此不能把自相对安全描述符缓冲区的逐字节相等作为稳定契约。新建文件没有既有描述符需要保留,因此仍继承目标目录的 DACL。
+`dsh-fs-local` 通过 `GetFileSecurityW` 读取现有目标文件的 DACL,在写入内容前将其以禁止继承的形式应用到空临时文件,并通过 `ReplaceFileW` 发布已关闭的临时文件。受保护的暂存安全描述符可防止暂存目录中的继承条目扩大访问权限;`ReplaceFileW` 会保留原目标文件的访问策略及其他替换元数据。其 ACL 合并过程可能重新序列化自动继承状态或复制等价 ACE,因此不能把自相对安全描述符缓冲区的逐字节相等作为稳定契约。新的 Windows 文件没有既有描述符需要保留,因此仍继承目标目录的 DACL;其暂存目录也因此位于目标文件旁。POSIX 继续使用仅所有者可访问的暂存 mode,并保留现有目标文件的 mode。
-Windows 原生覆盖率测试会保护目标文件的 DACL、检查写入完成的暂存文件,并对比最终替换文件中保持顺序且去重后的 ACE 策略。与宿主平台无关的绑定测试覆盖 Win32 错误转换以及每个原生调用边界。
+Windows 原生覆盖率测试会保护目标文件的 DACL、检查写入完成的暂存文件,并对比最终替换文件中保持顺序且去重后的 ACE 策略。与宿主平台无关的绑定测试覆盖 Win32 错误转换以及每个原生调用边界。mode 位断言仍仅适用于 POSIX;新文件的 DACL 继承由操作系统契约规定,不应通过特定机器的账户允许列表来断言。
## 备选方案
@@ -22,6 +22,10 @@ Windows 原生覆盖率测试会保护目标文件的 DACL、检查写入完成
**每次写入都设置仅所有者可访问的 DACL。** 不予采用,因为这会破坏项目有意设置的共享权限。复制目标文件的 DACL 可以保留部署中已有的访问策略,无需另行创设策略。
+**使用 `Get-Acl` 或 `icacls` 断言继承账户。** 不予采用,因为这类测试验证的是机器策略,而不是包行为;系统内置账户名会本地化,使输出在不同宿主上不稳定。
+
+**在 Windows 上跳过现有 `chmod` 调用。** 不予采用,因为 Node 会把这些可写 mode 映射为无害的空操作;平台条件判断只会增加分支,不会改变 DACL 行为。
+
## 影响
-替换 Windows 文件现在要求调用方有权读取目标 DACL 并设置临时文件 DACL;如果权限不足,系统会在写入内容前明确失败。该包(package)引入 Koffi 以执行少量 Win32 调用,并且只在 Windows 替换路径上加载。新建文件仍按目录继承,POSIX mode 行为保持不变。
+替换 Windows 文件现在要求调用方有权读取目标 DACL 并设置临时文件 DACL;如果权限不足,系统会在写入内容前明确失败。该包(package)引入 Koffi 以执行少量 Win32 调用,并且只在 Windows 替换路径上加载。新的 Windows 文件会在目录按设计开放较宽访问权限时继承该权限,而 POSIX 临时内容仍仅允许所有者访问;只读 Windows 目标文件仍会在发布时失败,早于重放合成 mode 可能产生影响的时点。
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.i18n.yaml
index 8453907b5c..bf8f19a570 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.i18n.yaml
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.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
-2026-07-20-code-mode-result-card-completeness.md: 03c14cd780832fa03977dade2c7d14feb0399369
-2026-07-20-code-mode-result-card-completeness.zh.md: 45047cc5bcb8b74668702302077ff91fd3ff6bdc
+2026-07-20-code-mode-result-card-completeness.md: 97cd9d722e8252b956e16da03c3b8418451350f3
+2026-07-20-code-mode-result-card-completeness.zh.md: fea162b073e3473f7a07d4bf054408d28851fea9
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.md b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.md
index 03c14cd780..97cd9d722e 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.md
@@ -6,7 +6,7 @@ English | [中文](2026-07-20-code-mode-result-card-completeness.zh.md)
## Problem
-The outer `run_code` tool persisted complete rendered content, but its editor presenter ignored that content and rebuilt the card body from a logs-only `presentationMeta` projection. A result-only run appeared correct because an empty presenter body let ACP and TUI fall back to `tool/result.content`. Once the program emitted a log, the presenter supplied non-empty content, that fallback stopped, and the returned value disappeared from the completed card. A spill policy's final head/tail preview was vulnerable to the same split ownership whenever captured logs made the stale projection non-empty.
+The outer `run_code` tool persisted complete rendered content, but its UI presenter ignored that content and rebuilt the card body from a logs-only `presentationMeta` projection. A result-only run appeared correct because an empty presenter body let consumers fall back to `tool/result.content`. Once the program emitted a log, the presenter supplied non-empty content, that fallback stopped, and the returned value disappeared from the completed card. A spill policy's final head/tail preview was vulnerable to the same split ownership whenever captured logs made the stale projection non-empty.
Nested Code calls never owned cards, so producing metadata for the outer call solely to reconstruct one incomplete card also obscured the intended one-card boundary.
@@ -22,7 +22,7 @@ Nested dispatch remains unchanged. Calls marked by `exec.parent` emit bounded `t
Tool unit coverage drives logs-only, result-only, logs-plus-result, no-output, spilled-result, and failure outcomes through the canonical registry, then pins the durable content and absence of a result presenter. A host-mux regression uses a call-only presenter to prove the result frame carries raw content exactly once and no view. These cases prove stale metadata cannot replace final content without making the host duplicate that content.
-The keyless ACP and TUI Code Mode snapshots execute one outer program that performs two nested bash calls, logs `captured output`, and returns `CODE_ONE+CODE_TWO`. Both surfaces show one completed outer card containing both lines and no nested cards.
+The keyless ACP backend and TUI Code Mode snapshots execute one outer program that performs two nested bash calls, logs `captured output`, and returns `CODE_ONE+CODE_TWO`. The persisted ACP log pins the complete result; the TUI surface shows one completed outer card containing both lines and no nested cards.
## Alternatives considered
@@ -30,10 +30,10 @@ The keyless ACP and TUI Code Mode snapshots execute one outer program that perfo
**Merge presenter metadata with `result.content`.** Rejected because the rendered content already contains the logs; merging would duplicate them and require brittle deduplication.
-**Forward `result.content` through a generic result presenter.** Rejected because the durable event already carries that content and ACP/TUI already have a generic raw-content fallback. The host mux serializes a tool-owned result view beside the event, so forwarding would duplicate the rendered content in one frame merely to recreate the fallback; the default worker alone admits a 64 MiB variable-payload budget before rendering.
+**Forward `result.content` through a generic result presenter.** Rejected because the durable event already carries that content and UI consumers already have a generic raw-content fallback. The host mux serializes a tool-owned result view beside the event, so forwarding would duplicate the rendered content in one frame merely to recreate the fallback; the default worker alone admits a 64 MiB variable-payload budget before rendering.
**Create one card per nested dispatch.** Rejected because intermediate values are intentionally execution-local and never model-facing. Multiple cards would expose an implementation trace instead of the single Code Mode operation the model and user invoked.
## Consequences
-ACP and TUI display the same complete content the model receives and replay persists, including post-policy spill previews, through their generic result fallback. The host API retains the pending program title without duplicating the raw result in a separate view payload. New `run_code` results no longer carry the optional logs metadata, but this requires no session-format bump: existing records remain valid because presentation reads their durable rendered content.
+TUI and JSON-RPC/Web display the same complete content the model receives and replay persists, including post-policy spill previews, through their generic result fallback. The host API retains the pending program title without duplicating the raw result in a separate view payload. New `run_code` results no longer carry the optional logs metadata, but this requires no session-format bump: existing records remain valid because presentation reads their durable rendered content.
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.zh.md b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.zh.md
index 45047cc5bc..fea162b073 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.zh.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-code-mode-result-card-completeness.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-外层 `run_code` 工具会持久化完整的渲染内容,但编辑器的卡片展示逻辑忽略了这些内容,转而根据仅含日志的 `presentationMeta` 投影重新构建卡片正文。仅有结果的运行看似正确,是因为展示逻辑未提供正文时,ACP 和 TUI 会回退到 `tool/result.content`。只要程序输出一条日志,展示逻辑就会提供非空内容,回退随即停止,返回值便会从完成态卡片中消失。当已捕获的日志使陈旧投影变为非空时,输出落盘策略最终生成的头尾预览也会受到同一职责拆分的影响。
+外层 `run_code` 工具会持久化完整的渲染内容,但其 UI 展示器忽略了这些内容,转而根据仅含日志的 `presentationMeta` 投影重新构建卡片正文。仅有结果的运行看似正确,是因为展示器正文为空时,消费方会回退到 `tool/result.content`。只要程序输出一条日志,展示器就会提供非空内容,回退随即停止,返回值便会从完成态卡片中消失。当已捕获的日志使陈旧投影变为非空时,输出落盘策略最终生成的头尾预览也会受到同一职责拆分的影响。
嵌套 Code 调用从不生成自己的卡片。因此,仅仅为了重建这一张不完整卡片而给外层调用生成元数据,还掩盖了每次外层调用只生成一张卡片的预期边界。
@@ -22,7 +22,7 @@ Status: implemented
工具单元测试通过规范注册表覆盖仅有日志、仅有结果、日志与结果并存、无输出、结果落盘和失败的结果,然后固定持久内容以及结果展示器不存在这一事实。宿主 mux 回归测试使用仅有调用的展示器,证明结果帧恰好携带一次原始内容,且不含视图。这些案例证明陈旧元数据无法替换最终内容,同时不会让宿主重复该内容。
-无密钥的 ACP 与 TUI Code Mode 快照会执行一个外层程序:程序进行两次嵌套 bash 调用,记录 `captured output`,并返回 `CODE_ONE+CODE_TWO`。两个界面都只显示一张完成态外层卡片,其中包含这两行内容,且没有嵌套卡片。
+无密钥的 ACP(Agent Client Protocol)后端快照与 TUI Code Mode 快照会执行一个外层程序:程序进行两次嵌套 bash 调用,记录 `captured output`,并返回 `CODE_ONE+CODE_TWO`。ACP 持久化日志固定完整结果;TUI 界面只显示一张完成态外层卡片,其中包含这两行内容,且没有嵌套卡片。
## 备选方案
@@ -30,10 +30,10 @@ Status: implemented
**把展示元数据与 `result.content` 合并:**不予采纳。渲染内容已经包含日志;合并会造成重复,还需要依赖脆弱的去重逻辑。
-**通过通用结果展示器转发 `result.content`:**不予采纳。持久事件已经携带该内容,ACP 和 TUI 也已有通用的原始内容回退机制。宿主 mux 会在事件旁序列化工具拥有的结果视图,因此转发仅仅是为了重建该回退机制,却会在一个帧中重复渲染内容;仅默认 worker 在渲染前允许 64 MiB 的可变载荷预算。
+**通过通用结果展示器转发 `result.content`:**不予采纳。持久事件已经携带该内容,UI 消费方也已有通用的原始内容回退机制。宿主 mux 会在事件旁序列化工具拥有的结果视图,因此转发仅仅是为了重建该回退机制,却会在一个帧中重复渲染内容;仅默认 worker 在渲染前允许 64 MiB 的可变载荷预算。
**为每次嵌套分发创建一张卡片:**不予采纳。中间值有意只存在于执行期间,永远不面向模型。多张卡片会暴露实现轨迹,而不是模型与用户调用的单次 Code Mode 操作。
## 影响
-ACP 和 TUI 通过通用结果回退机制显示与模型接收及回放持久化相同的完整内容,其中包括 post-policy 输出落盘预览。宿主 API 保留待完成的程序标题,同时不在单独的视图负载中重复原始结果。新的 `run_code` 结果不再携带可选的日志元数据,但无需提升会话格式版本:现有记录仍然有效,因为展示逻辑会读取其中持久化的渲染内容。
+TUI 与 JSON-RPC/Web 通过通用结果回退机制显示与模型接收及回放持久化相同的完整内容,其中包括 post-policy 输出落盘预览。宿主 API 保留待完成的程序标题,同时不在单独的视图负载中重复原始结果。新的 `run_code` 结果不再携带可选的日志元数据,但无需提升会话格式版本:现有记录仍然有效,因为展示逻辑会读取其中持久化的渲染内容。
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.i18n.yaml
index f907cf276b..7782ea3360 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.i18n.yaml
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.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
-2026-07-20-jsonl-storage-identity.md: 1ada16791f411a54fbcf9271c7d7963223bbe683
-2026-07-20-jsonl-storage-identity.zh.md: 8027c51dbf6c7d01463b7851d859a40890bf03e1
+2026-07-20-jsonl-storage-identity.md: 1079eb700c819951dbb81e99376c0b71e3e84617
+2026-07-20-jsonl-storage-identity.zh.md: d7ba5c646a7adaaa0ebd60fac7b9c2f030361ff9
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.md b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.md
index 1ada16791f..1079eb700c 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.md
@@ -6,11 +6,11 @@ English | [中文](2026-07-20-jsonl-storage-identity.zh.md)
## Problem
-JSONL lookup selects a physical log from the requested session id across cwd buckets, while the parsed `SessionHeader` supplies the metadata used by later repair and append operations. Without binding those two facts, a log selected for session A can declare session B's id or cwd and redirect a repair or later append to B's path. The bucket scan also needs a defined result when the same encoded id exists in more than one bucket. SQLite does not share this ambiguity because its primary-key query binds metadata and events to the requested id.
+JSONL lookup selects a physical log from the requested session id across project directories, while the parsed `SessionHeader` supplies the metadata used by later repair and append operations. Without binding those two facts, a log selected for session A can declare session B's id or cwd and redirect a repair or later append to B's path. The project scan also needs a defined result when the same encoded id exists in more than one project directory. SQLite does not share this ambiguity because its primary-key query binds metadata and events to the requested id.
## Decision
-`loadStored(id)` is the coordinator's single stored-prefix lookup. The JSONL backend scans every cwd bucket, requires at most one matching encoded filename, parses that file, then validates both `header.id === id` and `selectedPath === logPath(root, header.cwd, header.id)` before returning metadata. `list()` applies the same path validation and rejects duplicate ids across buckets.
+`loadStored(id)` is the coordinator's single stored-prefix lookup. The JSONL backend scans every project directory, requires at most one matching encoded session directory with a transcript, parses that file, then validates `header.id === id` and that the selected path either equals `logPath(root, header.cwd, header.id)` or filesystem canonicalization resolves both spellings to the same transcript. `list()` applies the same path validation and rejects duplicate ids across project directories.
The coordinator independently asserts the returned id and compares the stored cwd with a live session's cwd before repair, state publication, or suffix persistence. It keeps a detached copy of validated metadata; JSONL append and repair derive their path from that copy. The `PersistenceBackend` interface therefore needs neither a scope-specific live lookup nor a storage-locator type.
@@ -18,7 +18,7 @@ An existing configured JSONL root must be a readable directory when the plugin l
## Alternatives considered
-**Flatten storage by session id.** A flat namespace makes duplicate publication collide on one path, but path validation and duplicate rejection close the identity defect without changing the project-grouped cwd layout or its consumers.
+**Flatten storage by session id.** A flat namespace makes duplicate publication collide on one path, but path validation and duplicate rejection close the identity defect without making the check depend on a flat global namespace.
**Carry an opaque storage locator through the coordinator.** A locator binds JSONL mutations directly to a selected path, but JSONL can reproduce that path from metadata it has already validated. Adding another generic and argument to SQLite, test backends, append, and repair makes every implementation carry a concept only the file backend needs.
@@ -26,4 +26,4 @@ An existing configured JSONL root must be a readable directory when the plugin l
## Consequences
-Mismatched, misplaced, and duplicate JSONL logs fail before repair or coordinator state mutation. The cwd-bucket format stays unchanged and needs no migration. Lookup remains proportional to the number of buckets, and one-live-writer ownership remains an explicit limitation. Coordinator and JSONL tests pin rejection before repair, unchanged bytes for both affected logs, path validation during listing, duplicate-id rejection, cwd collision handling, and load-time root validation.
+Mismatched, misplaced, and duplicate JSONL logs fail before repair or coordinator state mutation. Lookup remains proportional to the number of project directories, and one-live-writer ownership remains an explicit limitation. Coordinator and JSONL tests pin rejection before repair, unchanged bytes for both affected logs, path validation during listing, duplicate-id rejection, normalized-project collisions and case aliases, and load-time root validation.
diff --git a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.zh.md b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.zh.md
index 8027c51dbf..d7ba5c646a 100644
--- a/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.zh.md
+++ b/.agents/notes/implemented/bug-fix/2026-07-20-jsonl-storage-identity.zh.md
@@ -6,11 +6,11 @@ Status: implemented
## 问题
-JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物理日志,而解析得到的 `SessionHeader` 会提供后续修复和追加操作使用的元数据。如果这两个事实没有绑定,为会话 A 选中的日志就能声明会话 B 的 id 或 cwd,并将修复或后续追加重定向到 B 的路径。当同一个编码后 id 出现在多个分桶目录中时,分桶扫描也必须给出确定的结果。SQLite 不存在这种歧义,因为主键查询会将元数据和事件绑定到请求的 id。
+JSONL 查找会根据请求的会话 id 在各个项目目录中选出物理日志,而解析得到的 `SessionHeader` 会提供后续修复和追加操作使用的元数据。如果这两个事实没有绑定,为会话 A 选中的日志就能声明会话 B 的 id 或 cwd,并将修复或后续追加重定向到 B 的路径。当同一个编码后 id 出现在多个项目目录中时,项目扫描也必须给出确定的结果。SQLite 不存在这种歧义,因为主键查询会将元数据和事件绑定到请求的 id。
## 决策
-`loadStored(id)` 是协调器唯一的已存前缀查找操作。JSONL 后端扫描所有 cwd 分桶目录,要求匹配编码文件名的日志至多有一个,解析该文件,然后在返回元数据前同时验证 `header.id === id` 和 `selectedPath === logPath(root, header.cwd, header.id)`。`list()` 执行相同的路径验证,并拒绝跨分桶目录重复的 id。
+`loadStored(id)` 是协调器唯一的已存前缀查找操作。JSONL 后端扫描所有项目目录,要求名称与该 id 的编码值匹配且其中包含 transcript(文本记录)的会话目录至多有一个,解析其中的 transcript,然后验证 `header.id === id`,并验证选定路径要么等于 `logPath(root, header.cwd, header.id)`,要么经文件系统路径规范化后,两种写法解析为同一份 transcript。`list()` 执行相同的路径验证,并拒绝跨项目目录重复的 id。
协调器会独立断言返回的 id,并在修复、发布状态或持久化后缀之前比较已存 cwd 和活动会话的 cwd。协调器保留一份已验证元数据的独立副本;JSONL 的追加和修复操作根据该副本派生路径。因此,`PersistenceBackend` 接口既不需要限定范围的活动会话查找,也不需要存储定位器类型。
@@ -18,7 +18,7 @@ JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物
## 考虑过的替代方案
-**按会话 id 扁平化存储。** 扁平命名空间会让重复发布在同一路径上冲突,但路径验证和重复项拒绝无需改变按项目分组的 cwd 布局及其消费方,也能消除身份缺陷。
+**按会话 id 扁平化存储。** 扁平命名空间会让重复发布在同一路径上冲突,但路径验证和重复项拒绝无需让检查依赖扁平的全局命名空间,也能消除身份缺陷。
**通过协调器传递不透明存储定位器。** 定位器可以将 JSONL 变更直接绑定到选定路径,但 JSONL 可以根据已经验证的元数据重新得到该路径。为 SQLite、测试后端、追加和修复操作增加一个泛型和参数,会让每个实现都承担只有文件后端需要的概念。
@@ -26,4 +26,4 @@ JSONL 查找会根据请求的会话 id 在各个 cwd 分桶目录中选出物
## 后果
-JSONL 日志的身份不匹配、位置错误和重复会在修复或协调器状态变更前失败。cwd 分桶格式保持不变,无需迁移。查找开销仍与分桶目录数量成正比,单一活动写入方的所有权仍是明确限制。协调器和 JSONL 测试固定了修复前拒绝、两个受影响日志的字节均保持不变、列出时的路径验证、重复 id 拒绝、cwd 冲突处理以及加载时的根目录验证。
+JSONL 日志的身份不匹配、位置错误和重复会在修复或协调器状态变更前失败。查找开销仍与项目目录数量成正比,单一活动写入方的所有权仍是明确限制。协调器和 JSONL 测试固定了修复前拒绝、两个受影响日志的字节均保持不变、列出时的路径验证、重复 id 拒绝、项目路径规范化冲突与大小写别名,以及加载时的根目录验证。
diff --git a/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.i18n.yaml b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.i18n.yaml
new file mode 100644
index 0000000000..4267e3b83f
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+# pnpm run verify-translation-pairing --write
+2026-07-24-empty-model-response-is-retryable.md: f4a6373178efd5ca1ba5882fb2aaf97dffb2526b
+2026-07-24-empty-model-response-is-retryable.zh.md: 4c3afe44140c029d274f34ade97803b958c6d669
diff --git a/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.md b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.md
new file mode 100644
index 0000000000..f4a6373178
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.md
@@ -0,0 +1,36 @@
+# Agent Note: Empty model completions are retryable EMPTY_RESPONSE failures
+
+Status: implemented
+
+English | [中文](2026-07-24-empty-model-response-is-retryable.zh.md)
+
+## Problem
+
+Providers occasionally return a degenerate completion: a well-formed stream that carries a terminal `stop` finish and zero content blocks — no text, no reasoning, no tool calls. If an adapter maps this shape to a successful `{kind: 'stop'}` finish, the loop logs an empty `assistant/message` and ends the turn as `completed`. Retry never runs, no failure reaches the caller, and a driver such as goal-session consumes a round without progress.
+
+## Decision
+
+An adapter classifies a completed empty response as a provider-boundary failure, and retry policy treats it as transient:
+
+- `dsh-llm` exports the canonical code `EMPTY_RESPONSE_CODE` (`'EMPTY_RESPONSE'`) beside `CONTEXT_WINDOW_EXCEEDED_CODE`/`QUOTA_EXCEEDED_CODE`.
+- `dsh-llm-pi-ai` (`mapStopReason`): a terminal `stop` whose assistant message has no content blocks becomes a `finish {kind: 'error'}` with that code. Context-overflow detection still wins where it applies (it is checked first and is the more actionable classification).
+- `dsh-llm-deepseek` (`translate`): at `[DONE]`, a `stop` (or absent) finish with no opened blocks becomes the same error finish. Reasoning-only streams count as content and stay successful.
+- `dsh-llm-retry` adds `EMPTY_RESPONSE` to `DEFAULT_RETRYABLE_CODES`: the attempt produced nothing durable, so repeating it is safe; deployments can still remove it via `retryableCodes`.
+
+Detection is scoped to `stop` finishes only. `max-tokens` with empty content keeps its existing meaning (pi-ai already normalizes the zero-output overflow case), `tool-calls` cannot be block-empty in practice, and error/aborted finishes already fail.
+
+The classification uses the existing loop machinery — `finishError` → `agent/request-error` → `dsh-llm-retry` — and keeps `agent-loop` provider-neutral. Exhausting the retry budget ends the turn with an explicit `EMPTY_RESPONSE` failure instead of an empty success.
+
+## Alternatives considered
+
+**Detect in the loop or `BlockAssembler`.** One shared implementation, but it moves provider-response judgment into the loop, against "plugins, not loop changes", and the assembler is a pure assembly algorithm. The adapter is where wire facts become harness classification, with the overflow reclassification as exact precedent.
+
+**A stream-transform plugin on the `llm/stream` waterfall.** Provider-neutral and one implementation, but it adds a package plus wiring for what is a boundary fact each adapter can state in a few lines, and default-on behavior would still require touching every bundle.
+
+**Treat whitespace-only or reasoning-only responses as empty too.** Rejected as overreach: those carry model-produced content, and misclassifying a legitimate (if useless) response as a transport-class failure risks retry loops on models that intentionally stop after reasoning. The scope is exactly "zero content blocks".
+
+## Consequences
+
+- A transiently misbehaving provider consumes a bounded retry instead of a turn with no output; a persistently empty model surfaces an actionable `EMPTY_RESPONSE` turn failure.
+- A model that genuinely intends to say nothing (rare, but possible after a tool result) is retried and, if consistently empty, fails the turn. This trade was accepted deliberately: an empty assistant message is indistinguishable from the provider defect and has no value to the user.
+- The `empty-response-retry` ACP snapshot (an authored keyless scenario with a deterministic 1 ms zero-jitter retry overlay, `examples/acp-agent/retry.cordis.yml`) pins the product-visible behavior: a durable `llm/retry` event, no ACP output for the discarded attempt, the recovered reply, and a clean completed turn.
diff --git a/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.zh.md b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.zh.md
new file mode 100644
index 0000000000..4c3afe4414
--- /dev/null
+++ b/.agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.zh.md
@@ -0,0 +1,36 @@
+# Agent Note: Empty model completions are retryable EMPTY_RESPONSE failures
+
+Status: implemented
+
+[English](2026-07-24-empty-model-response-is-retryable.md) | 中文
+
+## Problem
+
+提供方偶尔会返回一种退化的 completion:流本身格式完好,携带一个终止性的 `stop` 结束,却没有任何内容块——没有文本、没有 reasoning(推理)、没有工具调用。如果适配器把这种形态映射为成功的 `{kind: 'stop'}` 结束,主循环就会记录一条空的 `assistant/message`,并把该轮次以 `completed` 结束。系统不会重试,失败也不会向调用方暴露,而像 goal-session 这样的驱动方会消耗一个轮次,却没有取得任何进展。
+
+## Decision
+
+由适配器把「已完成但为空」的响应归类为一次提供方边界失败,重试策略则将其视为瞬时性问题:
+
+- `dsh-llm` 在 `CONTEXT_WINDOW_EXCEEDED_CODE`/`QUOTA_EXCEEDED_CODE` 之外,导出规范代码 `EMPTY_RESPONSE_CODE`(`'EMPTY_RESPONSE'`)。
+- `dsh-llm-pi-ai`(`mapStopReason`):当终止性 `stop` 所对应的 assistant 消息没有内容块时,它会变成一个携带该代码的 `finish {kind: 'error'}`。上下文溢出检测在其适用场景中仍然优先(它先被检查,也是更具可操作性的归类)。
+- `dsh-llm-deepseek`(`translate`):在 `[DONE]` 处,若 `stop`(或缺失)结束且没有打开过任何块,则同样变成该错误结束。仅含 reasoning 的流算作有内容,仍视为成功。
+- `dsh-llm-retry` 把 `EMPTY_RESPONSE` 加入 `DEFAULT_RETRYABLE_CODES`:这次尝试没有产生任何持久内容,因此重复它是安全的;部署方仍可通过 `retryableCodes` 将其移除。
+
+检测仅限于 `stop` 结束。内容为空的 `max-tokens` 保持其既有含义(pi-ai 已经把零输出的溢出场景归一化处理),`tool-calls` 在实践中不可能是空块,而 error/aborted 结束本身已经算失败。
+
+这套归类使用既有的主循环机制——`finishError` → `agent/request-error` → `dsh-llm-retry`——并让 `agent-loop` 保持提供方无关。重试预算耗尽时,该轮次会以显式的 `EMPTY_RESPONSE` 失败结束,而不是在没有内容的情况下成功结束。
+
+## Alternatives considered
+
+**在主循环或 `BlockAssembler` 中检测。** 只需一份共享实现,但这会把对提供方响应的判断挪进主循环,违背「插件优先,而非改动主循环」,且 assembler 是纯粹的组装算法。适配器才是把协议层面的事实转化为 harness 归类的地方,而溢出重归类正是精确的先例。
+
+**在 `llm/stream` waterfall(瀑布式事件)上做一个流转换插件。** 这种做法提供方无关且只需一份实现,但它为「每个适配器几行就能声明的边界事实」额外增加了一个包和相应接线,而且默认开启的行为仍需改动每一个 bundle。
+
+**把仅含空白或仅含 reasoning 的响应也当作空响应。** 作为过度设计予以否决:这类响应携带了模型产生的内容,把一个合法(哪怕无用)的响应误判为传输类失败,会在那些故意在 reasoning 之后停止的模型上引发重试循环。其范围严格限定为「零内容块」。
+
+## Consequences
+
+- 一个偶发异常的提供方会消耗一次有界重试,而不是一个没有输出的轮次;一个持续返回空内容的模型则会暴露为用户可据以行动的 `EMPTY_RESPONSE` 轮次失败。
+- 一个确实打算什么都不说的模型(罕见,但在一次工具结果之后有可能出现)会被重试,若始终为空,则该轮次失败。这个取舍是经过审慎权衡后接受的:一条空的 assistant 消息与提供方缺陷无法区分,且对用户毫无价值。
+- `empty-response-retry` ACP 快照(一个人工编写的无密钥场景,配有确定性的 1 ms 零抖动重试 overlay,`examples/acp-agent/retry.cordis.yml`)钉住了产品可见的行为:持久的 `llm/retry` 事件、被丢弃的尝试不产生任何 ACP 输出、恢复后的回复,以及一次干净的已完成轮次。
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.md b/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.md
deleted file mode 100644
index 61e2fd609b..0000000000
--- a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.md
+++ /dev/null
@@ -1,59 +0,0 @@
-# Agent Note: Agent Client Protocol (ACP) support — drive the coding agent from external editors
-
-Status: implemented
-
-English | [中文](2026-06-14-acp-agent-client-protocol.zh.md)
-
-## Problem
-
-The harness originally exposed agents only through a readline loop. That surface could carry text, but it gave an editor no structured way to create or resume sessions, correlate prompt completion, stream reasoning and tool activity, render tool-specific UI, ask for permission, or cancel one conversation without disturbing another. ACP defines those interactions as JSON-RPC over stdio, and Zed is the target client used to make concrete compatibility decisions.
-
-The bridge must preserve the harness's existing ownership boundaries. It cannot depend on the concrete agent loop, bypass the tool registry, execute shell commands in the editor, or invent a second source of session truth. stdout is also the protocol transport, so any accidental log output corrupts the connection.
-
-## Decision
-
-`@deepseek-ai/dsh-acp` is a UI/client-driver plugin under `packages/ui/acp`. It uses `@agentclientprotocol/sdk`'s `AgentSideConnection` over stdin/stdout and programs only interface services: the agent create/resume factory, session persistence, tool registry, user interaction, and optional approval/bash capabilities. It does not change the agent loop and is not a capability-seam implementation.
-
-The bridge implements the following stable session path:
-
-- `initialize` negotiates the protocol version, advertises text plus `resource_link` prompts, and advertises `loadSession`.
-- `session/new` validates an absolute `cwd`, stores it in `SessionHeader`, creates an agent through `ctx.agents`, and returns any composition-backed config options.
-- `session/load` validates the requested cwd against persisted metadata before constructing an agent, reserves the id across the asynchronous resume, replays user/assistant/tool events as ACP updates, and reports the resumed config-option fold.
-- `session/prompt` accepts text and resource links, rejects unsupported or empty content, allows one in-flight prompt per session, and settles against that prompt's owning `turn/end`. An error turn rejects the RPC; other closed turn reasons map through a total ACP stop-reason codec.
-- `session/cancel` calls the queue-aware agent cancel path and settles only the addressed session's prompt.
-
-Tool-call presentation remains tool-owned. A tool's `presentCall` and `presentResult` return the `generic`, `terminal`, or `diff` render-intent variants; the bridge switches on that union and maps it to ACP. Presenter-less tools receive a generic fallback. Bash terminal cards use Zed's capability-gated `_meta.terminal_info`, `_meta.terminal_output`, and `_meta.terminal_exit` convention; the harness still executes the command through `ctx.bash`, preserving sandbox, environment scrub, ownership, and cwd. Clients without that extension receive ordinary text content. Filesystem tools provide diff cards and file locations without hard-coded tool-name branches in the bridge.
-
-Permission handling is an answerer on the [user-approval seam](2026-07-06-approval-seam.md), not an ask-every-tool policy in ACP. An `approval/request` for a bridge-owned agent with a call id becomes `session/request_permission` on that agent's editor session, with one-shot allow/reject choices. Foreign or call-less requests delegate; a missing or failed answerer remains fail-closed. The plugin that asks—such as a pre-execute policy or bash escalation—owns the decision to ask.
-
-When `ctx.permission` is composed, the bridge exposes one `permission` select from the deployment's preset table. The shipped `workspace-write` and `danger-full-access` presets each bundle a sandbox mode with an approval policy; unmatched effective knobs produce the switch-away-only `custom` state. `session/set_config_option` validates through `PermissionService.set()` and writes both owning knob events. A switch during an open turn appends immediately; an idle switch is overlaid in responses and anchored at the next `agent/prompt-submit`, before request assembly. Until then it is memory-only, so a crash restores the durable fold. ACP session modes are not modeled because config options are the forward protocol surface; `AcpConfig.model` remains connection-wide.
-
-The bridge also provides the ACP-backed `UserInteractionProvider`: `ask_user_question` requests become form elicitations on the owning session. Select, multi-select, option descriptions, and custom-answer override semantics are preserved.
-
-Lifecycle ownership is explicit. The bridge holds an `AgentHandle` per live session. Disconnect and Cordis disposal cancel pending prompts, dispose every handle in parallel, await loop quiescence and persistence flush, and then remove the records. Stream notification failures are contained so a vanished client cannot corrupt an agent turn. The ACP app composition loads no stdout logger; a test guards stdout as framed JSON-RPC only.
-
-The precise supported and deferred protocol rows live in [`packages/ui/acp/acp-feature-support.md`](../../../../packages/ui/acp/acp-feature-support.md); the package README is the operational contract.
-
-## Alternatives considered
-
-**A prepended `tools/execute` listener that asks on every ACP-owned call** — rejected. It would hard-code permission policy into the UI bridge, ask even when no policy requires it, and could not serve approval requests that arise after execution begins. The shared user-approval seam keeps mechanism, asking policy, and UI answerer separate.
-
-**Inject the concrete `agentLoop`** — rejected. Agent creation, resume, idle observation, and disposal are interface-level ownership operations on `dsh-agent`; a UI plugin does not need a dependency-rule exception.
-
-**Execute bash through ACP `terminal/*`** — rejected. That would move execution outside the harness and bypass its sandbox, credential scrub, task ownership, cwd resolution, and session log. Terminal metadata is presentation only.
-
-**Represent permission presets as ACP session modes** — rejected. The deployment-defined preset is already one config-option select, while session modes are the legacy surface slated for removal in ACP v2.
-
-**Hijack stdout defensively** — rejected. Process-wide monkey-patching is outside Cordis effect ownership and races the protocol transport. The app composition owns stdout purity.
-
-## Consequences
-
-Editors can create, load, prompt, cancel, render, ask, and reconfigure multiple harness sessions over one ACP connection without a loop-specific dependency. The session event log remains the durable source for replay, prompt settlement, cwd, and per-session configuration. Tool presentation and human-answer channels remain extensible plugin contracts instead of ACP-specific behavior.
-
-The bridge deliberately does not implement session list/delete/resume/close capabilities, MCP passthrough, additional directories, image/audio/embedded-resource prompts, plans, slash commands, usage updates, editor filesystem delegation, or the ACP terminal execution sub-protocol. Runtime model selection was added later through standard session config options by the [LLM catalog and ACP selection Agent Note](../architecture/2026-07-15-llm-model-catalog-and-acp-selection.md).
-
-An idle config selection is truthful in the live response but not durable until the next `agent/step` anchors it inside the open turn. Crashing before that boundary loses the pending selection; this is the cost of keeping session events turn-enclosed and replay-safe.
-
-## Verification
-
-The ACP suites cover the in-memory protocol codec, create/load replay, exact prompt settlement, cancellation races, unsupported content, tool presentation, terminal capability fallback, permission outcome mapping, config-option validation and persistence, multi-session isolation, disconnect/disposal quiescence, and HMR cleanup. Snapshot and built-bin tests exercise the app composition, while the real-API e2e self-skips without a key.
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.zh.md b/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.zh.md
deleted file mode 100644
index 608f445432..0000000000
--- a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.zh.md
+++ /dev/null
@@ -1,59 +0,0 @@
-# Agent Note: Agent Client Protocol(ACP)支持——从外部编辑器驱动编码 agent
-
-Status: implemented
-
-[English](2026-06-14-acp-agent-client-protocol.md) | 中文
-
-## 问题
-
-harness 最初仅通过 readline 循环暴露 agent。该接口能传输文本,但编辑器无法以结构化方式创建或恢复会话、关联提示词完成、流式输出推理(reasoning)与工具活动、渲染工具专属 UI、请求权限,或在不干扰其他对话的前提下取消某个对话。ACP(Agent Client Protocol)将这些交互定义为基于 stdio 的 JSON-RPC,Zed 是用于做出具体兼容性决策的目标客户端。
-
-桥接层必须保持 harness 既有的所有权边界。它不能依赖具体的 agent loop(智能体循环),不能绕过工具注册表,不能在编辑器中执行 shell 命令,也不能发明第二个会话真源。stdout 同时也是协议传输通道,因此任何意外的日志输出都会破坏连接。
-
-## 决策
-
-`@deepseek-ai/dsh-acp` 是位于 `packages/ui/acp` 的 UI/客户端驱动插件。它使用 `@agentclientprotocol/sdk` 的 `AgentSideConnection`(基于 stdin/stdout),仅编排接口服务:agent 创建/恢复工厂、会话持久化、工具注册表、用户交互,以及可选的审批/bash 能力。它不修改 agent loop,也不是能力 seam 的实现。
-
-桥接层实现以下稳定的会话路径:
-
-- `initialize` 协商协议版本,声明支持 text 与 `resource_link` 类型的提示词,并声明 `loadSession` 能力。
-- `session/new` 校验绝对路径 `cwd`,将其存入 `SessionHeader`,通过 `ctx.agents` 创建 agent,并返回由组合层支持的配置选项。
-- `session/load` 在构造 agent 之前校验请求的 cwd 与持久化元数据是否一致,在异步恢复期间保留 id,将用户/助手/工具事件作为 ACP update 回放,并报告恢复后的 config-option 折叠结果。
-- `session/prompt` 接受文本和 resource link,拒绝不支持的或空的内容,每个会话同时只允许一个 in-flight 提示词,并在该提示词所属的 `turn/end` 时结算。错误轮次拒绝 RPC;其他关闭轮次的原因通过一个全覆盖的 ACP stop-reason 编解码器映射。
-- `session/cancel` 调用队列感知的 agent 取消路径,仅结算被寻址会话的提示词。
-
-工具调用的展示仍由工具自身负责。工具的 `presentCall` 和 `presentResult` 返回 `generic`、`terminal` 或 `diff` 渲染意图变体;桥接层对该联合类型做 switch 并映射到 ACP。没有 presenter 的工具获得通用回退。Bash 终端卡片使用 Zed 的能力门控约定 `_meta.terminal_info`、`_meta.terminal_output` 和 `_meta.terminal_exit`;harness 仍通过 `ctx.bash` 执行命令,保留沙箱、环境清洗、所有权和 cwd。不支持该扩展的客户端收到普通文本内容。文件系统工具提供 diff 卡片和文件位置,桥接层中无需硬编码工具名分支。
-
-权限处理是[用户审批 seam](2026-07-06-approval-seam.md)上的一个 answerer,而非 ACP 中的「每次工具调用都询问」策略。对桥接层所属 agent 且带有 call id 的 `approval/request`,会变为该 agent 编辑器会话上的 `session/request_permission`,提供一次性允许/拒绝选项。外部请求或无 call id 的请求委托给下游;缺失或失败的 answerer 会在故障时保持拒绝。发起询问的插件(如预执行策略或 bash 升级)拥有「是否询问」的决策权。
-
-当 `ctx.permission` 被组合时,桥接层从部署的预设表中暴露一个 `permission` select。已发布的 `workspace-write` 和 `danger-full-access` 预设各自捆绑一个沙箱模式与一条审批策略;无法匹配的有效旋钮组合产生只能切走的 `custom` 状态。`session/set_config_option` 通过 `PermissionService.set()` 校验并写入两个所属旋钮事件。在开放轮次中的切换立即追加;空闲时的切换叠加在响应中,并在下一次 `agent/prompt-submit` 时锚定到开放轮次之前的请求组装阶段。在此之前它仅存于内存,因此崩溃后恢复的是持久化的折叠结果。ACP session mode 不被建模,因为 config option 是面向未来的协议表面;`AcpConfig.model` 保持连接级别。
-
-桥接层还提供基于 ACP 的 `UserInteractionProvider`:`ask_user_question` 请求变为所属会话上的表单引导。select、multi-select、选项描述与自定义回答覆盖语义均被保留。
-
-生命周期所有权是显式的。桥接层为每个活跃会话持有一个 `AgentHandle`。断连和 Cordis dispose(资源释放)会取消待处理的提示词,并行 dispose 所有 handle,等待循环完全停稳与持久化刷写,然后移除记录。流通知失败被隔离,因此消失的客户端不会破坏 agent 轮次。ACP 应用组合不加载 stdout logger;一个测试守卫 stdout 仅包含帧化的 JSON-RPC。
-
-精确的已支持与已推迟的协议行列表见 [`packages/ui/acp/acp-feature-support.md`](../../../../packages/ui/acp/acp-feature-support.md);包 README 是操作契约。
-
-## 曾考虑的替代方案
-
-**在 `tools/execute` 监听器前置一层,对每个 ACP 所属调用都询问权限**:否决。这会将权限策略硬编码到 UI 桥接层,即使没有策略要求也会询问,且无法服务于执行开始后才产生的审批请求。共享的 user-approval seam 将机制、询问策略和 UI answerer 分离。
-
-**注入具体的 `agentLoop`**:否决。agent 的创建、恢复、空闲观察与释放是 `dsh-agent` 上的接口级所有权操作;UI 插件不需要依赖规则例外。
-
-**通过 ACP `terminal/*` 执行 bash**:否决。这会将执行移到 harness 之外,绕过其沙箱、凭证清洗、任务所有权、cwd 解析与会话日志。终端元数据仅用于展示。
-
-**将权限预设表示为 ACP session mode**:否决。部署定义的预设已经是一个 config-option select,而 session mode 是 ACP v2 计划移除的遗留接口。
-
-**防御性劫持 stdout**:否决。进程级 monkey-patching 超出 Cordis 副作用所有权范围,且与协议传输存在竞争。应用组合拥有 stdout 纯净性。
-
-## 后果
-
-编辑器可以通过一条 ACP 连接创建、加载、提交提示词、取消、渲染、询问和重新配置多个 harness 会话,无需依赖特定的循环实现。会话事件日志仍是回放、提示词结算、cwd 与每会话配置的持久真源。工具展示与人工回答通道仍是可扩展的插件契约,而非 ACP 专属行为。
-
-桥接层有意不实现会话列表/删除/恢复/关闭能力、MCP 透传、附加目录、图片/音频/嵌入资源提示词、plan、斜杠命令、用量更新、编辑器文件系统委托或 ACP 终端执行子协议。后续已通过标准会话配置选项加入运行时模型选择,见 [LLM 目录与 ACP 选择 Agent Note](../architecture/2026-07-15-llm-model-catalog-and-acp-selection.md)。
-
-空闲时的配置选择在实时响应中是真实的,但在下一次 `agent/step` 将其锚定到开放轮次之前不具持久性。在该边界之前崩溃会丢失待定选择;这是保持会话事件封闭于轮次内且回放安全的代价。
-
-## 验证
-
-ACP 测试套件覆盖内存协议编解码器、创建/加载回放、精确的提示词结算、取消竞争、不支持的内容、工具展示、终端能力回退、权限结果映射、config-option 校验与持久化、多会话隔离、断连/释放后的完全停稳,以及 HMR(热模块替换)清理。快照测试与 built-bin 测试验证应用组合,真实 API 的 e2e 测试在无 key 时自动跳过。
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.i18n.yaml b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.i18n.yaml
index 4969861f15..d4665517a8 100644
--- a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.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
-2026-06-14-acp-multi-session.md: 43aa32a43fef7aa69f5efd45d4c8176da9f459cd
-2026-06-14-acp-multi-session.zh.md: 6835f9623b89dc20d6c7b3054365cd9c39402aeb
+2026-06-14-acp-multi-session.md: 088fe984fbc94fe0d8654702d5fce9eb0581cd3b
+2026-06-14-acp-multi-session.zh.md: 2a803a6eff9b16b48fb90b2b986f74e762e5419c
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.md b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.md
index 43aa32a43f..088fe984fb 100644
--- a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.md
+++ b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.md
@@ -4,19 +4,21 @@ Status: implemented
English | [中文](2026-06-14-acp-multi-session.zh.md)
+> Written when ACP was an editor bridge, motivated by Zed's multi-session client model. [ACP as an automation-only protocol](../simplification/2026-07-23-acp-automation-only-protocol.md) removed the editor surfaces; the multiplexing decision itself is unchanged and this note now states it against the automation contract.
+
## Problem
-An ACP editor can keep several conversations alive over one agent subprocess. A single-active-session bridge would force extra processes and would not match Zed's client model, which tracks multiple session ids and concurrent loads. Multiplexing introduces isolation risks: events, prompt completion, cancellation, permission prompts, config selections, and predictable background-task ids must never cross session boundaries.
+An ACP automation client can keep several conversations alive over one agent subprocess. A single-active-session bridge would force extra processes and prevent one parent controller from driving independent children over one connection. Multiplexing introduces isolation risks: committed answers, prompt completion, cancellation, permission requests, and predictable background-task ids must never cross session boundaries.
## Decision
-The ACP bridge stores live sessions in `Map`. Agent-scoped callbacks use `ownedRecord`: look up `agent.session.id` in that forward map and accept the record only when it owns the exact agent object, so a foreign same-id object cannot claim the session. A record owns its agent handle, in-flight prompt, live tool-call presentation state, pending idle config switches, session cwd, and client capability snapshot. A separate loading-id set reserves each id before asynchronous resume so two pipelined loads cannot construct duplicate agents; distinct ids may load concurrently.
+The ACP bridge stores live sessions in `Map`. Agent-scoped callbacks use `ownedRecord`: look up `agent.session.id` in that forward map and accept the record only when it owns the exact agent object, so a foreign same-id object cannot claim the session. A record owns its agent, exact disposer, and optional in-flight prompt with the durable turn number that eventually settles it. The session header owns its cwd; the bridge keeps no parallel workspace or client-capability state.
-Every `session/event` and `agent/status` callback resolves the owning record before sending or settling anything. Each session permits one in-flight prompt independently. The prompt records a log watermark, captures its own `turn/start`, and settles only on the matching `turn/end`; a late end from a cancelled prior turn cannot resolve a newer prompt. `session/cancel` addresses one record and calls only that agent's queue-aware cancel path.
+Every `session/event` callback resolves the owning record before sending or settling anything. Each session permits one in-flight prompt independently. The prompt captures its own user-sourced message `turn/start` and settles only on the matching `turn/end`; injection turns, autonomous plugin or goal turns, and a late end from a cancelled prior turn cannot resolve it. `session/cancel` addresses one record and calls only that agent's queue-aware cancel path.
-Permission ownership uses the same exact-agent check against the forward map. The ACP `approval/request` answerer prompts only the editor session that owns the requesting agent and delegates foreign requests. User-interaction elicitations likewise route by agent ownership. Per-session sandbox and approval config values fold only that session's events, with pending idle switches stored on that record until the next turn anchors them.
+Permission ownership uses the same exact-agent check against the forward map. The ACP `approval/request` answerer sends a one-shot machine-policy request only for the session that owns the requesting agent and delegates foreign or call-less requests. The bridge has no elicitation, config-selection, or other human-interaction state.
-Background bash tasks carry an opaque owner token equal to the owning session id. `bash_output` and `bash_kill` compare the caller's token with the executor's task ownership before reading or killing; a predictable task id alone grants no access. Ownership is stored with the executor task, so a tool plugin reload does not erase it.
+Background bash tasks carry an opaque owner token equal to the owning session id. `task_output` and `task_kill` compare the caller's token with the executor's task ownership before reading or killing; a predictable task id alone grants no access. Ownership is stored with the executor task, so a tool plugin reload does not erase it.
Connection teardown clears the live map, settles each pending prompt as cancelled, and disposes all `AgentHandle`s in parallel. Each handle stops and awaits its loop, flushes the session while attached, unregisters the agent, and removes the session. Teardown is memoized and shared by client disconnect and plugin disposal.
@@ -24,13 +26,13 @@ Connection teardown clears the live map, settles each pending prompt as cancelle
[ACP v1 expressly permits several concurrent sessions on one connection](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/get-started/architecture.mdx#L16-L24), and each new session carries its own primary `cwd`. This bridge implements that session-level multiplexing, including different primary workspaces as recorded by the [per-session cwd decision](../architecture/2026-07-02-fs-per-session-cwd.md); it does not create one agent subprocess per session.
-A multi-root project inside one session is a separate optional capability: ACP defines the [effective roots as the primary `cwd` plus `additionalDirectories`](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/session-setup.mdx#L313-L367). [Zed sends the remaining project work directories only when the agent advertises that capability](https://github.com/zed-industries/zed/blob/ea77ca2818f3e059a2b61ecc7e63b67e01e1cec5/crates/agent_servers/src/acp.rs#L1139-L1145), otherwise it [drops them from the session request](https://github.com/zed-industries/zed/blob/ea77ca2818f3e059a2b61ecc7e63b67e01e1cec5/crates/agent_servers/src/acp.rs#L1454-L1472). The bridge does not advertise this capability and rejects non-empty values, as recorded in its [known limitations](../../../../packages/ui/acp/README.md#known-limitations-and-deferred-work), so a current Zed multi-root project reaches it with only the first work directory.
+A multi-root project inside one session is a separate optional capability: ACP defines the [effective roots as the primary `cwd` plus `additionalDirectories`](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/session-setup.mdx#L313-L367). The automation bridge advertises no multi-root capability and rejects non-empty `additionalDirectories`; each fresh session has exactly one workspace, as recorded in the [package contract](../../../../packages/acp/acp/README.md#protocol-contract).
-[The standard transport is one editor-launched agent subprocess per stdio connection](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/transports.mdx#L17-L42); multiple editor connections therefore require multiple subprocesses or a custom transport, while this decision guarantees multiple sessions within one connection. Within that connection, `ctx.sandboxPolicy` resolves every session's `cwd` as its own `workspace-write` root, so the shared bash and filesystem services can serve concurrent projects without granting cross-project writes. This does not add ACP `additionalDirectories`; it removes the process-wide root limit from the already-supported one-primary-root-per-session path.
+[The standard transport is one agent subprocess per stdio connection](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/transports.mdx#L17-L42); multiple connections therefore require multiple subprocesses or a custom transport, while this decision guarantees multiple sessions within one connection. Within that connection, `ctx.sandboxPolicy` resolves every session's `cwd` as its own `workspace-write` root, so the shared bash and filesystem services can serve concurrent projects without granting cross-project writes. This does not add ACP `additionalDirectories`; it removes the process-wide root limit from the already-supported one-primary-root-per-session path.
## Alternatives considered
-**One live session per connection** — rejected. It adds process overhead and contradicts the target client's multi-session shape without removing multiplexing needs from the editor.
+**One live session per connection** — rejected. It adds process overhead and prevents a programmatic parent from multiplexing independently cancellable work.
**A per-session `ctx.extend()`** — rejected. A child context does not by itself create a child plugin fiber, so listeners would still belong to the bridge fiber. The implemented bridge instead uses global listeners with explicit O(1) demultiplexing and per-session owned records; agent lifecycle is owned by `AgentHandle`.
@@ -38,10 +40,10 @@ A multi-root project inside one session is a separate optional capability: ACP d
## Consequences
-N sessions can stream, prompt, request permission, switch config, and run background tasks concurrently without interleaving or cross-settling. A cancel or dispose in one session does not affect its neighbors. The bridge pays for explicit maps and isolation tests, but it does not add one listener set per session and therefore avoids listener fan-out during long-lived connections.
+N sessions can return committed answers, prompt, request permission, and run background tasks concurrently without interleaving or cross-settling. A cancel in one session does not affect its neighbors. The bridge pays for explicit maps and isolation tests, but it does not add one listener set per session and therefore avoids listener fan-out during long-lived connections.
-The bridge still exposes no protocol method to close one live session independently. Today records leave together on connection teardown; session close/resume lifecycle capabilities remain deferred in the ACP feature checklist.
+The bridge exposes no protocol method to close one live session independently. Records leave together on connection teardown; navigation and resume belong to host APIs rather than this automation protocol.
## Verification
-The multi-session suite drives concurrent sessions through interleaved updates, independent in-flight prompts, targeted cancellation, same-id and distinct-id load races, permission routing, config isolation, and teardown. Tool-bash tests prove one session cannot read or kill another session's background task.
+The multi-session suite drives concurrent sessions through routed committed answers, independent in-flight prompts, targeted cancellation, and shared teardown; the approval and output-boundary suites cover permission routing and exact-agent rejection. Tool-bash tests prove one session cannot read or kill another session's background task.
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.zh.md b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.zh.md
index 6835f9623b..2a803a6eff 100644
--- a/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-14-acp-multi-session.zh.md
@@ -4,19 +4,21 @@ Status: implemented
[English](2026-06-14-acp-multi-session.md) | 中文
+> 本 Agent Note 写于 ACP 还是编辑器桥接层的时期,动机来自 Zed 的多会话客户端模型。[ACP 作为仅面向自动化的协议](../simplification/2026-07-23-acp-automation-only-protocol.md)移除了编辑器接口;多路复用决策本身不变,本 Agent Note 现依照自动化契约陈述它。
+
## 问题
-一个 ACP(Agent Client Protocol)编辑器可以在同一个 agent(智能体)子进程上保持多个对话。如果桥接层只支持单活跃会话,就不得不启动额外进程,也无法匹配 Zed 的客户端模型——该模型跟踪多个会话 id 和并发加载。多路复用引入了隔离风险:事件、提示词完成、取消、权限提示、配置选择以及可预测的后台 task id 绝不能跨越会话边界。
+一个 ACP(Agent Client Protocol)自动化客户端可以在同一个 agent(智能体)子进程上保持多个对话。如果桥接层只支持单活跃会话,就不得不启动额外进程,也会阻止一个父控制器通过一条连接驱动多个独立子任务。多路复用引入了隔离风险:已提交的回答、提示词完成、取消、权限请求以及可预测的后台 task id 绝不能跨越会话边界。
## 决策
-ACP 桥接层将活跃会话存储在 `Map` 中。agent 作用域的回调使用 `ownedRecord`:在正向 map 中查找 `agent.session.id`,且仅当该记录拥有精确的 agent 对象时才接纳它,使外部的同 id 对象无法冒领会话。一条记录拥有其 agent 句柄、进行中的提示词、活跃的工具调用展示状态、待处理的空闲配置切换、会话 cwd 以及客户端能力快照。一个独立的 loading-id 集合在异步恢复之前预留每个 id,使两个流水线化的加载请求无法构造出重复的 agent;不同 id 可以并发加载。
+ACP 桥接层将活跃会话存储在 `Map` 中。agent 作用域的回调使用 `ownedRecord`:在正向 map 中查找 `agent.session.id`,且仅当该记录拥有精确的 agent 对象时才接纳它,使外部的同 id 对象无法冒领会话。一条记录拥有其 agent、精确的释放器,以及可选的进行中提示词和最终结算它的持久轮次号。会话 header 拥有其 cwd;桥接层不保留平行的工作区或客户端能力状态。
-每个 `session/event` 和 `agent/status` 回调在发送或结算任何内容之前,先解析出所属记录。每个会话独立允许一个进行中的提示词。提示词记录一个日志水位线,捕获自己的 `turn/start`,并仅在匹配的 `turn/end` 到达时结算;来自已取消的前一轮次的迟到 end 不能 resolve 更新的提示词。`session/cancel` 定位到一条记录,只调用该 agent 的队列感知取消路径。
+每个 `session/event` 回调在发送或结算任何内容之前,先解析出所属记录。每个会话独立允许一个进行中的提示词。提示词捕获自己源自用户消息的 `turn/start`,并仅在匹配的 `turn/end` 到达时结算;注入轮次、插件或 goal 的自主轮次,以及来自已取消的前一轮次的迟到 end 都不能 resolve 它。`session/cancel` 定位到一条记录,只调用该 agent 的队列感知取消路径。
-权限归属使用对正向 map 的同一精确 agent 检查。ACP `approval/request` 应答器只向拥有发起请求的 agent 的编辑器会话发起提示,并将外部请求委托出去。用户交互引出同样按 agent 归属路由。每会话的沙箱和审批配置值只折叠该会话自身的事件,待处理的空闲切换存储在该记录上,直到下一轮次将其锚定。
+权限归属使用对正向 map 的同一精确 agent 检查。ACP `approval/request` 应答器只为拥有发起请求的 agent 的会话发送一次性机器策略请求,并将外部请求或不带 call id 的请求委托出去。桥接层没有表单引导、配置选择或其他人机交互状态。
-后台 bash 任务携带一个不透明的 owner token,其值等于所属会话 id。`bash_output` 和 `bash_kill` 在读取或终止之前,将调用方的 token 与执行器的任务归属进行比较;仅凭可预测的 task id 不能获得访问权。归属信息与执行器任务一起存储,因此工具插件重载不会擦除它。
+后台 bash 任务携带一个不透明的 owner token,其值等于所属会话 id。`task_output` 和 `task_kill` 在读取或终止之前,将调用方的 token 与执行器的任务归属进行比较;仅凭可预测的 task id 不能获得访问权。归属信息与执行器任务一起存储,因此工具插件重载不会擦除它。
连接拆除时清空活跃 map,将每个待处理的提示词以取消状态结算,并并行 dispose(资源释放)所有 `AgentHandle`。每个句柄停止并等待其循环完成、在仍然附着时刷新会话、注销 agent 并移除会话。拆除操作被 memoize 化,由客户端断连和插件 dispose 共享。
@@ -24,13 +26,13 @@ ACP 桥接层将活跃会话存储在 `Map` 中。agen
[ACP v1 明确允许一个连接上存在多个并发会话](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/get-started/architecture.mdx#L16-L24),每个新会话都携带自己的主 `cwd`。本桥实现该会话级多路复用,其中包括[按会话 cwd 决策](../architecture/2026-07-02-fs-per-session-cwd.md)所记录的不同主工作区;它不会为每个会话创建一个 agent 子进程。
-一个会话内部的多根项目是另一项可选能力:ACP 把[有效根目录定义为主 `cwd` 加 `additionalDirectories`](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/session-setup.mdx#L313-L367)。[Zed 仅在 agent 公布该能力时发送其余项目工作目录](https://github.com/zed-industries/zed/blob/ea77ca2818f3e059a2b61ecc7e63b67e01e1cec5/crates/agent_servers/src/acp.rs#L1139-L1145),否则会[从会话请求中丢弃这些目录](https://github.com/zed-industries/zed/blob/ea77ca2818f3e059a2b61ecc7e63b67e01e1cec5/crates/agent_servers/src/acp.rs#L1454-L1472)。如其[已知限制](../../../../packages/ui/acp/README.md#known-limitations-and-deferred-work)所记录,桥不公布该能力,并拒绝非空值,因此当前 Zed 多根项目到达桥时只携带第一个工作目录。
+一个会话内部的多根项目是另一项可选能力:ACP 把[有效根目录定义为主 `cwd` 加 `additionalDirectories`](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/session-setup.mdx#L313-L367)。自动化桥接层不公布任何多根能力,并拒绝非空的 `additionalDirectories`;如[包契约](../../../../packages/acp/acp/README.md#protocol-contract)所记录,每个全新会话恰好有一个工作区。
-[标准传输是每个 stdio 连接一个由编辑器启动的 agent 子进程](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/transports.mdx#L17-L42);多个编辑器连接因此需要多个子进程或自定义传输,而本决策保证的是一个连接内部存在多个会话。在该连接内,`ctx.sandboxPolicy` 把每个会话的 `cwd` 解析为其自己的 `workspace-write` 根目录,因此共享的 bash 和文件系统服务可以服务并发项目而不授予跨项目写入。这不会添加 ACP `additionalDirectories`;它只是从已经支持的「每会话一个主根目录」路径中移除了进程级根目录限制。
+[标准传输是每个 stdio 连接一个 agent 子进程](https://github.com/agentclientprotocol/agent-client-protocol/blob/01beb5fb5eec60e9f516a80d85eb03594bac61e3/docs/protocol/v1/transports.mdx#L17-L42);多个连接因此需要多个子进程或自定义传输,而本决策保证的是一个连接内部存在多个会话。在该连接内,`ctx.sandboxPolicy` 把每个会话的 `cwd` 解析为其自己的 `workspace-write` 根目录,因此共享的 bash 和文件系统服务可以服务并发项目而不授予跨项目写入。这不会添加 ACP `additionalDirectories`;它只是从已经支持的「每会话一个主根目录」路径中移除了进程级根目录限制。
## 曾考虑的替代方案
-**每连接单活跃会话**:否决。增加进程开销,与目标客户端的多会话形态相矛盾,且并未消除编辑器端的多路复用需求。
+**每连接单活跃会话**:否决。增加进程开销,并阻止程序化的父控制器多路复用可独立取消的工作。
**每会话 `ctx.extend()`**:否决。子上下文本身不会创建子插件 fiber,因此监听器仍属于桥接层 fiber。实际实现的桥接层使用全局监听器加显式 O(1) 解复用,以及每会话拥有的记录;agent 生命周期由 `AgentHandle` 管理。
@@ -38,10 +40,10 @@ ACP 桥接层将活跃会话存储在 `Map` 中。agen
## 后果
-N 个会话可以并发地进行流式输出、提示词、权限请求、配置切换和后台任务运行,而不会交错或跨会话结算。一个会话中的取消或 dispose 不影响相邻会话。桥接层为此付出了显式 map 和隔离测试的代价,但它不会为每个会话添加一组监听器,从而避免了长连接期间的监听器扇出。
+N 个会话可以并发地返回已提交的回答、提交提示词、请求权限和运行后台任务,而不会交错或跨会话结算。一个会话中的取消不影响相邻会话。桥接层为此付出了显式 map 和隔离测试的代价,但它不会为每个会话添加一组监听器,从而避免了长连接期间的监听器扇出。
-桥接层目前仍未暴露独立关闭单个活跃会话的协议方法。当前所有记录在连接拆除时一起离开;会话关闭/恢复的生命周期能力在 ACP 功能清单中仍处于延期状态。
+桥接层不暴露独立关闭单个活跃会话的协议方法。所有记录在连接拆除时一起离开;会话导航与恢复属于 host API,而非这个自动化协议。
## 验证
-多会话测试套件通过交错更新、独立的进行中提示词、定向取消、相同 id 与不同 id 的加载竞争、权限路由、配置隔离以及拆除来驱动并发会话。工具 bash 测试证明一个会话无法读取或终止另一个会话的后台任务。
+多会话测试套件通过按路由投递的已提交回答、独立的进行中提示词、定向取消以及共享拆除来驱动并发会话;审批与输出边界套件覆盖权限路由和精确 agent 拒绝。工具 bash 测试证明一个会话无法读取或终止另一个会话的后台任务。
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 9375a02315..7ee0bfc88c 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
-2026-06-15-code-mode.md: 94a95ba2c5760c0fed6ebfe09330546f401c42c4
-2026-06-15-code-mode.zh.md: 0ba5d69258bce23a24db9a0b470ddb074840cba7
+2026-06-15-code-mode.md: 8e964eeb36e430b58312427e45cf8e4a99582457
+2026-06-15-code-mode.zh.md: 3de1adfcf287b69ca8307244bde35f286ce4b99d
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 94a95ba2c5..8e964eeb36 100644
--- a/.agents/notes/implemented/feature/2026-06-15-code-mode.md
+++ b/.agents/notes/implemented/feature/2026-06-15-code-mode.md
@@ -50,7 +50,7 @@ Under `'code'` and `'both'` the registry owns `run_code` as a reserved presentat
**Concurrency is serialized.** Each run owns a dispatch queue, so even `Promise.all` executes tool calls in submission order. Settlement abandons queued calls that have not started. Parallelism requires per-tool concurrency-safety metadata.
-**Presentation.** `run_code`'s render intent is decided here per the [render-intent Agent Note](../architecture/2026-07-02-tool-render-intent-union.md): `presentCall` creates a `generic` card with `kind: 'execute'`, the program text as its title, and the same program text as `rawInput`; `run_code` intentionally declares no `presentResult`, so ACP and TUI complete that card through their generic raw-content fallback using the final durable `tool/result.content`, including captured logs plus the returned value, failure, or post-policy spill preview. The program is the title because ACP execute cards reliably render that field while some clients omit body and raw-input content. This is not a `terminal` card: that card's semantics are "a shell command in a working directory", which a program is not. See the [result-card completeness note](../bug-fix/2026-07-20-code-mode-result-card-completeness.md).
+**Presentation.** `run_code`'s render intent is decided here per the [render-intent Agent Note](../architecture/2026-07-02-tool-render-intent-union.md): `presentCall` creates a `generic` card with `kind: 'execute'`, the program text as its title, and the same program text as `rawInput`; `run_code` intentionally declares no `presentResult`, so the TUI and host/client runtime (Web) complete that card through their generic raw-content fallback using the final durable `tool/result.content`, including captured logs plus the returned value, failure, or post-policy spill preview. This is not a `terminal` card: that card's semantics are "a shell command in a working directory", which a program is not. See the [result-card completeness note](../bug-fix/2026-07-20-code-mode-result-card-completeness.md).
### Observability: `tool/code-dispatch`
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 0ba5d69258..3de1adfcf2 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
@@ -50,7 +50,7 @@ Cloudflare 的 [Code Mode](https://blog.cloudflare.com/code-mode/) 提出了一
**并发被序列化。** 每次 run 拥有一个分发队列,因此即使 `Promise.all` 也按提交顺序执行工具调用。结算时放弃尚未开始的排队调用。并行化需要每个工具的并发安全元数据。
-**呈现。** `run_code` 的 render intent 按[呈现意图 Agent Note](../architecture/2026-07-02-tool-render-intent-union.md)在此决定:`presentCall` 创建一个 `generic` 卡片,`kind: 'execute'`,以程序文本作为标题,并将同一程序文本作为 `rawInput`;`run_code` 有意不声明 `presentResult`,因此 ACP 和 TUI 会通过通用原始内容回退机制,使用最终持久化的 `tool/result.content` 补全该卡片,其中包括捕获的日志,以及返回值、失败信息或 post-policy 输出落盘预览。程序作为标题是因为 ACP execute 卡片可靠地渲染该字段,而某些客户端会省略 body 和 raw-input 内容。这不是 `terminal` 卡片:该卡片的语义是「工作目录中的 shell 命令」,程序不是。参见[结果卡片完整性说明](../bug-fix/2026-07-20-code-mode-result-card-completeness.md)。
+**呈现。** `run_code` 的 render intent 按[呈现意图 Agent Note](../architecture/2026-07-02-tool-render-intent-union.md)在此决定:`presentCall` 创建一个 `generic` 卡片,`kind: 'execute'`,以程序文本作为标题,并将同一程序文本作为 `rawInput`;`run_code` 有意不声明 `presentResult`,因此 TUI 和宿主/客户端运行时(Web)会通过通用原始内容回退机制,使用最终持久化的 `tool/result.content` 补全该卡片,其中包括捕获的日志,以及返回值、失败信息或 post-policy 输出落盘预览。这不是 `terminal` 卡片:该卡片的语义是「工作目录中的 shell 命令」,程序不是。参见[结果卡片完整性说明](../bug-fix/2026-07-20-code-mode-result-card-completeness.md)。
### 可观测性:`tool/code-dispatch`
diff --git a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md b/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md
deleted file mode 100644
index 166e52f8e6..0000000000
--- a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md
+++ /dev/null
@@ -1,48 +0,0 @@
-# Agent Note: Rich ACP bash rendering — the terminal card via the `_meta` convention
-
-Status: implemented
-
-English | [中文](2026-06-18-acp-terminal-and-tool-rendering.zh.md)
-
-## Problem
-
-The ACP bridge lets each tool own its call rendering via `presentCall`/`presentResult` (see [tool-call UI presentation](2026-06-14-acp-agent-client-protocol.md) and `packages/core/tools`). For `bash` we surface the exact command as the `tool_call` title, the model's `description` as a content text block, `kind: 'execute'`, and the completed output wrapped in a fenced ` ```console ` text block.
-
-Reference editors render terminal metadata as a dedicated card with cwd, command, live-style output, and exit status; plain text loses that structure. The command is the title because execute cards hide raw input, while the human-readable description remains a separate block above the card.
-
-## Key finding: agent-executed terminals use a `_meta` convention, NOT `terminal/create`
-
-The ACP spec has a *client-side* terminal sub-protocol — the agent calls the client's `terminal/create` with `{ command, args, cwd, env }` and the **editor** executes the process, then the agent reads `terminal/output` / `wait_for_exit`. That model is wrong for us: our harness executes bash itself through `dsh-bash` (sandboxed env-scrub, background-task ownership, per-session cwd). Routing execution to the editor would bypass all of that and fork execution into two backends.
-
-Studying the two reference agents (2026-06-18) shows neither uses `terminal/create` for their own shell tool — **both keep agent-side execution and emit a `_meta` convention** that Zed special-cases:
-
-- **`claude-agent-acp`** (`tools.ts`, `acp-agent.ts`): gated on `clientCapabilities._meta.terminal_output`. The `tool_call` carries `content: [{ type: 'terminal', terminalId }]` and `_meta.terminal_info.{ terminal_id, cwd }`; output/exit arrive on the `tool_call_update`'s `_meta.terminal_output.{ terminal_id, data }` and `_meta.terminal_exit.{ terminal_id, exit_code, signal }`.
-- **`codex-acp`** (`CodexToolCallMapper.ts`, `TerminalOutputMode.ts`): same `terminal_info` on the call; output via `_meta.terminal_output` (full) or `_meta.terminal_output_delta` (incremental), selected from the same `_meta.terminal_output` capability.
-
-Zed's side (`crates/agent_servers/src/acp.rs`, verified): on a `ToolCall` whose `_meta.terminal_info.terminal_id` is set, it registers a **display-only** terminal (header = `terminal_info.cwd`, label = `tool_call.title`); on a `ToolCallUpdate`, `_meta.terminal_output.data` writes to that terminal and `_meta.terminal_exit.{exit_code,signal}` sets the status. It advertises the capability as `clientCapabilities._meta.terminal_output = true`. `_meta` itself is a spec-blessed ACP extensibility point (typed `{[k]: unknown} | null` on `ToolCall`/`ToolCallUpdate`); the *specific keys* here (`terminal_info`/`terminal_output`/`terminal_exit`) are a Zed convention, not part of the ACP spec — but they are the de-facto contract for the Zed integration and the only way to get the terminal card while keeping execution agent-side.
-
-## Decision
-
-Keep `dsh-bash` agent-side execution; render the terminal card via the `_meta` convention, capability-gated, with the ` ```console ` text block as the guaranteed fallback.
-
-1. **Capability.** `initialize` reads `clientCapabilities._meta.terminal_output` and the bridge remembers it per connection.
-2. **Neutral presentation vocabulary.** `dsh-tools` gains a terminal-shaped presentation a tool can return — provider-neutral (`cwd`, the output `data`, an `exitCode`/`signal`), NO ACP types. `dsh-tool-bash` returns it for `bash` (cwd from the resolved workdir; output + exit parsed from the run result).
-3. **Bridge mapping.** When the client advertised the capability, the bridge maps that presentation to: on `tool_call`, `content:[…, {type:'terminal', terminalId}]` (any tool `content`, e.g. the description, rendered BEFORE the terminal block) + `_meta.terminal_info.{terminal_id,cwd}`; on `tool_call_update`, `_meta.terminal_output.{terminal_id,data}` (the captured output) + `_meta.terminal_exit.{terminal_id, exit_code|signal}` (the parsed exit), with the update's text `content` OMITTED (an ACP `tool_call_update.content` REPLACES the call's content collection, so re-sending the fenced block would clobber the terminal content block). `terminalId` is derived from the harness `callId` (stable, unique per call). When the capability is absent, the bridge sends the description content block on the call and the existing ` ```console ` text content on the update — unchanged.
-4. **The exit pill is parsed from the rendered output; no new execution path, no live streaming.** Output is attached at completion (from the agent's own `tool/result`), not streamed token-by-token. The exit-status pill (`_meta.terminal_exit.{exit_code,signal}`) IS emitted: the pure `presentResult(args, result)` seam sees only content blocks, so `dsh-tool-bash` recovers the structured exit by parsing the status markers (`[exit code: N]` / `[killed by signal: …]`) that `renderResult` appended — the parse is the exact inverse of the marker emission, the two co-evolve in one file, and a round-trip test guards the pair. Disposal is unaffected: nothing new to tear down, since the bridge never creates a client-side terminal.
-
-## Alternatives considered
-
-- **The ACP client-side terminal sub-protocol (`terminal/create`)** — explicitly rejected: the editor would execute the process, bypassing `dsh-bash`'s env scrub, background-task ownership, and per-session cwd, and forking execution into two backends. Both reference agents reject it the same way (the key finding above); agent-side execution plus the `_meta` convention is the only shape that yields the terminal card while keeping the harness's execution policy.
-- **Threading a structured exit through the event schema** — rejected in favor of the marker round-trip: the pure `presentResult(args, result)` seam sees only content blocks, and the parse is the exact inverse of the marker emission, co-evolving in one file under a round-trip test.
-
-## Consequences
-
-- **Zed-convention `_meta` keys.** The terminal card rides on Zed-specific keys (`terminal_info`/`terminal_output`/`terminal_exit`) inside ACP's spec-blessed `_meta` extensibility point, NOT on the ACP terminal sub-protocol. A client that doesn't recognize the keys still gets the text fallback (the capability gate ensures we only emit them when the client opted in via `_meta.terminal_output`), so a non-Zed client is never worse off. If ACP later standardizes agent-executed terminals, migrate to that and drop the convention keys.
-- **Capability honesty.** Emit terminal metadata ONLY when the client advertised `_meta.terminal_output`; the text fallback is the contract for everyone else and must never regress. Covered by a no-capability test asserting the ` ```console ` path.
-- **terminalId collisions.** Deriving it from the per-call `callId` keeps it unique within a session and stable across the call/result pair; never reuse one across calls.
-- **Exit parsed from rendered text.** The exit pill recovers `exit_code`/`signal` by parsing `renderResult`'s status markers rather than threading a structured exit through the event schema (which the pure `presentResult` seam never sees). The parse is the exact inverse of the marker emission and lives in the same file; a round-trip test pins the pair so a marker-format change that breaks the parse fails the suite. If the markers ever need to diverge from what the pill wants, surface a structured exit on the result event instead.
-- **Provider-neutral vocabulary creep.** The terminal presentation widens the `dsh-tools` surface; keep it neutral (no ACP types leak into `dsh-tools`) and only as rich as a second UI consumer would also want.
-
-## Out of scope / non-goals
-
-The text-block baseline stays the no-capability default. Two follow-ups are deliberately NOT built here and would each warrant their own Agent Note when someone takes them on: **live incremental streaming** (`_meta.terminal_output_delta` as chunks arrive, which needs an incremental-output seam on `dsh-bash`), and **command classification** (parsing a `cat`/`sed` as a `read` card with a file location, a `grep` as a `search`, etc., falling back to the terminal card — display-only, must never change what executes).
diff --git a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.zh.md b/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.zh.md
deleted file mode 100644
index 1623c8f51f..0000000000
--- a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.zh.md
+++ /dev/null
@@ -1,48 +0,0 @@
-# Agent Note: 富 ACP bash 渲染——通过 `_meta` 约定实现终端卡片
-
-Status: implemented
-
-[English](2026-06-18-acp-terminal-and-tool-rendering.md) | 中文
-
-## 问题
-
-ACP(Agent Client Protocol)桥接层允许每个工具通过 `presentCall`/`presentResult` 自行控制调用渲染(见[工具调用 UI 呈现](2026-06-14-acp-agent-client-protocol.md)与 `packages/core/tools`)。对于 `bash`,我们将确切命令作为 `tool_call` 标题呈现,模型的 `description` 作为一个内容文本块,`kind: 'execute'`,完成后的输出包裹在 ` ```console ` 围栏文本块中。
-
-参考编辑器将终端元数据渲染为一张专用卡片,包含 cwd、命令、实时风格的输出和退出状态;纯文本则丢失了这些结构。命令之所以作为标题,是因为执行卡片隐藏原始输入,而人类可读的描述保留为卡片上方的独立块。
-
-## 关键发现:agent 执行的终端使用 `_meta` 约定,而非 `terminal/create`
-
-ACP 规范有一个*客户端侧*终端子协议:agent(智能体)调用客户端的 `terminal/create`(传入 `{ command, args, cwd, env }`),由**编辑器**执行进程,然后 agent 读取 `terminal/output` / `wait_for_exit`。这个模型不适合我们:我们的 harness 通过 `dsh-bash` 自行执行 bash(沙箱化的环境清理、后台任务所有权、按会话的 cwd)。将执行路由到编辑器会绕过所有这些机制,并将执行分叉到两个后端。
-
-研究两个参考 agent(2026-06-18)发现,二者都没有为自己的 shell 工具使用 `terminal/create`——**两者都保持 agent 侧执行,并发出一套 `_meta` 约定**,由 Zed 特殊处理:
-
-- **`claude-agent-acp`**(`tools.ts`、`acp-agent.ts`):以 `clientCapabilities._meta.terminal_output` 为门控。`tool_call` 携带 `content: [{ type: 'terminal', terminalId }]` 与 `_meta.terminal_info.{ terminal_id, cwd }`;输出和退出通过 `tool_call_update` 的 `_meta.terminal_output.{ terminal_id, data }` 与 `_meta.terminal_exit.{ terminal_id, exit_code, signal }` 到达。
-- **`codex-acp`**(`CodexToolCallMapper.ts`、`TerminalOutputMode.ts`):调用上同样携带 `terminal_info`;输出通过 `_meta.terminal_output`(完整)或 `_meta.terminal_output_delta`(增量),由同一个 `_meta.terminal_output` 能力选择。
-
-Zed 侧(`crates/agent_servers/src/acp.rs`,已验证):收到 `ToolCall` 且其 `_meta.terminal_info.terminal_id` 已设置时,注册一个**仅展示**的终端(header = `terminal_info.cwd`,label = `tool_call.title`);收到 `ToolCallUpdate` 时,`_meta.terminal_output.data` 写入该终端,`_meta.terminal_exit.{exit_code,signal}` 设置状态。客户端通过 `clientCapabilities._meta.terminal_output = true` 声明此能力。`_meta` 本身是 ACP 规范认可的扩展点(在 `ToolCall`/`ToolCallUpdate` 上类型为 `{[k]: unknown} | null`);这里的*具体键*(`terminal_info`/`terminal_output`/`terminal_exit`)是 Zed 约定,不属于 ACP 规范,但它们是 Zed 集成的事实契约,也是在保持 agent 侧执行的前提下获得终端卡片的唯一方式。
-
-## 决策
-
-保持 `dsh-bash` 的 agent 侧执行;通过 `_meta` 约定渲染终端卡片,以能力声明为门控,以 ` ```console ` 文本块作为保底回退。
-
-1. **能力声明。** `initialize` 读取 `clientCapabilities._meta.terminal_output`,桥接层按连接记住它。
-2. **提供方无关的展示词汇。** `dsh-tools` 新增一种终端形态的展示结构,工具可返回它——提供方无关(`cwd`、输出 `data`、`exitCode`/`signal`),不含 ACP 类型。`dsh-tool-bash` 为 `bash` 返回该结构(cwd 来自解析后的工作目录;输出与退出从运行结果解析)。
-3. **桥接映射。** 当客户端声明了该能力时,桥接层将展示结构映射为:在 `tool_call` 上,`content:[…, {type:'terminal', terminalId}]`(工具的任何 `content`,如描述,渲染在终端块之前)+ `_meta.terminal_info.{terminal_id,cwd}`;在 `tool_call_update` 上,`_meta.terminal_output.{terminal_id,data}`(捕获的输出)+ `_meta.terminal_exit.{terminal_id, exit_code|signal}`(解析后的退出),且 update 的文本 `content` 被省略(ACP 的 `tool_call_update.content` 会替换调用的 content 集合,因此重新发送围栏块会覆盖终端内容块)。`terminalId` 由 harness 的 `callId` 派生(稳定、每次调用唯一)。当能力未声明时,桥接层在调用上发送描述内容块,在 update 上发送既有的 ` ```console ` 文本内容——行为不变。
-4. **退出信息从渲染输出中解析;无新执行路径,无实时流式传输。** 输出在完成时附加(来自 agent 自身的 `tool/result`),不逐 token 流式传输。退出状态(`_meta.terminal_exit.{exit_code,signal}`)确实会发出:纯 `presentResult(args, result)` seam 只能看到内容块,因此 `dsh-tool-bash` 通过解析 `renderResult` 追加的状态标记(`[exit code: N]` / `[killed by signal: …]`)来恢复结构化退出信息——解析是标记发出的精确逆操作,二者在同一文件中共同演进,一个往返测试守护这对关系。资源释放不受影响:无需新增拆除逻辑,因为桥接层从未创建客户端侧终端。
-
-## 曾考虑的替代方案
-
-- **ACP 客户端侧终端子协议(`terminal/create`)**:明确否决。编辑器将执行进程,绕过 `dsh-bash` 的环境清理、后台任务所有权和按会话的 cwd,并将执行分叉到两个后端。两个参考 agent 以同样的方式否决了它(见上述关键发现);agent 侧执行加 `_meta` 约定是在保持 harness 执行策略的同时获得终端卡片的唯一形态。
-- **通过事件 schema 传递结构化退出信息**:否决,改用标记往返方案。纯 `presentResult(args, result)` seam 只能看到内容块,而解析是标记发出的精确逆操作,二者在同一文件中共同演进,由往返测试守护。
-
-## 后果
-
-- **Zed 约定的 `_meta` 键。** 终端卡片依赖 Zed 特有的键(`terminal_info`/`terminal_output`/`terminal_exit`),位于 ACP 规范认可的 `_meta` 扩展点内,而非 ACP 终端子协议。不识别这些键的客户端仍然获得文本回退(能力门控确保我们仅在客户端通过 `_meta.terminal_output` 声明支持时才发出这些键),因此非 Zed 客户端不会变差。如果 ACP 日后标准化了 agent 执行的终端,则迁移到该标准并移除约定键。
-- **能力诚实。** 仅在客户端声明了 `_meta.terminal_output` 时才发出终端元数据;文本回退是对其他所有客户端的契约,绝不可退化。由一个无能力测试覆盖,断言 ` ```console ` 路径。
-- **terminalId 冲突。** 从每次调用的 `callId` 派生,保证在会话内唯一且在 call/result 对之间稳定;绝不跨调用复用。
-- **退出信息从渲染文本解析。** 退出信息通过解析 `renderResult` 的状态标记恢复 `exit_code`/`signal`,而非通过事件 schema 传递结构化退出(纯 `presentResult` seam 看不到后者)。解析是标记发出的精确逆操作,且位于同一文件中;往返测试固定了这对关系,标记格式变更若破坏解析则测试套件失败。如果标记格式日后需要与退出信息分道扬镳,则改为在 result 事件上暴露结构化退出。
-- **提供方无关词汇的蔓延。** 终端展示结构扩大了 `dsh-tools` 的接口面;保持其中立性(不让 ACP 类型泄漏到 `dsh-tools`),且只提供第二个 UI 消费方同样需要的丰富度。
-
-## 超出范围 / 非目标
-
-文本块基线仍为无能力声明时的默认行为。以下两项后续工作有意不在此处构建,各自需要单独的 Agent Note:**实时增量流式传输**(在分片到达时发出 `_meta.terminal_output_delta`,需要在 `dsh-bash` 上新增增量输出 seam);**命令分类**(将 `cat`/`sed` 解析为带文件位置的 `read` 卡片,将 `grep` 解析为 `search`,回退到终端卡片——仅展示,绝不改变实际执行内容)。
diff --git a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.i18n.yaml b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.i18n.yaml
index cd85b98708..6a06911100 100644
--- a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.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
-2026-06-25-ask-user-question.md: 9eab7aeb17774961dfc40224accdcfcbd5eb9d5f
-2026-06-25-ask-user-question.zh.md: eb166acfef7ed8f7ab4eb3018bf161a9ecc3549e
+2026-06-25-ask-user-question.md: 51b9a67bfc0fdc88d84a9fa51662d02b127f7925
+2026-06-25-ask-user-question.zh.md: 4bd2c2b8664ba085919eb9d4b8c5dee92d4bf72e
diff --git a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.md b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.md
index 9eab7aeb17..51b9a67bfc 100644
--- a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.md
+++ b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.md
@@ -18,7 +18,7 @@ The model-facing request vocabulary is deliberately aligned with the product-res
Providers return `{ answers: [{ id, selected, custom? }] }`. `selected` is always an array of selected option labels, so single-select and `multi_select` answers share one result shape. `custom` carries a free-text "Other" answer; optionless questions collect `custom` directly. When `custom` is present, it overrides any selected choices and `selected` is empty. A provider that supports partial completion represents a deliberately skipped item with the existing `{ id, selected: [] }` shape, preserving the other answers without extending the tool result vocabulary.
-`UserInteractionError` extends `HarnessError`, so failures such as `NO_PROVIDER`, `ASK_ABORTED`, ACP cancellation, or missing session routing survive `ctx.tools.execute()` as machine-routable `{ name, code }` tool errors. This matches the structured-error taxonomy and lets the model or a wrapping plugin distinguish "user cancelled" from a generic thrown exception.
+`UserInteractionError` extends `HarnessError`, so failures such as `NO_PROVIDER`, `ASK_ABORTED`, or missing request ownership survive `ctx.tools.execute()` as machine-routable `{ name, code }` tool errors. This matches the structured-error taxonomy and lets the model or a wrapping plugin distinguish "user cancelled" from a generic thrown exception.
## UI mappings
@@ -28,9 +28,7 @@ The Web composer shows one question at a time while retaining every request in t
`dsh-tui` renders each question as a keyboard overlay, shows option descriptions, supports single- and multi-select choices plus free-form custom answers, and rejects pending questions on abort, provider disposal, or terminal shutdown. Batched and simultaneous requests are queued so one overlay owns keyboard focus at a time.
-`dsh-acp` provides the same seam for ACP sessions. It resolves the calling `Agent` through `ownedRecord`, requiring the forward session-map record at `agent.session.id` to own that exact agent object, and calls ACP `unstable_createElicitation` with a session-scoped form for each question. Single-select options become a `choice` string enum; `multi_select` options become a `choice` array enum; optionless questions use a required `custom` text field. If the client returns both `choice` and non-empty `custom`, the custom answer wins. ACP `decline`/`cancel`, a missing answer, a missing session, and a client without elicitation support all become structured `UserInteractionError`s.
-
-The ACP mapping deliberately uses elicitation, not `session/request_permission`. `request_permission` is still reserved for the separate permission gate: it is a yes/no-or-policy authorization protocol around tool execution. `ask_user_question` is a general information-gathering tool with optional free-form answers, so ACP form elicitation is the closer protocol fit. The bridge's session routing is shared with the future permission gate, but the user intent is different.
+An ACP elicitation mapping existed while the bridge was an editor UI; [ACP as an automation-only protocol](../simplification/2026-07-23-acp-automation-only-protocol.md) removed that third mapping.
## Alternatives considered
@@ -38,18 +36,16 @@ The ACP mapping deliberately uses elicitation, not `session/request_permission`.
**Core-owned ask-user packages.** The first implementation split the seam and the model-facing tool across `packages/core` and `packages/ui`, but both names describe one UI-backed human-interaction affordance. The seam remains provider-neutral, but it is not providerless core infrastructure like sessions, tools, or the agent registry. Keeping `dsh-user-interaction` and `dsh-tool-ask-user` together under `packages/ui` makes the package map match the product boundary: apps and bridges provide the human-answer provider, and the stdio app opts into the model-facing tool.
-**ACP `session/request_permission`.** Permission requests are authorization around tool execution; `ask_user_question` is information gathering with optional free-form answers. Using permission for general questions would collapse two different product concepts and make the future permission gate harder to reason about.
+**Use a permission request for general questions.** Permission requests authorize tool execution; `ask_user_question` gathers information with optional free-form answers. Reusing the permission channel would collapse two different product concepts.
**A loop-level pause primitive.** The agent loop already knows how to await a tool call and resume from a tool result. Adding a new loop special case would duplicate that async shape and make every loop implementation learn about a UI concern.
## Consequences
-ACP elicitation is currently marked unstable in the SDK. The fallback is still structured: if a client does not implement it, the tool returns `ASK_FAILED` rather than hanging. A later ACP stabilization may rename or reshape the method; that migration should stay inside `dsh-acp` because the core `ctx.userInteraction` vocabulary is provider-neutral.
-
The feature gives the model a powerful pause primitive, so prompt guidance matters. The tool description tells the model to ask concise questions and use options when possible. Product policy can later wrap `tools/execute` to restrict when the tool is allowed, but the loop should not special-case it.
-`dsh-user-interaction` and `dsh-tool-ask-user` both live in `packages/ui` because they form one product-facing human-interaction capability. `agent-core` does not load either the tool or a provider. `dsh-tui-demo` opts into the seam, TUI provider, and model-facing tool. `dsh web` boots the seam/provider in the host runtime and exposes the tool through the selected Web question plugin. `acp-agent` keeps only the `userInteraction` seam/provider by default: ACP elicitation support is still client-dependent, so an ACP leaf must opt into the model-facing tool deliberately once its client can complete elicitation requests.
+`dsh-user-interaction` and `dsh-tool-ask-user` both live in `packages/ui` because they form one product-facing human-interaction capability. `agent-core` does not load either the tool or a provider. `dsh-tui-demo` opts into the seam, TUI provider, and model-facing tool. `dsh web` boots the seam/provider in the host runtime and exposes the tool through the selected Web question plugin. The ACP automation app mounts neither the seam nor the tool.
## Testing
-Unit coverage pins provider registration/disposal, duplicate-provider rejection, abort-before-provider, empty-question rejection, structured tool errors through `ctx.tools.execute()`, batched answers, multi-select answers, custom answers, explicit per-item skips, and the model schema including the removal of `value`, `recommended`, `allow_custom`, and `desc`. TUI tests cover option descriptions, queued requests, shutdown/abort cleanup, optionless free-form input, invalid choices, duplicate multi-select selections, and batched question flows. ACP bridge tests drive a real in-memory ACP connection with the real `ask_user_question` tool and verify selected-option, custom-overrides-choice, multi-select, and optionless free-form elicitation paths continue the agent loop. Web tests pin stable-id replay, response validation, first-wins settlement, duplicate and late responses, whole-request cancellation versus owner abort, single-select advance, IME-safe Enter submission, per-item skip preservation, composer takeover, structured batch submission, and restoration of the normal composer.
+Unit coverage pins provider registration/disposal, duplicate-provider rejection, abort-before-provider, empty-question rejection, structured tool errors through `ctx.tools.execute()`, batched answers, multi-select answers, custom answers, explicit per-item skips, and the model schema including the removal of `value`, `recommended`, `allow_custom`, and `desc`. TUI tests cover option descriptions, queued requests, shutdown/abort cleanup, optionless free-form input, invalid choices, duplicate multi-select selections, and batched question flows. Web tests pin stable-id replay, response validation, first-wins settlement, duplicate and late responses, whole-request cancellation versus owner abort, single-select advance, IME-safe Enter submission, per-item skip preservation, composer takeover, structured batch submission, and restoration of the normal composer.
diff --git a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.zh.md b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.zh.md
index eb166acfef..4bd2c2b866 100644
--- a/.agents/notes/implemented/feature/2026-06-25-ask-user-question.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-25-ask-user-question.zh.md
@@ -18,7 +18,7 @@ agent(智能体)有时仅凭模型推理(inference)无法安全地继续
提供方返回 `{ answers: [{ id, selected, custom? }] }`。`selected` 始终是选中选项 label 的数组,因此单选和 `multi_select` 的答案共享同一种结果形状。`custom` 承载自由文本的「其他」答案;无选项的问题直接收集 `custom`。当 `custom` 存在时,它覆盖任何已选择的选项,`selected` 为空。支持部分完成的提供方使用现有的 `{ id, selected: [] }` 形状表示某项被有意跳过,在不扩展工具结果词汇的前提下保留其他答案。
-`UserInteractionError` 继承 `HarnessError`,因此 `NO_PROVIDER`、`ASK_ABORTED`、ACP(Agent Client Protocol)取消或会话路由缺失等失败会以机器可路由的 `{ name, code }` 工具错误形式通过 `ctx.tools.execute()` 传出。这与结构化错误分类体系一致,使模型或包装插件能够区分「用户取消」与一般的抛出异常。
+`UserInteractionError` 继承 `HarnessError`,因此 `NO_PROVIDER`、`ASK_ABORTED` 或请求归属缺失等失败会以机器可路由的 `{ name, code }` 工具错误形式通过 `ctx.tools.execute()` 传出。这与结构化错误分类体系一致,使模型或包装插件能够区分「用户取消」与一般的抛出异常。
## UI 映射
@@ -28,9 +28,7 @@ Web 输入区一次显示一个问题,同时在会话对象层保留每个请
`dsh-tui` 将每个问题渲染为键盘叠层,展示选项描述,支持单选、多选和自由格式自定义答案,并在中止、提供方 dispose(资源释放)或终端关闭时拒绝待处理的问题。批量请求和并发请求都会排队,确保同一时刻只有一个叠层占用键盘焦点。
-`dsh-acp` 为 ACP 会话提供同一 seam。它通过 `ownedRecord` 解析调用方 `Agent`,要求位于 `agent.session.id` 的正向会话 map 记录拥有该精确 agent 对象,并为每个问题调用 ACP `unstable_createElicitation`(附带会话范围的表单)。单选选项变为 `choice` 字符串枚举;`multi_select` 选项变为 `choice` 数组枚举;无选项的问题使用必填的 `custom` 文本字段。如果客户端同时返回 `choice` 和非空 `custom`,以 custom 答案为准。ACP `decline`/`cancel`、缺失答案、缺失会话以及客户端不支持 elicitation,都会转为结构化的 `UserInteractionError`。
-
-ACP 映射有意使用 elicitation 而非 `session/request_permission`。`request_permission` 仍保留给独立的权限门禁:它是围绕工具执行的 yes/no 或策略式授权协议。`ask_user_question` 是一个通用的信息收集工具,支持可选的自由格式答案,因此 ACP 表单 elicitation 是更贴合的协议。bridge 的会话路由与未来的权限门禁共享,但用户意图不同。
+在桥接层还是编辑器 UI 时曾存在一个 ACP(Agent Client Protocol)elicitation 映射;[ACP 作为仅面向自动化的协议](../simplification/2026-07-23-acp-automation-only-protocol.md)移除了这第三个映射。
## 曾考虑的替代方案
@@ -38,18 +36,16 @@ ACP 映射有意使用 elicitation 而非 `session/request_permission`。`reques
**核心拥有的 ask-user 包。** 最初实现将 seam 和面向模型的工具分别放在 `packages/core` 和 `packages/ui`,但两者描述的是同一个由 UI 支撑的人机交互功能。seam 仍然是提供方无关的,但它不是像会话、工具或 agent 注册表那样的无提供方核心基础设施。将 `dsh-user-interaction` 和 `dsh-tool-ask-user` 一起放在 `packages/ui` 下,使包的划分与产品边界一致:应用和 bridge 提供人类答案的提供方,stdio 应用选择性加载面向模型的工具。
-**ACP `session/request_permission`。** 权限请求是围绕工具执行的授权;`ask_user_question` 是带可选自由格式答案的信息收集。将权限用于通用提问会混淆两个不同的产品概念,并使未来的权限门禁更难推理。
+**用权限请求处理通用提问。** 权限请求是对工具执行的授权;`ask_user_question` 是带可选自由格式答案的信息收集。复用权限通道会混淆两个不同的产品概念。
**循环级别的暂停原语。** agent loop 已经知道如何等待工具调用并从工具结果恢复。添加新的循环特殊分支会重复这一异步形状,并迫使每个循环实现都了解一个 UI 关注点。
## 后果
-ACP elicitation 目前在 SDK 中标记为 unstable。回退仍然是结构化的:如果客户端未实现它,工具返回 `ASK_FAILED` 而非挂起。后续 ACP 稳定化可能重命名或重塑该方法;该迁移应限制在 `dsh-acp` 内部,因为核心 `ctx.userInteraction` 词汇是提供方无关的。
-
该功能赋予模型一个强大的暂停原语,因此提示词引导很重要。工具描述告诉模型:提问要简洁,尽可能使用选项。产品策略后续可以包装 `tools/execute` 来限制工具何时可用,但循环不应对其做特殊处理。
-`dsh-user-interaction` 和 `dsh-tool-ask-user` 都位于 `packages/ui`,因为它们共同构成一个面向产品的人机交互能力。`agent-core` 不加载工具或提供方。`dsh-tui-demo` 选择性加载 seam、TUI 提供方和面向模型的工具。`dsh web` 在 host 运行时启动 seam/提供方,并通过选定的 Web question 插件暴露该工具。`acp-agent` 默认只保留 `userInteraction` seam/提供方:ACP elicitation 支持仍取决于客户端,因此 ACP 叶节点必须在其客户端能完成 elicitation 请求后才有意加载面向模型的工具。
+`dsh-user-interaction` 和 `dsh-tool-ask-user` 都位于 `packages/ui`,因为它们共同构成一个面向产品的人机交互能力。`agent-core` 不加载工具或提供方。`dsh-tui-demo` 选择性加载 seam、TUI 提供方和面向模型的工具。`dsh web` 在 host 运行时启动 seam/提供方,并通过选定的 Web question 插件暴露该工具。ACP 自动化应用既不挂载 seam 也不挂载该工具。
## 测试
-单元覆盖率固定了以下场景:提供方注册/释放、重复提供方拒绝、提供方就绪前中止、空问题拒绝、通过 `ctx.tools.execute()` 传出的结构化工具错误、批量答案、多选答案、自定义答案、显式按项跳过,以及模型 schema(包括移除 `value`、`recommended`、`allow_custom` 和 `desc`)。TUI 测试覆盖选项描述、排队请求、关闭/中止清理、无选项自由格式输入、无效选择、重复多选和批量问题流。ACP bridge 测试驱动一个真实的内存 ACP 连接(使用真实的 `ask_user_question` 工具),验证选中选项、custom 覆盖 choice、多选和无选项自由格式 elicitation 路径能继续 agent loop。Web 测试固定稳定 id 重放、响应校验、首个响应胜出的结算、重复和迟到响应、整个请求的取消与拥有方中止的区别、单选后前进、IME 安全的 Enter 提交、按项跳过保留、输入区接管、结构化批量提交,以及常规输入区的恢复。
+单元覆盖率固定了以下场景:提供方注册/释放、重复提供方拒绝、提供方就绪前中止、空问题拒绝、通过 `ctx.tools.execute()` 传出的结构化工具错误、批量答案、多选答案、自定义答案、显式按项跳过,以及模型 schema(包括移除 `value`、`recommended`、`allow_custom` 和 `desc`)。TUI 测试覆盖选项描述、排队请求、关闭/中止清理、无选项自由格式输入、无效选择、重复多选和批量问题流。Web 测试固定稳定 id 重放、响应校验、首个响应胜出的结算、重复和迟到响应、整个请求的取消与拥有方中止的区别、单选后前进、IME 安全的 Enter 提交、按项跳过保留、输入区接管、结构化批量提交,以及常规输入区的恢复。
diff --git a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.i18n.yaml b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.i18n.yaml
index e6fc4aba97..d7babf16b3 100644
--- a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.i18n.yaml
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write
-2026-06-29-todo-write-tool.md: bf5fceaafd475224914212beb460fdbc42e3dd68
-2026-06-29-todo-write-tool.zh.md: 3afc03a393284e2a9c2d6e234bc95f0faebbfb48
+2026-06-29-todo-write-tool.md: df1bee2801b0e01b290b63f6edbe2e5b1be80cb7
+2026-06-29-todo-write-tool.zh.md: 7fa5cb2aad2b32ef0662df04ff6576be14a3a8e7
diff --git a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.md b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.md
index bf5fceaafd..df1bee2801 100644
--- a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.md
+++ b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.md
@@ -6,31 +6,27 @@ English | [中文](2026-06-29-todo-write-tool.zh.md)
## Problem
-The harness gives the model bash and subagent tools but no way to record a structured task list. A todo list serves two co-equal purposes: it steers the model to plan multi-step work and keep the active task unambiguous (at most one active, exactly one while work remains), and it gives the human a live progress checklist. The ACP protocol has a native `plan` sessionUpdate that editors (Zed) already render, but the bridge never emitted one. Every reference coding agent surveyed (claude-code, opencode, codex, oh-my-pi, pi) ships some form of this; the harness had nothing.
+The harness gives the model bash and subagent tools but no way to record a structured task list. A todo list serves two co-equal purposes: it steers the model to plan multi-step work and keep the active task unambiguous (at most one active, exactly one while work remains), and it gives an interactive host a live progress checklist. Every reference coding agent surveyed (claude-code, opencode, codex, oh-my-pi, pi) ships some form of this; the harness had nothing.
## Decision
-Add a model-facing `todo_write(todos: [{ content, status }])` tool whose whole-list state lives on the event-sourced session log as a new `todo/write` `SessionEventMap` variant. Both the stdio UI and the ACP bridge render off the existing `session/event` — the ACP bridge maps the list to a `plan` sessionUpdate.
+Add a model-facing `todo_write(todos: [{ content, status }])` tool whose whole-list state lives on the event-sourced session log as a new `todo/write` `SessionEventMap` variant. Interactive hosts render from the durable event; the TUI folds it directly, while the [automation-only ACP bridge](../simplification/2026-07-23-acp-automation-only-protocol.md) deliberately omits todo presentation.
### Whole-list replace, three-state status
-The model sends the ENTIRE list every call; the new list replaces the old (last-write-wins on replay). This is the shape claude-code V1, opencode, and codex `update_plan` all use, and the shape the model is most trained on — no per-item ids, no delta protocol. `status` is exactly `pending | in_progress | completed`: the same triple as codex `update_plan` and, crucially, **identical to the ACP `PlanEntryStatus`**, so the bridge maps it 1:1 with no lossy translation.
+The model sends the entire list every call; the new list replaces the old (last-write-wins on replay). This is the shape claude-code V1, opencode, and codex `update_plan` all use, and the shape the model is most trained on — no per-item ids, no delta protocol. `status` is exactly `pending | in_progress | completed`, the same triple as codex `update_plan`; it also matched the ACP `PlanEntryStatus` 1:1 while the bridge projected todo lists as `plan` updates, a mapping retired with the [automation-only ACP contract](../simplification/2026-07-23-acp-automation-only-protocol.md).
### State on the session log, not a service
-The list is appended as a `todo/write` event carrying the full `{ todos }` snapshot. The harness is event-sourced — the LLM history, tool calls, and turn structure all live on the log — so the todo list lives there too. This buys durability, replay, and `session/load` reconstruction for free: a reopened session re-derives the current list (the last `todo/write`) and the ACP bridge re-emits the `plan` on load, with no separate persistence backend, no in-memory service to rehydrate, and no extra wiring. An in-memory `ctx.todos` service would have had to reinvent all of that.
+The list is appended as a `todo/write` event carrying the full `{ todos }` snapshot. The harness is event-sourced — the LLM history, tool calls, and turn structure all live on the log — so the todo list lives there too. This buys durability, replay, and resume reconstruction for free: a reopened session re-derives the current list from the latest `todo/write`, with no separate persistence backend, in-memory service to rehydrate, or extra wiring. An in-memory `ctx.todos` service would have to reinvent all of that.
### NOT a surface event
`todo/write` is deliberately excluded from `SurfaceEventType`. The surface is the projection that produces the LLM message history (`deriveMessages()`); a todo write produces no conversation message. So it carries no `surfaceOp`, never joins the ordered surface, and never reaches `deriveMessages()` — it is durable, replayable *UI* state that travels alongside the conversation without being part of it. (The dev-mode invariants still require it to sit inside an open turn, which it always does: it is appended mid-step during a tool call.)
-### Priority synthesized only at the ACP boundary
-
-ACP's `PlanEntry` requires `content` + `priority` + `status`, but a `TodoItem` has no priority — the model never reasons about it. Rather than burden the schema with a field the model must always supply, the bridge synthesizes a constant `priority: 'medium'` on every entry when it builds the `plan`. Priority is an ACP wire requirement, not a harness concept, so it lives at exactly the boundary that needs it.
-
### Dropped vs claude-code V1: `activeForm`, id, priority
-claude-code V1's item is `{ content, status, activeForm }`; later (V2) it grew ids, dependencies, and ownership — but only to support agent *swarms* (disk-backed, lock-guarded, per-item mutation). This tool keeps the item at the minimum: `{ content, status }`. No `activeForm` (the present-continuous label) — the UI shows `content`; no id — whole-list replace needs no stable identity; no priority — see above. Each dropped field is one less thing the model must produce on every call.
+claude-code V1's item is `{ content, status, activeForm }`; later (V2) it grew ids, dependencies, and ownership — but only to support agent *swarms* (disk-backed, lock-guarded, per-item mutation). This tool keeps the item at the minimum: `{ content, status }`. No `activeForm` (the present-continuous label) — the UI shows `content`; no id — whole-list replace needs no stable identity; no priority — that was only ever an ACP `PlanEntry` wire requirement, synthesized as a constant at the bridge boundary rather than modeled, and it left with that projection. Each dropped field is one less thing the model must produce on every call.
### Single owner — no swarm machinery (YAGNI)
@@ -47,18 +43,18 @@ The schema enforces type/required/enum. Beyond that, `execute` rejects empty or
## Testing
Four tiers, designed up front:
-- **Unit** — the session event (append/snapshot-clone/last-write-wins/not-on-surface); the tool (schema shape, arg validation via the real `ctx.tools.execute`, value validation, the event append + replacement, no-agent rejection, `presentCall`, HMR-safety); the ACP `todosToPlan` mapping; the stdio render arm.
+- **Unit** — the session event (append/snapshot-clone/last-write-wins/not-on-surface); the tool (schema shape, arg validation via the real `ctx.tools.execute`, value validation, the event append + replacement, no-agent rejection, `presentCall`, HMR-safety); and TUI folding.
- **Real-Loader path** — the plugin run through `Loader.unwrapExports`, asserting the namespace export shape survives (it HAS `inject`, so a stray default would crash at load — postmortem/0001).
- **Full-loop integration** — a scripted mock model calls `todo_write` through the real agent loop; the `todo/write` event lands and a second call replaces it.
-- **`session/load` replay** — a persisted `todo/write` re-emits the `plan` update when a fresh ACP bridge loads the session.
-- **With-key e2e + snapshot** — a real prompt induces a `todo_write`; the snapshot expected output gains the `plan` notification and the log event.
+- **Resume/replay** — a persisted `todo/write` folds back into the current task list.
+- **With-key e2e + snapshots** — a real prompt induces `todo_write`; assembled snapshots pin the log event and interactive rendering.
## Alternatives considered
-- **In-memory `ctx.todos` service** — would reinvent durability, replay, and `session/load` reconstruction the log gives for free.
+- **In-memory `ctx.todos` service** — would reinvent durability, replay, and resume reconstruction the log gives for free.
- **Per-item delta protocol** — only needed for a shared multi-owner list, which is out of scope; whole-list replace is simpler and matches the references.
- **Tool in `core/`** — `todo_write` is an extension tool registering on `ctx.tools`, not part of the spine; it lives in its own `packages/todo/` group like other tool families.
## Consequences
-The todo list is durable, replayable session state: a persisted `todo/write` re-emits the editor's `plan` update on `session/load`, and the log — not plugin memory — is the single source of truth. Whole-list replace means one tool call per update with last-write-wins; there is no delta protocol to reconcile. The event stays off the surface, so a todo update never perturbs the derived model history — the model sees only its own tool call and result.
+The todo list is durable, replayable session state: an interactive host re-derives it from the latest persisted `todo/write`, and the log — not plugin memory — is the single source of truth. Whole-list replace means one tool call per update with last-write-wins; there is no delta protocol to reconcile. The event stays off the model surface, so a todo update never perturbs derived model history — the model sees only its own tool call and result.
diff --git a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.zh.md b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.zh.md
index 3afc03a393..7fa5cb2aad 100644
--- a/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-29-todo-write-tool.zh.md
@@ -6,31 +6,27 @@ Status: implemented
## 问题
-harness 为模型提供了 bash 和 subagent 工具,却没有办法记录结构化的任务列表。todo 列表有两个同等重要的用途:引导模型规划多步骤工作并保持当前活跃任务明确(最多一个活跃,有剩余工作时恰好一个);同时为人类提供实时进度清单。ACP(Agent Client Protocol)协议原生支持 `plan` sessionUpdate,编辑器(Zed)已能渲染它,但 bridge 从未发出过。调研的所有参考编码 agent(智能体)(claude-code、opencode、codex、oh-my-pi、pi)都提供了某种形式的此功能;本 harness 此前没有。
+harness 为模型提供了 bash 和 subagent 工具,却没有办法记录结构化的任务列表。todo 列表有两个同等重要的用途:引导模型规划多步骤工作并保持当前活跃任务明确(最多一个活跃,有剩余工作时恰好一个);同时为交互式宿主提供实时进度清单。调研的所有参考编码 agent(智能体)(claude-code、opencode、codex、oh-my-pi、pi)都提供了某种形式的此功能;本 harness 此前没有。
## 决策
-新增一个面向模型的 `todo_write(todos: [{ content, status }])` 工具,其整列表状态作为新的 `todo/write` `SessionEventMap` 变体存储在事件溯源的会话日志上。stdio UI 和 ACP bridge 均从现有的 `session/event` 渲染;ACP bridge 将列表映射为 `plan` sessionUpdate。
+新增一个面向模型的 `todo_write(todos: [{ content, status }])` 工具,其整列表状态作为新的 `todo/write` `SessionEventMap` 变体存储在事件溯源的会话日志上。交互式宿主从持久事件渲染;TUI 直接折叠它,而[仅面向自动化的 ACP(Agent Client Protocol)桥接层](../simplification/2026-07-23-acp-automation-only-protocol.md)有意省略 todo 展示。
### 整列表替换,三态 status
-模型每次调用发送完整列表;新列表替换旧列表(回放时 last-write-wins)。这是 claude-code V1、opencode 和 codex `update_plan` 共同采用的形状,也是模型训练最多的形状——没有逐项 id,没有 delta 协议。`status` 恰好是 `pending | in_progress | completed`:与 codex `update_plan` 相同的三元组,且关键的是**与 ACP `PlanEntryStatus` 完全一致**,bridge 因此可以 1:1 映射,无需有损转换。
+模型每次调用发送完整列表;新列表替换旧列表(回放时 last-write-wins)。这是 claude-code V1、opencode 和 codex `update_plan` 共同采用的形状,也是模型训练最多的形状——没有逐项 id,没有 delta 协议。`status` 恰好是 `pending | in_progress | completed`,与 codex `update_plan` 相同的三元组;在 bridge 还把 todo 列表投影为 `plan` 更新时,它也与 ACP `PlanEntryStatus` 1:1 对应,该映射已随[仅面向自动化的 ACP 契约](../simplification/2026-07-23-acp-automation-only-protocol.md)退役。
### 状态在会话日志上,而非服务
-列表作为 `todo/write` 事件追加到日志,携带完整的 `{ todos }` 快照。harness 是事件溯源的——LLM(大语言模型)历史、工具调用和轮次结构都在日志上——所以 todo 列表也在那里。这免费获得了持久性、回放和 `session/load` 重建:重新打开的会话从最后一条 `todo/write` 重新推导当前列表,ACP bridge 在加载时重新发出 `plan`,无需独立的持久化后端、无需重新注水的内存服务、无需额外接线。一个内存中的 `ctx.todos` 服务需要重新发明以上所有。
+列表作为 `todo/write` 事件追加到日志,携带完整的 `{ todos }` 快照。harness 是事件溯源的——LLM(大语言模型)历史、工具调用和轮次结构都在日志上——所以 todo 列表也在那里。这免费获得了持久性、回放和恢复重建:重新打开的会话从最新的 `todo/write` 重新推导当前列表,无需独立的持久化后端、无需重新注水的内存服务、无需额外接线。一个内存中的 `ctx.todos` 服务需要重新发明以上所有。
### 不是 surface 事件
`todo/write` 被有意排除在 `SurfaceEventType` 之外。surface 是产出 LLM 消息历史(`deriveMessages()`)的投影;todo write 不产生对话消息。因此它不携带 `surfaceOp`,不加入有序 surface,不进入 `deriveMessages()`——它是持久、可回放的 *UI* 状态,与对话并行传输但不属于对话的一部分。(dev-mode 不变式仍要求它位于一个打开的轮次内,而它始终如此:它在工具调用的步骤中途追加。)
-### Priority 仅在 ACP 边界合成
-
-ACP 的 `PlanEntry` 要求 `content` + `priority` + `status`,但 `TodoItem` 没有 priority——模型从不推理它。与其在 schema 中增加一个模型每次都必须提供的字段,bridge 在构建 `plan` 时为每条条目合成常量 `priority: 'medium'`。Priority 是 ACP 协议格式(wire format)的要求,不是 harness 概念,因此它恰好存在于需要它的边界上。
-
### 相比 claude-code V1 舍弃的字段:`activeForm`、id、priority
-claude-code V1 的条目是 `{ content, status, activeForm }`;后来(V2)增加了 id、依赖和所有权——但仅为支持 agent *集群*(磁盘持久、锁保护、逐项变更)。本工具将条目保持在最小集:`{ content, status }`。不要 `activeForm`(现在进行时标签)——UI 直接展示 `content`;不要 id——整列表替换不需要稳定标识;不要 priority——见上文。每舍弃一个字段,模型每次调用就少产出一项。
+claude-code V1 的条目是 `{ content, status, activeForm }`;后来(V2)增加了 id、依赖和所有权——但仅为支持 agent *集群*(磁盘持久、锁保护、逐项变更)。本工具将条目保持在最小集:`{ content, status }`。不要 `activeForm`(现在进行时标签)——UI 直接展示 `content`;不要 id——整列表替换不需要稳定标识;不要 priority——它只曾是 ACP `PlanEntry` 的协议格式(wire format)要求,在 bridge 边界合成为常量而非建模,并已随该投影一起离开。每舍弃一个字段,模型每次调用就少产出一项。
### 单一所有者——无集群机制(YAGNI)
@@ -47,18 +43,18 @@ schema 强制 type/required/enum。在此之上,`execute` 拒绝为空或重
## 测试
四个层级,预先设计:
-- **单元测试**——会话事件(append/snapshot-clone/last-write-wins/not-on-surface);工具(schema 形状、通过真实 `ctx.tools.execute` 的参数校验、值校验、事件追加与替换、非 agent 拒绝、`presentCall`、HMR(热模块替换)安全性);ACP `todosToPlan` 映射;stdio 渲染分支。
+- **单元测试**——会话事件(append/snapshot-clone/last-write-wins/not-on-surface);工具(schema 形状、通过真实 `ctx.tools.execute` 的参数校验、值校验、事件追加与替换、非 agent 拒绝、`presentCall`、HMR(热模块替换)安全性);以及 TUI 折叠。
- **真实 Loader 路径**——插件通过 `Loader.unwrapExports` 运行,断言命名空间导出形状存活(它有 `inject`,因此一个意外的 default 导出会在加载时崩溃——postmortem/0001)。
- **全循环集成**——一个脚本化的 mock 模型通过真实 agent loop(智能体循环)调用 `todo_write`;`todo/write` 事件落地,第二次调用替换它。
-- **`session/load` 回放**——持久化的 `todo/write` 在新的 ACP bridge 加载会话时重新发出 `plan` 更新。
-- **带密钥 e2e + 快照**——真实提示词诱导一次 `todo_write`;快照预期输出获得 `plan` 通知和日志事件。
+- **恢复/回放**——持久化的 `todo/write` 折叠回当前任务列表。
+- **带密钥 e2e + 快照**——真实提示词诱导 `todo_write`;组装后的快照固定日志事件和交互式渲染。
## 曾考虑的替代方案
-- **内存中的 `ctx.todos` 服务**——需要重新发明日志免费提供的持久性、回放和 `session/load` 重建。
+- **内存中的 `ctx.todos` 服务**——需要重新发明日志免费提供的持久性、回放和恢复重建。
- **逐项 delta 协议**——仅在共享多所有者列表时需要,超出当前范围;整列表替换更简单,且与参考实现一致。
- **工具放在 `core/` 中**——`todo_write` 是注册在 `ctx.tools` 上的扩展工具,不属于主干;它像其他工具族一样位于自己的 `packages/todo/` 分组中。
## 后果
-todo 列表是持久、可回放的会话状态:持久化的 `todo/write` 在 `session/load` 时重新发出编辑器的 `plan` 更新,日志(而非插件内存)是唯一真源。整列表替换意味着每次更新一次工具调用,last-write-wins;没有需要协调的 delta 协议。事件不进入 surface,因此 todo 更新永远不会扰动推导出的模型历史——模型只看到自己的工具调用和结果。
+todo 列表是持久、可回放的会话状态:交互式宿主从最新持久化的 `todo/write` 重新推导它,日志(而非插件内存)是唯一真源。整列表替换意味着每次更新一次工具调用,last-write-wins;没有需要协调的 delta 协议。事件不进入模型 surface,因此 todo 更新永远不会扰动推导出的模型历史——模型只看到自己的工具调用和结果。
diff --git a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.i18n.yaml b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.i18n.yaml
index 08e3acdef5..1ad47cd1f6 100644
--- a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.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
-2026-06-30-hook-bridges.md: c944a7699eda1878c1ec2db20f45c788da89c607
-2026-06-30-hook-bridges.zh.md: cd4886e146d3d4c8b016aea166819750aa497823
+2026-06-30-hook-bridges.md: c207b6901155925215548676364e903f5de2f29b
+2026-06-30-hook-bridges.zh.md: d396279d7ed1991536da2cea39e2aec5e50960c2
diff --git a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.md b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.md
index c944a7699e..04cbef6738 100644
--- a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.md
+++ b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.md
@@ -31,7 +31,7 @@ Each bridge maps the neutral `MergedHookOutcome` from the shared lib onto the se
| `subagent/start` (emit) | additionalContext → inject into a live in-process child; a remote child has no local injection target | unsupported by this bridge |
| `subagent/end` (emit) | observe-only | unsupported by this bridge |
-The CC bridge's `ask` result is a real permission path, not a terminal bridge decision: `dsh-tools` resolves it through the optional [approval seam](2026-07-06-approval-seam.md). A composed ACP answerer prompts the owning editor session and `allowed-once` proceeds; without an ApprovalService or answerer, the call fails closed to `deny`.
+The CC bridge's `ask` result is a real permission path, not a terminal bridge decision: `dsh-tools` resolves it through the optional [approval seam](2026-07-06-approval-seam.md). An ACP automation client may answer the owning session's one-shot machine-policy request and `allowed-once` proceeds; without an ApprovalService or answerer, the call fails closed to `deny`.
### Context source is always the plugin (the mislabel guard)
@@ -57,7 +57,7 @@ Hooks run in the agent's session workspace, so relative paths target the user's
## Deferred compatibility gaps
-- **Tool-input rewrite.** A CC/Codex `updatedInput` is logged + warned, not honored — input rewrite is a deferred consistency-design problem ([the pre-tool-input-rewrite Agent Note](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)), because the pre-execution args are read by `tool/call` audit + `assistant/message` history + ACP/tool-bash presentation, so an honest rewrite is a design unit, not a field.
+- **Tool-input rewrite.** A CC/Codex `updatedInput` is logged + warned, not honored — input rewrite is a deferred consistency-design problem ([the pre-tool-input-rewrite Agent Note](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)), because the pre-execution args are read by `tool/call` audit + `assistant/message` history + tool presentation, so an honest rewrite is a design unit, not a field.
- **Stop loop-guard** (`TODO(stop-loop-guard)`). Claude Code supplies `stop_hook_active` and overrides a hook after eight consecutive blocks; Codex supplies `stop_hook_active` but documents no equivalent cap. Both bridges always report `false`, so a Stop hook that unconditionally blocks force-continues every step — a hook author must self-limit until state tracking lands.
- **Hook `continue:false` (hard halt).** A hook can ask to halt the whole run (CC/Codex `continue:false`); the shared merge folds it into `MergedHookOutcome.stop`/`stopReason`, but no bridge acts on it (`TODO(hook-continue-false)`) — the interception seams have no "hard-halt the agent" primitive yet (a Decision blocks/steers a single point, not the run). Deferred with the loop-guard work; mid-turn requests record the halt in `hook/result`, and the hook keeps its per-point effect (decision/context) meanwhile.
- **Config discovery.** The path is explicit in `cordis.yml` and process-level (see above); the full multi-layer CC/Codex precedence walk, per-session project-local discovery, and the trust/hash model are not reimplemented (`TODO(per-session-hook-config)`).
diff --git a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.zh.md b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.zh.md
index cd4886e146..60146edcba 100644
--- a/.agents/notes/implemented/feature/2026-06-30-hook-bridges.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-30-hook-bridges.zh.md
@@ -31,7 +31,7 @@ harness 的扩展面是其类型化的拦截 seam(见[拦截 seam Agent Note](
| `subagent/start`(emit) | additionalContext → 注入到存活的进程内 subagent;远程 subagent 无本地注入目标 | 本桥接不支持 |
| `subagent/end`(emit) | 仅观察 | 本桥接不支持 |
-CC 桥接的 `ask` 结果是一条真正的权限路径,而非终态桥接决策:`dsh-tools` 通过可选的[审批 seam](2026-07-06-approval-seam.md) 来解析它。组合式 ACP 应答器向拥有该会话的编辑器会话发起提示,`allowed-once` 后继续执行;如果没有 ApprovalService 或应答器,调用以 `deny` 安全关闭。
+CC 桥接的 `ask` 结果是一条真正的权限路径,而非终态桥接决策:`dsh-tools` 通过可选的[审批 seam](2026-07-06-approval-seam.md) 来解析它。ACP 自动化客户端可以应答所属会话的一次性机器策略请求,`allowed-once` 后继续执行;如果没有 ApprovalService 或应答器,调用以 `deny` 安全关闭。
### 上下文来源始终是插件(误标签防护)
@@ -57,7 +57,7 @@ Claude Code 始终导出 `CLAUDE_PROJECT_DIR`,常见的未修改钩子引用 `
## 推迟的兼容性缺口
-- **工具输入重写。** CC/Codex 的 `updatedInput` 被记录日志并发出警告,但不予执行——输入重写是一个推迟的一致性设计问题(见 [pre-tool-input-rewrite Agent Note](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)),因为 pre-execution 参数被 `tool/call` 审计、`assistant/message` 历史和 ACP/tool-bash 展示共同读取,诚实的重写是一个设计单元,而非一个字段。
+- **工具输入重写。** CC/Codex 的 `updatedInput` 被记录日志并发出警告,但不予执行——输入重写是一个推迟的一致性设计问题(见 [pre-tool-input-rewrite Agent Note](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)),因为 pre-execution 参数被 `tool/call` 审计、`assistant/message` 历史和工具展示共同读取,诚实的重写是一个设计单元,而非一个字段。
- **Stop 循环防护**(`TODO(stop-loop-guard)`)。Claude Code 提供 `stop_hook_active` 并在连续八次阻塞后覆盖钩子;Codex 提供 `stop_hook_active` 但未记录等效上限。两个桥接始终报告 `false`,因此一个无条件阻塞的 Stop 钩子会在每一步强制继续——在状态追踪落地之前,钩子作者必须自行限制。
- **钩子 `continue:false`(硬停止)。** 钩子可以请求终止整个运行(CC/Codex `continue:false`);共享合并将其折叠为 `MergedHookOutcome.stop`/`stopReason`,但没有桥接对其采取行动(`TODO(hook-continue-false)`)——拦截 seam 尚无「硬停止 agent」原语(Decision 阻塞/引导的是单个点,而非整个运行)。与循环防护工作一同推迟;轮中请求会将停止请求记录在 `hook/result` 中,钩子在此期间保留其逐点效果(决策/上下文)。
- **配置发现。** 路径在 `cordis.yml` 中显式指定且为进程级(见上文);完整的多层 CC/Codex 优先级遍历、按会话的项目本地发现以及信任/hash 模型未被重新实现(`TODO(per-session-hook-config)`)。
diff --git a/.agents/notes/implemented/feature/2026-06-30-interception-seams.i18n.yaml b/.agents/notes/implemented/feature/2026-06-30-interception-seams.i18n.yaml
index 8fc176cd9a..ff40078df0 100644
--- a/.agents/notes/implemented/feature/2026-06-30-interception-seams.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-30-interception-seams.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
-2026-06-30-interception-seams.md: fb0532741239058e7bcdf9e42a5030431b418f96
-2026-06-30-interception-seams.zh.md: 7a87f4cb7130e677ab3b376e5d08a1a8ab567c6f
+2026-06-30-interception-seams.md: 81799f4c6e3e7a4c6b9605cd97f5728b99d11995
+2026-06-30-interception-seams.zh.md: 65ae16842c632641e7ac65908162f4784dc6e1e0
diff --git a/.agents/notes/implemented/feature/2026-06-30-interception-seams.md b/.agents/notes/implemented/feature/2026-06-30-interception-seams.md
index fb05327412..9c293c273e 100644
--- a/.agents/notes/implemented/feature/2026-06-30-interception-seams.md
+++ b/.agents/notes/implemented/feature/2026-06-30-interception-seams.md
@@ -43,7 +43,7 @@ Core dispatch and the tool body sit inside normalization boundaries, so tool, li
### Pre-tool input rewrite is a separate consistency decision
-`PreToolDecision` cannot rewrite arguments. History and the audit call are logged before execution, and ACP presentation reads the same input, so the registry seals arguments before policy. A valid rewrite must update history, audit, presentation, and execution before identity is created; that contract belongs to the [input-rewrite proposal](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md).
+`PreToolDecision` cannot rewrite arguments. History and the audit call are logged before execution, and UI presentation reads the same input, so the registry seals arguments before policy. A valid rewrite must update history, audit, presentation, and execution before identity is created; that contract belongs to the [input-rewrite proposal](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md).
### Boundaries
diff --git a/.agents/notes/implemented/feature/2026-06-30-interception-seams.zh.md b/.agents/notes/implemented/feature/2026-06-30-interception-seams.zh.md
index 7a87f4cb71..39eb68983d 100644
--- a/.agents/notes/implemented/feature/2026-06-30-interception-seams.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-30-interception-seams.zh.md
@@ -43,7 +43,7 @@ harness 需要一套钩子子系统:用户像 Claude Code(CC)和 Codex 那
### 工具执行前输入重写是一个独立的一致性决策
-`PreToolDecision` 不能重写参数。历史和审计调用在执行前记录,ACP 展示读取相同的输入,因此注册表在策略之前封存参数。有效的重写必须在身份创建之前同时更新历史、审计、展示和执行;该契约属于[输入重写提案](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)。
+`PreToolDecision` 不能重写参数。历史和审计调用在执行前记录,UI 展示读取相同的输入,因此注册表在策略之前封存参数。有效的重写必须在身份创建之前同时更新历史、审计、展示和执行;该契约属于[输入重写提案](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)。
### 边界
diff --git a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.i18n.yaml b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.i18n.yaml
index b1ad660010..b535ef6081 100644
--- a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.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
-2026-06-30-session-store-fork-api.md: 4e4ea7f1d3aa3290bdb21999a292dbdfd016302e
-2026-06-30-session-store-fork-api.zh.md: d17daa464ad59c6b5ac4d5041b4af30c44321fa7
+2026-06-30-session-store-fork-api.md: 13d411e915b2e34b15a12e632f1a5e047f4aeedc
+2026-06-30-session-store-fork-api.zh.md: bcf15eb581af3993ed2d71a2b7dc604faa6e4433
diff --git a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.md b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.md
index 4e4ea7f1d3..13d411e915 100644
--- a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.md
+++ b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.md
@@ -40,4 +40,4 @@ An empty prefix is forkable; any non-empty boundary must be a safe existing sequ
The public surface stays small and discoverable: live session branching is part of `ctx.sessions`, next to `create({ seed })`, rather than a standalone service or a two-step helper pair. Persistence continues to work through existing `session/created` and `session/flush` behavior: a forked child starts life with seeded events, so existing backends persist that seed once and preserve `parentSession` / `seedLength` in the header.
-The v1 scope still excludes ACP `session/fork`, unloaded persisted-session forking, model-facing tools, and subagent refactors. If a future ACP method is added, it should advertise the capability only after it has transcript/snapshot coverage; this Agent Note adds no editor-facing updates, so no ACP snapshot is required now. Fork-child replay remains covered by the existing [seed-boundary testing Agent Note](../testing/2026-06-22-fork-child-replay-seed-boundary.md), while this API gets focused `dsh-session` unit tests plus JSONL persistence coverage.
+The v1 scope still excludes ACP `session/fork`, unloaded persisted-session forking, model-facing tools, and subagent refactors. If a future ACP method is added, it should advertise the capability only after it has protocol and snapshot coverage; this Agent Note adds no ACP wire behavior, so no ACP snapshot is required. Fork-child replay remains covered by the existing [seed-boundary testing Agent Note](../testing/2026-06-22-fork-child-replay-seed-boundary.md), while this API gets focused `dsh-session` unit tests plus JSONL persistence coverage.
diff --git a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.zh.md b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.zh.md
index d17daa464a..bcf15eb581 100644
--- a/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.zh.md
+++ b/.agents/notes/implemented/feature/2026-06-30-session-store-fork-api.zh.md
@@ -40,4 +40,4 @@ class SessionStore extends Service {
公开接口保持精简且易于发现:活跃会话分支是 `ctx.sessions` 的一部分,紧邻 `create({ seed })`,而非一个独立服务或一对两步辅助函数。持久化继续通过现有的 `session/created` 和 `session/flush` 行为运作:fork 出的子会话以种子事件开始生命,因此现有后端只需持久化该种子一次,并在 header 中保存 `parentSession`/`seedLength`。
-v1 范围仍然排除 ACP(Agent Client Protocol) `session/fork`、对未加载的已持久化会话的 fork、面向模型的工具,以及 subagent 重构。如果未来添加 ACP 方法,应在具备 transcript(文本记录)/快照覆盖后才广播该能力;本 Agent Note 不添加面向编辑器的更新,因此当前不需要 ACP 快照。fork 子会话的回放仍由现有的[种子边界测试 Agent Note](../testing/2026-06-22-fork-child-replay-seed-boundary.md) 覆盖,而本 API 则获得专门的 `dsh-session` 单元测试加 JSONL 持久化覆盖。
+v1 范围仍然排除 ACP(Agent Client Protocol) `session/fork`、对未加载的已持久化会话的 fork、面向模型的工具,以及 subagent 重构。如果未来添加 ACP 方法,应在具备协议与快照覆盖后才广播该能力;本 Agent Note 不添加任何 ACP 协议行为,因此不需要 ACP 快照。fork 子会话的回放仍由现有的[种子边界测试 Agent Note](../testing/2026-06-22-fork-child-replay-seed-boundary.md) 覆盖,而本 API 则获得专门的 `dsh-session` 单元测试加 JSONL 持久化覆盖。
diff --git a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.i18n.yaml b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.i18n.yaml
index 2b02053f05..cc2e199e60 100644
--- a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.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
-2026-07-05-dynamic-workflows.md: 4c8606fb617b3fb6b2e8ad9c35d77c585b1fe8b1
-2026-07-05-dynamic-workflows.zh.md: 757eafb5d21f42b2ca529ebeb6b16eb8205e823d
+2026-07-05-dynamic-workflows.md: bba62098c66477a3f1929f9029e81c645bfc4d41
+2026-07-05-dynamic-workflows.zh.md: 6aa1ce63f0edf9dbf296d12d3bc0c62594fa33a6
diff --git a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.md b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.md
index 4c8606fb61..bba62098c6 100644
--- a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.md
+++ b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.md
@@ -60,7 +60,7 @@ Worker-side logic runs through an in-process `MessageChannel` so V8 coverage mea
- **Nested `workflow()`**, **token `budget`**, and the `effort`/`isolation`/`agentType` agent options (each rejects loud with a message naming it deferred).
- **An overall run wall-clock timeout** — cancellation always frees the caller (result settles within the grace), so a cap on total run time is a policy knob for the background redesign, not a correctness need here.
- **Engine hardening beyond worker threads**: an isolated-vm or separate-process engine behind the same seam (actual sandboxing; memory limits).
-- **ACP progress UI** over the `workflow/*` events (a `/workflows`-style view); the events exist for it.
+- **Human-interface progress UI** over the `workflow/*` events (a `/workflows`-style view); the events exist for it.
- **ACP-backend structured output** and **`toolFilter`** (both still capability-gated `false`).
## Alternatives considered
diff --git a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.zh.md b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.zh.md
index 757eafb5d2..6aa1ce63f0 100644
--- a/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-05-dynamic-workflows.zh.md
@@ -60,7 +60,7 @@ worker 侧逻辑通过进程内 `MessageChannel` 运行,使 V8 覆盖率能够
- **嵌套 `workflow()`**、**token `budget`**,以及 `effort`/`isolation`/`agentType` agent 选项(每个都以命名延迟的消息大声拒绝)。
- **整体运行的挂钟超时**:取消总能释放调用方(result 在宽限期内 settle),因此总运行时间上限是后台重设计的策略旋钮,不是此处的正确性需求。
- **超越 worker 线程的引擎加固**:在同一 seam 背后使用 isolated-vm 或独立进程引擎(真正的沙箱化;内存限制)。
-- **ACP 进度 UI**(基于 `workflow/*` 事件的 `/workflows` 风格视图);事件已为此而存在。
+- **面向人类界面的进度 UI**(基于 `workflow/*` 事件的 `/workflows` 风格视图);事件已为此而存在。
- **ACP 后端结构化输出**和 **`toolFilter`**(两者仍以能力标志 `false` 门控)。
## 曾考虑的替代方案
diff --git a/.agents/notes/implemented/feature/2026-07-05-skill-system.i18n.yaml b/.agents/notes/implemented/feature/2026-07-05-skill-system.i18n.yaml
index b5882ad5ea..9cd21cbb39 100644
--- a/.agents/notes/implemented/feature/2026-07-05-skill-system.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-05-skill-system.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
-2026-07-05-skill-system.md: 3c328adf09ab0f6ad3388a01bac3661d45c9247e
-2026-07-05-skill-system.zh.md: ba30a3735b73b00afb46baa24f969fd687a759b5
+2026-07-05-skill-system.md: 95772067b3b224f15fd290876c297eccbe5ab97e
+2026-07-05-skill-system.zh.md: 589410076b6da1b108b01f55bf217001b2a8404f
diff --git a/.agents/notes/implemented/feature/2026-07-05-skill-system.md b/.agents/notes/implemented/feature/2026-07-05-skill-system.md
index 3c328adf09..95772067b3 100644
--- a/.agents/notes/implemented/feature/2026-07-05-skill-system.md
+++ b/.agents/notes/implemented/feature/2026-07-05-skill-system.md
@@ -12,7 +12,7 @@ DeepSeek Harness uses the same primitive so project-specific review, plugin-auth
## Decision
-`@deepseek-ai/dsh-skill` is the pure provider registry (`ctx.skills`), `@deepseek-ai/dsh-skill-local` is the shipped local filesystem provider, and `@deepseek-ai/dsh-tool-skill` owns the session-prefix catalog and model-facing loader tool. `dsh-agent-spine-demo` loads the registry, local provider, and consumer by default so stdio and ACP apps get the same behavior while embedded or remote providers contribute skills without changing the registry or consumer. Its `skills` config forwards `registry`, `local`, and `tool` branches to those owners.
+`@deepseek-ai/dsh-skill` is the pure provider registry (`ctx.skills`), `@deepseek-ai/dsh-skill-local` is the shipped local filesystem provider, and `@deepseek-ai/dsh-tool-skill` owns the session-prefix catalog and model-facing loader tool. `dsh-agent-spine-demo` loads the registry, local provider, and consumer by default so TUI, headless, and ACP apps get the same behavior while embedded or remote providers contribute skills without changing the registry or consumer. Its `skills` config forwards `registry`, `local`, and `tool` branches to those owners.
Provider plugins register synchronously during `apply()`. Provider membership is direct effect-owned state: registration and disposal invalidate completed catalogs synchronously, and discovery reads the current provider map on demand rather than observing registry-change events. Provider catalogs return ranked candidates from awaited `list()` calls, where remote providers perform initialization, authentication, and discovery while honoring the lookup abort signal. The registry validates each candidate, resolves same-name skills first-wins by rank, provider registration order, and provider-local order, then sorts summaries by skill name for deterministic consumers. It caches only completed catalog snapshots and retries when a provider/runtime revision changes during discovery, so an unload cannot freeze a stale, unresolvable skill into a session prefix. Runtime `ctx.skills.register(...)` remains a convenience for embedded in-process skills and uses project-over-user priority; `runtime` is reserved as the registry-owned provider name.
@@ -32,7 +32,7 @@ The data structures and catalog/tool contract are documented in [skills.md](../.
**Inject full skill bodies into every system prompt.** Rejected because it destroys progressive disclosure and makes every request pay for instructions that may not apply.
-**Expose skills only as slash commands.** Rejected because model-initiated loading is the core capability; slash/ACP command advertisement does not change discovery.
+**Expose skills only as slash commands.** Rejected because model-initiated loading is the core capability; human command advertisement does not change discovery.
**Put local filesystem scanning directly inside `ctx.skills`.** Rejected because coding agents, web agents, and future plugin ecosystems need different skill sources. A provider registry mirrors the subagent seam: the registry owns conflict resolution and consumers, while implementations own loading.
diff --git a/.agents/notes/implemented/feature/2026-07-05-skill-system.zh.md b/.agents/notes/implemented/feature/2026-07-05-skill-system.zh.md
index ba30a3735b..589410076b 100644
--- a/.agents/notes/implemented/feature/2026-07-05-skill-system.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-05-skill-system.zh.md
@@ -12,7 +12,7 @@ DeepSeek Harness 使用同一原语,使项目特定的评审、插件编写和
## 决策
-`@deepseek-ai/dsh-skill` 是纯提供方注册表(`ctx.skills`),`@deepseek-ai/dsh-skill-local` 是随附的本地文件系统提供方,`@deepseek-ai/dsh-tool-skill` 负责会话前缀目录与面向模型的 loader 工具。`dsh-agent-spine-demo` 默认加载注册表、本地提供方和消费方,使 stdio 与 ACP(Agent Client Protocol)应用获得相同行为,同时嵌入式或远程提供方可在不修改注册表或消费方的前提下贡献 skill。其 `skills` 配置将 `registry`、`local` 和 `tool` 分支分别转发给对应的所有者。
+`@deepseek-ai/dsh-skill` 是纯提供方注册表(`ctx.skills`),`@deepseek-ai/dsh-skill-local` 是随附的本地文件系统提供方,`@deepseek-ai/dsh-tool-skill` 负责会话前缀目录与面向模型的 loader 工具。`dsh-agent-spine-demo` 默认加载注册表、本地提供方和消费方,使 TUI、headless 与 ACP(Agent Client Protocol)应用获得相同行为,同时嵌入式或远程提供方可在不修改注册表或消费方的前提下贡献 skill。其 `skills` 配置将 `registry`、`local` 和 `tool` 分支分别转发给对应的所有者。
提供方插件在 `apply()` 期间同步注册。提供方成员资格是由直接 effect 持有的状态:注册与 dispose(资源释放)同步地使已完成的目录失效,发现操作按需读取当前提供方映射而非监听注册表变更事件。提供方目录从等待的 `list()` 调用返回排序后的候选项,远程提供方在此过程中执行初始化、认证和发现,同时遵守查找的 abort 信号。注册表校验每个候选项,按排名、提供方注册顺序和提供方内部顺序以先到先得方式解决同名 skill 冲突,然后按 skill 名称排序摘要以保证消费方获得确定性结果。它仅缓存已完成的目录快照,并在发现过程中提供方/运行时修订版本发生变化时重试,因此卸载操作不会将一个陈旧且不可解析的 skill 冻结到会话前缀中。运行时 `ctx.skills.register(...)` 仍作为嵌入式进程内 skill 的便捷方式保留,使用 project 优先于 user 的优先级;`runtime` 保留为注册表拥有的提供方名称。
@@ -32,7 +32,7 @@ DeepSeek Harness 使用同一原语,使项目特定的评审、插件编写和
**将完整 skill 正文注入每条系统提示词。** 否决,因为这破坏了渐进式披露,使每个请求都为可能不适用的指令付出代价。
-**仅以斜杠命令暴露 skill。** 否决,因为模型主动加载是核心能力;斜杠/ACP 命令广播不改变发现机制。
+**仅以斜杠命令暴露 skill。** 否决,因为模型主动加载是核心能力;面向人类的命令广播不改变发现机制。
**将本地文件系统扫描直接放入 `ctx.skills`。** 否决,因为编码 agent、Web agent 和未来的插件生态需要不同的 skill 来源。提供方注册表与 subagent seam 镜像:注册表拥有冲突解决和消费方,实现拥有加载。
diff --git a/.agents/notes/implemented/feature/2026-07-06-approval-seam.i18n.yaml b/.agents/notes/implemented/feature/2026-07-06-approval-seam.i18n.yaml
index ec785c65c4..d225ddf2d5 100644
--- a/.agents/notes/implemented/feature/2026-07-06-approval-seam.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-06-approval-seam.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
-2026-07-06-approval-seam.md: e28587e0185530164f300562f7dd5e57255ac515
-2026-07-06-approval-seam.zh.md: 94ffb330cd36ae88975e8ae9d579b3b4606a5c25
+2026-07-06-approval-seam.md: 70ccd4d486ad6e0126fa2eb638a064e9fc89bba6
+2026-07-06-approval-seam.zh.md: d218f79888957735305db14cd97cc74480297d29
diff --git a/.agents/notes/implemented/feature/2026-07-06-approval-seam.md b/.agents/notes/implemented/feature/2026-07-06-approval-seam.md
index e28587e018..70ccd4d486 100644
--- a/.agents/notes/implemented/feature/2026-07-06-approval-seam.md
+++ b/.agents/notes/implemented/feature/2026-07-06-approval-seam.md
@@ -6,13 +6,13 @@ English | [中文](2026-07-06-approval-seam.zh.md)
## Problem
-Two callers need to put one question — "may this specific action proceed?" — to a human: `tools/pre-execute`'s `ask` decision (including the Claude-Code hook bridge's `permissionDecision: ask`) and the [sandbox Agent Note](2026-07-06-sandbox.md)'s post-denial one-shot escalation retry. A shared seam keeps them from inventing separate outcome vocabularies, UI routing, cancellation, and audit trails, while guaranteeing that a deployment with no UI can never grant an unanswerable request.
+Two callers need one closed decision — "may this specific action proceed?": `tools/pre-execute`'s `ask` decision (including the Claude-Code hook bridge's `permissionDecision: ask`) and the [sandbox Agent Note](2026-07-06-sandbox.md)'s post-denial one-shot escalation retry. A shared seam keeps them from inventing separate outcome vocabularies, channel routing, cancellation, and audit trails, while guaranteeing that a deployment with no answerer can never grant an unanswerable request. The answerer may be an interactive host or an automated controller.
-The routing problem is ownership: an approval prompt must reach the editor session that owns the asking agent (the ACP bridge multiplexes N sessions over one connection), fail closed for agents nobody owns (in-process subagents, tests), and stay out of deployments that compose no UI (headless, CI).
+The routing problem is ownership: a permission request must reach the channel that owns the asking agent, fail closed for agents nobody owns, and stay out of deployments that compose no answerer.
## Decision
-One package, `dsh-user-approval` (`packages/ui/user-approval`), owning the vocabulary and the `ctx.approval` service — the MECHANISM. The POLICY — who answers, and whether a session is asked at all — lives outside it: answerers are `approval/request` waterfall listeners registered by the plugins that own the channel (the ACP bridge; future terminal UIs; test scripts), and a per-session policy tier can decide before any human is involved. Consumers (`dsh-tools`' ask routing, the sandbox escalation gate) resolve a question to a closed outcome and derive their own tool results from it. Deliberately ONE package, not the capability-seam three (see Alternatives).
+One package, `dsh-user-approval` (`packages/ui/user-approval`), owns the vocabulary and the `ctx.approval` service — the mechanism. The policy — who answers, and whether a session is asked at all — lives outside it: answerers are `approval/request` waterfall listeners registered by channel-owning plugins (the ACP bridge, host adapters, and test scripts), and a per-session policy tier can decide before a channel is involved. Consumers (`dsh-tools`' ask routing and the sandbox escalation gate) resolve a question to a closed outcome and derive their own tool results from it. This is deliberately one package, not the capability-seam three (see Alternatives).
### How a deployment uses it
@@ -25,11 +25,11 @@ One `cordis.yml` entry mounts the seam. Not loading it is the fail-closed opt-ou
# policy: never # deployment default for sessions without an override; 'ask' when omitted
```
-The entry alone provides mechanism, not a channel: with no answerer composed, every ask resolves `unavailable` and the asking tool call denies — fail-closed needs no configuration. Composing the ACP app (`@deepseek-ai/dsh-acp-demo`, as in [the acp-agent example's default tree](../../../../examples/acp-agent/README.md)) completes the loop: its bridge registers an answerer that prompts the owning editor session via `session/request_permission`, so a hook's `ask` or an escalation request surfaces as a one-shot Allow/Reject prompt attached to the already-streamed tool call. `policy: never` is the unattended stance — every ask auto-rejects deterministically, stated in the system prompt, no human in the loop. `policy` is validated against the closed list at plugin load; anything else throws.
+The entry alone provides mechanism, not a channel: with no answerer composed, every ask resolves `unavailable` and the asking tool call denies — fail-closed needs no configuration. Composing the ACP app (`@deepseek-ai/dsh-acp-demo`, as in [the acp-agent example's default tree](../../../../examples/acp-agent/README.md)) completes the loop: its [automation-only bridge](../simplification/2026-07-23-acp-automation-only-protocol.md) registers an answerer that sends `session/request_permission` to the owning client with the exact tool-call id and one-shot allow/reject options. `policy: never` is the unattended stance — every ask auto-rejects deterministically and is stated in the system prompt. `policy` is validated against the closed list at plugin load; anything else throws.
What a composed deployment observes: `allowed-once` lets exactly that call proceed; rejection, dismissal, and channel absence deny with three distinct reasons the model can tell apart; a successful in-turn request lands a durable `approval/asked`/`approval/decided` pair on the asking agent's session log; nothing about a grant persists past the call that asked. An idle request or audit append failure rejects instead of returning an unaudited decision.
-One ask under this composition, verbatim from the sandbox example's recorded `escalation-approved` scenario — the model requests a sandbox escalation, the gate asks, the bridge prompts the owning editor, the user clicks Allow once:
+One ask under this composition, from the sandbox example's recorded `escalation-approved` scenario — the model requests a sandbox escalation, the gate asks, and the automation client selects Allow once:
```
tool/call bash {"command": "printf 'escalated\n' > escalated.txt && cat escalated.txt",
@@ -40,12 +40,12 @@ approval/asked {"toolName": "bash", "callId": "call_00_…",
→ session/request_permission {"toolCall": {"toolCallId": "call_00_…"},
"options": [{"optionId": "allow-once", "name": "Allow once", "kind": "allow_once"},
{"optionId": "reject-once", "name": "Reject", "kind": "reject_once"}]}
- ← the user picks "Allow once" on the prompt the editor attaches to the streamed bash call
+ ← the client selects "Allow once"
approval/decided {"outcome": "allowed-once"}
tool/result "escalated" — this one call ran under the wider mode; the grant died with it
```
-The `escalation-rejected` twin ends in `{"outcome": "rejected"}` instead: nothing executes, and the model's result carries the asker's verbatim fail-closed text (`the user rejected escalating this command to "workspace-write"`). A hook's `permissionDecision: ask` rides the identical wire; only the asker and its deny texts differ (§ Ask routing in dsh-tools). Headless, the same request skips the prompt entirely and settles `unavailable`.
+The `escalation-rejected` twin ends in `{"outcome": "rejected"}` instead: nothing executes, and the model's result carries the asker's verbatim fail-closed text (`the user rejected escalating this command to "workspace-write"`). A hook's `permissionDecision: ask` rides the identical wire; only the asker and its deny texts differ (§ Ask routing in dsh-tools). Without an answerer, the same request settles `unavailable`.
### Design detail
@@ -55,11 +55,11 @@ After validation and a successful `approval/asked` append, the service resolves
Answerers are `approval/request` waterfall listeners. Zero listeners fall through to `unavailable`; a recognizing listener occupies the first-wins decision slot, while an unrecognized agent must delegate with `next()`. Listeners dispose with their fibers, so an unloaded channel fails closed. Because sibling registration order is not deterministic, a deployment composes one terminal answerer and reserves `prepend` for decide-or-delegate gates.
-`ApprovalRequest` carries the asking `agent`, `toolName`, optional exact `callId`, human-readable `reason`, and optional `signal`. It uses the `CallId` brand without importing `dsh-tools`, which depends on this seam. Tool arguments stay on the already-streamed call that a UI references by `callId`.
+`ApprovalRequest` carries the asking `agent`, `toolName`, optional exact `callId`, human-readable `reason`, and optional `signal`. It uses the `CallId` brand without importing `dsh-tools`, which depends on this seam. Channel adapters correlate any richer call state by `callId`; the approval request does not duplicate tool arguments.
#### Ask routing in dsh-tools
-`ToolRegistry.execute()` resolves `ask` before dispatch: `allowed-once` proceeds, while rejection, cancellation, and channel absence produce distinct deny reasons. Opportunistic `ctx.get('approval')` consumption lets an absent or unmounted service fail closed without gating the registry fiber. Agent-less execution also fails closed because it has neither an audit session nor a UI owner.
+`ToolRegistry.execute()` resolves `ask` before dispatch: `allowed-once` proceeds, while rejection, cancellation, and channel absence produce distinct deny reasons. Opportunistic `ctx.get('approval')` consumption lets an absent or unmounted service fail closed without gating the registry fiber. Agent-less execution also fails closed because it has neither an audit session nor a channel owner.
#### The per-session policy tier
@@ -67,9 +67,9 @@ The seam also owns the session-scoped `'ask' | 'never'` policy described by [the
#### The ACP answerer
-The ACP bridge answers only for an exact agent object owned by its forward session map. It attaches `session/request_permission` to the existing `callId`, advertises one-shot allow/reject options, maps cancellation separately, and never grants an unknown option. Foreign or call-less requests delegate; a failed client RPC becomes `unavailable`. Hooks and `tools/pre-execute` decide whether a call asks at all.
+The ACP bridge answers only for an exact agent object owned by its session map. It sends `session/request_permission` with the existing `callId`, advertises one-shot allow/reject options, maps cancellation separately, and never grants an unknown option. Foreign or call-less requests delegate; a failed client RPC becomes `unavailable`. Hooks and `tools/pre-execute` decide whether a call asks at all. This channel is machine policy between an automated client and its agent, not ACP presentation.
-The answerer routes through the bridge's exact-agent ownership check described by [the ACP support Agent Note](2026-06-14-acp-agent-client-protocol.md), implementing the per-session permission ownership required by [the multi-session Agent Note](2026-06-14-acp-multi-session.md).
+The answerer routes through the bridge's exact-agent ownership check described by [the automation-only ACP Agent Note](../simplification/2026-07-23-acp-automation-only-protocol.md), preserving the per-session permission ownership required by [the multi-session Agent Note](2026-06-14-acp-multi-session.md).
#### Audit, and what the model sees
@@ -88,13 +88,13 @@ Snapshots record allowed and rejected sandbox escalation through `session/reques
## Deferred
- **`allow_always` grant storage** — honoring a persistent grant means designing storage, scope identity (call? path? prefix? session? time window?), and revocation; until designed, only the one-shot options are advertised ([the sandbox Agent Note](2026-07-06-sandbox.md) § Escalation records the open scope question).
-- **A recorded hook-driven `ask` through a composed answerer** — the human-prompt wire is recorded through the sandbox example's escalation branches. The hook matrix's `hook-cc-pretool-ask` pins the no-ApprovalService fallback denial, while the hook-producer-plus-answerer composition remains on the unit tier.
-- **Routing a child agent's approvals to the parent session** — `subagent-acp`'s child auto-answers its own `permission` requests; surfacing them to the parent's editor is its own design.
+- **A recorded hook-driven `ask` through a composed answerer** — the permission wire is recorded through the sandbox example's escalation branches. The hook matrix's `hook-cc-pretool-ask` pins the no-ApprovalService fallback denial, while the hook-producer-plus-answerer composition remains on the unit tier.
+- **Routing a child agent's approvals to the parent session** — `subagent-acp`'s child auto-answers its own permission requests; delegating them to the parent controller is its own design.
## Alternatives considered
- **A single registered provider instead of waterfall listeners** — rejected: a `registerProvider()` surface forces every composition question — allowlist pre-filters, external hook deciders, scripted test answers, a policy gate in front of a human — inside one provider implementation. The waterfall gets composition, fail-closed absence, and HMR disposal from machinery the runtime already has; the seam's JSDoc pins the single-decision-slot convention instead of inventing a provider registry.
-- **An inline `tools/pre-execute` permission gate in the ACP bridge** — rejected: prompting for every bridge-owned call hardwires the asking POLICY into the UI plugin, cannot serve a second asker (sandbox escalation happens after execution starts, with no pre-execute moment), and leaves hook-produced `ask` decisions without a shared mechanism.
+- **An inline `tools/pre-execute` permission gate in the ACP bridge** — rejected: prompting for every bridge-owned call hardwires the asking policy into the transport, cannot serve a second asker (sandbox escalation happens after execution starts, with no pre-execute moment), and leaves hook-produced `ask` decisions without a shared mechanism.
- **The generic user-interaction seam (`ctx.userInteraction`)** — rejected as the approval mechanism: the two share a skeleton (route by agent, block for a human, handle absence), but approval's contract is narrower in every dimension that matters: a closed outcome vocabulary instead of free text, a protocol-native prompt attached to a tool call instead of a generic form, mandatory fail-closed absence, and audit events. Approval therefore does not ride the shipped `packages/ui/user-interaction` / `ask_user_question` elicitation path — an elicitation form is not a permission prompt, and a free-text answer is not a closed outcome; sharing provider plumbing stays open if the two ever converge.
- **Static optional injection in `dsh-tools`** — rejected: the vendored cordis `Inject` type has no optional flag — the object form maps service names to intercept config, and a declared inject gates the fiber. `ctx.get('approval')` is the documented opportunistic-consumption pattern (the `tool-bash` owner-token lookup, the loop's persistence probe), reads presence per call, and degrades correctly across HMR without extra machinery.
- **The capability-seam three-package split** — rejected: interface/implementation/consumer fits a seam whose implementation is swappable (bash-local vs bash-sandbox). Here the service body is fixed mechanism and the variable part is listeners that live with their owners — splitting would manufacture an implementation package with nothing in it ("don't split preemptively").
@@ -107,13 +107,13 @@ The implemented contract is pinned by the suites in Testing:
- `allowed-once` dispatches one action; every other outcome denies with a distinct reason, and `'never'` rejects before prompting.
- Missing, foreign, agent-less, throwing, invalid, and disconnected answer paths fail closed.
- Successful requests route by exact agent ownership and append one replayable, model-invisible audit pair; idle and pre-commit failures reject.
-- ACP ownership keeps prompts inside their session, while a deployment without the service emits no prompt or audit events.
+- ACP ownership keeps decisions inside their session, while a deployment without the service emits no request or audit events.
Costs and accepted limits:
- **Two decide-eager answerers race for the slot.** Sibling-plugin listener order is not deterministic, so the seam cannot referee competing terminal answerers — mitigated by convention (one terminal answerer per deployment; `prepend` only for decide-or-delegate gates) rather than a priority mechanism the event bus does not have.
- **Production exercise rests on one composition.** `ask` has two producer families — the hook bridges through `tools/pre-execute`, and sandbox escalation through its own gate — with the wire recorded in the sandbox example's snapshot suite, so the seam's real-world coverage is that one composition until more deployments compose it.
-- **Ownership keys on `Agent` object identity.** The answerer resolves the forward session-map record at `agent.session.id`, then requires that record to own the exact agent object; every current path hands the same object through the loop and the seams, but a future boundary that clones or proxies agents would make the bridge delegate and fail closed — safe, but silently UI-less — and would need a different ownership contract.
+- **Ownership keys on `Agent` object identity.** The answerer resolves the session-map record at `agent.session.id`, then requires that record to own the exact agent object; every current path hands the same object through the loop and the seams, but a future boundary that clones or proxies agents would make the bridge delegate and fail closed, and would need a different ownership contract.
## FAQ
@@ -123,10 +123,10 @@ Costs and accepted limits:
- **Who decides whether a call asks in the first place?** Policy producers: a hook returning `permissionDecision: ask`, any `tools/pre-execute` listener, or the sandbox escalation gate. The seam and the bridge only route and answer; neither injects its own judgment about what deserves a prompt.
- **What happens when the user dismisses the prompt, or the turn aborts mid-ask?** Dismissal maps to `cancelled` with its own deny text. An already-aborted signal settles `cancelled` without dispatching; an abort during the ask discards the late answer. When both audit appends commit, either path records one pair, never two.
- **What if the client answers with an option the harness never offered?** Any selection other than the offered `allow_once` maps to `rejected` — an unknown optionId from a non-conforming client can never grant.
-- **How do subagents' approvals route?** An agent no answerer owns delegates through the whole waterfall and fails closed — in-process subagents are deliberately unanswerable. `subagent-acp`'s child-side auto-answer is separate; routing a child's asks to the parent's editor is deferred (§ Deferred).
+- **How do subagents' approvals route?** An agent no answerer owns delegates through the whole waterfall and fails closed — in-process subagents are deliberately unanswerable. `subagent-acp`'s child-side auto-answer is separate; routing a child's asks to the parent controller is deferred (§ Deferred).
- **What does `policy: 'never'` actually change at runtime?** The service resolves every ask for that session to `rejected` before dispatching any answerer (in-service, so no registration order can bypass it); the system prompt states the policy; switches are narrated at boundaries; each successful auto-rejection records the audit pair.
-- **What happens across a hot reload, or when the UI plugin unloads mid-session?** Answerers dispose with their owning fiber, so the next ask degrades to `unavailable` instead of hanging on a dead channel; remounting re-registers the answerer with no catch-up state.
-- **Where does the user see what they are approving?** On the tool call itself: the prompt attaches to the already-streamed call via `callId` — arguments included — and adds the asker's human-readable `reason`; the request carries no argument copy of its own.
+- **What happens across a hot reload, or when an answerer unloads mid-session?** Answerers dispose with their owning fiber, so the next ask degrades to `unavailable` instead of hanging on a dead channel; remounting re-registers the answerer with no catch-up state.
+- **Where does a client get approval context?** The request carries the exact `callId` and the asker's human-readable `reason`; channel adapters may correlate richer tool-call state without duplicating arguments in the approval seam.
## Prior art
@@ -135,5 +135,5 @@ In-repo precedents this design copies or contrasts with:
- The `fs/write-intent` gate (`packages/fs/fs/`) — the documented single-occupancy decision-slot waterfall semantics (first answer wins, delegate via `next()`) the answerer contract reuses.
- `hook/invoked`/`hook/result` — the log-only audit-pair precedent `approval/asked`/`approval/decided` follows; [the hook-bridges Agent Note](2026-06-30-hook-bridges.md) ships `permissionDecision: ask`, the first producer.
- [The interception-seams Agent Note](2026-06-30-interception-seams.md) — the `tools/pre-execute` `allow`/`deny`/`ask` vocabulary whose `ask` this seam services.
-- [The ACP support Agent Note](2026-06-14-acp-agent-client-protocol.md) — the exact-agent ownership check against the forward session map that the answerer routes through; [the multi-session Agent Note](2026-06-14-acp-multi-session.md) — the per-session permission-ownership blocker this implements.
+- [The automation-only ACP Agent Note](../simplification/2026-07-23-acp-automation-only-protocol.md) — the exact-agent ownership check against the session map that the answerer routes through; [the multi-session Agent Note](2026-06-14-acp-multi-session.md) — the per-session permission-ownership blocker this implements.
- The opportunistic `ctx.get()` consumption pattern (`tool-bash`'s owner-token lookup, the loop's persistence probe) — how `dsh-tools` consumes the seam without gating its fiber on it.
diff --git a/.agents/notes/implemented/feature/2026-07-06-approval-seam.zh.md b/.agents/notes/implemented/feature/2026-07-06-approval-seam.zh.md
index 94ffb330cd..d218f79888 100644
--- a/.agents/notes/implemented/feature/2026-07-06-approval-seam.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-06-approval-seam.zh.md
@@ -6,13 +6,13 @@ Status: implemented
## 问题
-两个调用方需要向人类提出同一个问题——「这个具体操作可以继续吗?」:`tools/pre-execute` 的 `ask` 决策(包括 Claude-Code 钩子桥的 `permissionDecision: ask`)以及[沙箱 Agent Note](2026-07-06-sandbox.md) 中拒绝后的一次性升级重试。一个共享的 seam 使它们无需各自发明独立的结果词汇、UI 路由、取消机制和审计轨迹,同时保证没有 UI 的部署永远不会批准一个无法应答的请求。
+两个调用方需要同一个封闭决策——「这个具体操作可以继续吗?」:`tools/pre-execute` 的 `ask` 决策(包括 Claude-Code 钩子桥的 `permissionDecision: ask`)以及[沙箱 Agent Note](2026-07-06-sandbox.md) 中拒绝后的一次性升级重试。一个共享的 seam 使它们无需各自发明独立的结果词汇、通道路由、取消机制和审计轨迹,同时保证没有应答者的部署永远不会批准一个无法应答的请求。应答者可以是交互式宿主,也可以是自动化控制器。
-路由问题的核心是归属:审批提示必须到达拥有发起请求的 agent(智能体)的编辑器会话(ACP(Agent Client Protocol)桥在一条连接上多路复用 N 个会话),对无人拥有的 agent(进程内 subagent、测试)失败关闭,并且不侵入没有组合 UI 的部署(headless、CI)。
+路由问题的核心是归属:权限请求必须到达拥有发起请求的 agent(智能体)的通道,对无人拥有的 agent 失败关闭,并且不侵入没有组合应答者的部署。
## 决策
-一个包 `dsh-user-approval`(`packages/ui/user-approval`),拥有词汇表和 `ctx.approval` 服务——即机制。策略——谁来应答、某个会话是否需要被询问——不在其中:应答者是 `approval/request` waterfall 监听器,由拥有通道的插件注册(ACP 桥、未来的终端 UI、测试脚本),而每会话的策略层可以在任何人类介入之前做出决定。消费方(`dsh-tools` 的 ask 路由、沙箱升级门禁)将问题解析为一个封闭结果,并从中派生各自的工具结果。刻意设计为一个包,而非能力 seam 的三包拆分(见「替代方案」)。
+一个包 `dsh-user-approval`(`packages/ui/user-approval`),拥有词汇表和 `ctx.approval` 服务——即机制。策略——谁来应答、某个会话是否需要被询问——不在其中:应答者是 `approval/request` waterfall 监听器,由拥有通道的插件注册(ACP(Agent Client Protocol)桥、宿主适配器、测试脚本),而每会话的策略层可以在任何通道介入之前做出决定。消费方(`dsh-tools` 的 ask 路由和沙箱升级门禁)将问题解析为一个封闭结果,并从中派生各自的工具结果。刻意设计为一个包,而非能力 seam 的三包拆分(见「替代方案」)。
### 部署如何使用它
@@ -25,11 +25,11 @@ Status: implemented
# policy: never # deployment default for sessions without an override; 'ask' when omitted
```
-仅有这条条目只提供机制,不提供通道:没有组合应答者时,每次 ask 都解析为 `unavailable`,发起请求的工具调用被拒绝——失败关闭无需配置。组合 ACP 应用(`@deepseek-ai/dsh-acp-demo`,如 [acp-agent 示例的默认树](../../../../examples/acp-agent/README.md))即可闭环:其桥注册一个应答者,通过 `session/request_permission` 向拥有该会话的编辑器发出提示,于是钩子的 `ask` 或升级请求会以一次性 Allow/Reject 提示的形式呈现,附着在已流式输出的工具调用上。`policy: never` 是无人值守姿态:每次 ask 确定性地自动拒绝,在系统提示词中声明,无人类参与。`policy` 在插件加载时对照封闭列表校验;非法值直接抛异常。
+仅有这条条目只提供机制,不提供通道:没有组合应答者时,每次 ask 都解析为 `unavailable`,发起请求的工具调用被拒绝——失败关闭无需配置。组合 ACP 应用(`@deepseek-ai/dsh-acp-demo`,如 [acp-agent 示例的默认树](../../../../examples/acp-agent/README.md))即可闭环:其[仅面向自动化的桥接层](../simplification/2026-07-23-acp-automation-only-protocol.md)注册一个应答者,向拥有该会话的客户端发送 `session/request_permission`,携带精确的工具调用 id 和一次性 allow/reject 选项。`policy: never` 是无人值守姿态:每次 ask 确定性地自动拒绝,并在系统提示词中声明。`policy` 在插件加载时对照封闭列表校验;非法值直接抛异常。
组合部署的可观测行为:`allowed-once` 仅允许该次调用继续;拒绝、关闭和通道缺失以三种不同原因拒绝,模型可以区分;轮次内成功的请求会在发起请求的 agent 的会话日志上落一对持久的 `approval/asked`/`approval/decided` 事件;授权不会在发起请求的调用结束后继续存在。空闲时的请求或审计追加失败会拒绝,而不会返回未经审计的决策。
-以下是该组合下的一次 ask,逐字取自沙箱示例录制的 `escalation-approved` 场景——模型请求沙箱升级,门禁发起 ask,桥向拥有该会话的编辑器发出提示,用户点击 Allow once:
+以下是该组合下的一次 ask,取自沙箱示例录制的 `escalation-approved` 场景——模型请求沙箱升级,门禁发起 ask,自动化客户端选择 Allow once:
```
tool/call bash {"command": "printf 'escalated\n' > escalated.txt && cat escalated.txt",
@@ -40,12 +40,12 @@ approval/asked {"toolName": "bash", "callId": "call_00_…",
→ session/request_permission {"toolCall": {"toolCallId": "call_00_…"},
"options": [{"optionId": "allow-once", "name": "Allow once", "kind": "allow_once"},
{"optionId": "reject-once", "name": "Reject", "kind": "reject_once"}]}
- ← the user picks "Allow once" on the prompt the editor attaches to the streamed bash call
+ ← the client selects "Allow once"
approval/decided {"outcome": "allowed-once"}
tool/result "escalated" — this one call ran under the wider mode; the grant died with it
```
-`escalation-rejected` 孪生场景以 `{"outcome": "rejected"}` 结束:不执行任何操作,模型的结果携带发起方的逐字失败关闭文本(`the user rejected escalating this command to "workspace-write"`)。钩子的 `permissionDecision: ask` 走完全相同的协议;只有发起方和拒绝文本不同(§ dsh-tools 中的 Ask 路由)。在 headless 环境下,同一请求完全跳过提示,直接结算为 `unavailable`。
+`escalation-rejected` 孪生场景以 `{"outcome": "rejected"}` 结束:不执行任何操作,模型的结果携带发起方的逐字失败关闭文本(`the user rejected escalating this command to "workspace-write"`)。钩子的 `permissionDecision: ask` 走完全相同的协议;只有发起方和拒绝文本不同(§ dsh-tools 中的 Ask 路由)。没有应答者时,同一请求直接结算为 `unavailable`。
### 设计细节
@@ -55,11 +55,11 @@ tool/result "escalated" — this one call ran under the wider mode; the gra
应答者是 `approval/request` waterfall 监听器。零监听器会一路委派至 `unavailable`;识别该 agent 的监听器占用先到先得的决策槽,而不识别的监听器必须调用 `next()` 委派。监听器随其 fiber dispose,因此卸载通道会失败关闭。由于兄弟插件的注册顺序不确定,部署应组合一个终端应答者,并保留 `prepend` 给「决策或委派」门禁。
-`ApprovalRequest` 携带发起请求的 `agent`、`toolName`、可选的精确 `callId`、人类可读的 `reason` 和可选的 `signal`。它使用 `CallId` brand 而不导入依赖本 seam 的 `dsh-tools`。工具参数仍留在 UI 通过 `callId` 引用的、已流式输出的调用上。
+`ApprovalRequest` 携带发起请求的 `agent`、`toolName`、可选的精确 `callId`、人类可读的 `reason` 和可选的 `signal`。它使用 `CallId` brand 而不导入依赖本 seam 的 `dsh-tools`。通道适配器可按 `callId` 关联任何更丰富的调用状态;审批请求本身不重复携带工具参数。
#### dsh-tools 中的 Ask 路由
-`ToolRegistry.execute()` 在派发前解析 `ask`:`allowed-once` 继续执行,而拒绝、取消和通道不可用产生三种不同的拒绝原因。机会性消费 `ctx.get('approval')`,让缺失或未挂载的服务失败关闭而不阻塞注册表 fiber。无 agent 的执行同样失败关闭,因为它既没有审计会话,也没有 UI 所有者。
+`ToolRegistry.execute()` 在派发前解析 `ask`:`allowed-once` 继续执行,而拒绝、取消和通道不可用产生三种不同的拒绝原因。机会性消费 `ctx.get('approval')`,让缺失或未挂载的服务失败关闭而不阻塞注册表 fiber。无 agent 的执行同样失败关闭,因为它既没有审计会话,也没有通道所有者。
#### 每会话策略层
@@ -67,9 +67,9 @@ seam 还拥有[沙箱 Agent Note](2026-07-06-sandbox.md) 所描述的会话级 `
#### ACP 应答者
-ACP 桥只应答其正向会话映射所拥有的精确 agent 对象。它把 `session/request_permission` 附着到既有 `callId`,声明一次性的 allow/reject 选项,单独映射取消,并且绝不批准未知选项。外部或无调用标识的请求会委派;客户端 RPC 失败变为 `unavailable`。钩子和 `tools/pre-execute` 决定一次调用是否需要询问。
+ACP 桥只应答其会话映射所拥有的精确 agent 对象。它携带既有 `callId` 发送 `session/request_permission`,声明一次性的 allow/reject 选项,单独映射取消,并且绝不批准未知选项。外部或无调用标识的请求会委派;客户端 RPC 失败变为 `unavailable`。钩子和 `tools/pre-execute` 决定一次调用是否需要询问。该通道是自动化客户端与其 agent 之间的机器策略,不是 ACP 展示层。
-应答者通过 [ACP 支持 Agent Note](2026-06-14-acp-agent-client-protocol.md) 描述的桥精确 agent 归属检查进行路由,实现了[多会话 Agent Note](2026-06-14-acp-multi-session.md) 要求的每会话权限归属。
+应答者通过[仅面向自动化的 ACP Agent Note](../simplification/2026-07-23-acp-automation-only-protocol.md)描述的桥精确 agent 归属检查进行路由,保留了[多会话 Agent Note](2026-06-14-acp-multi-session.md) 要求的每会话权限归属。
#### 审计,以及模型看到什么
@@ -88,13 +88,13 @@ ACP 桥只应答其正向会话映射所拥有的精确 agent 对象。它把 `s
## 延后
- **`allow_always` 授权存储**:兑现持久授权意味着设计存储、作用域标识(调用?路径?前缀?会话?时间窗口?)和撤销;在设计完成之前,只展示一次性选项([沙箱 Agent Note](2026-07-06-sandbox.md) § Escalation 记录了开放的作用域问题)。
-- **通过组合应答者录制由钩子驱动的 `ask`**:人类提示协议格式已通过沙箱示例的升级分支录制。钩子矩阵中的 `hook-cc-pretool-ask` 固定无 ApprovalService 时的后备拒绝,而钩子生产者与应答者的组合仍留在单元测试层。
-- **将子 agent 的审批路由到父会话**:`subagent-acp` 的子侧自动应答自己的 `permission` 请求;将其呈现给父会话的编辑器是独立的设计。
+- **通过组合应答者录制由钩子驱动的 `ask`**:权限协议格式已通过沙箱示例的升级分支录制。钩子矩阵中的 `hook-cc-pretool-ask` 固定无 ApprovalService 时的后备拒绝,而钩子生产者与应答者的组合仍留在单元测试层。
+- **将子 agent 的审批路由到父会话**:`subagent-acp` 的子侧自动应答自己的权限请求;将其委派给父控制器是独立的设计。
## 曾考虑的替代方案
- **单一注册提供方而非 waterfall 监听器**:否决。`registerProvider()` 接口迫使所有组合问题——允许列表预过滤、外部钩子决策者、脚本化测试应答、人类前面的策略门禁——都塞进一个提供方实现。waterfall 从运行时已有的机制中获得组合能力、缺失时失败关闭和 HMR(热模块替换) dispose(资源释放);seam 的 JSDoc 以约定固定单决策槽语义,而非发明一个提供方注册表。
-- **在 ACP 桥中内联 `tools/pre-execute` 权限门禁**:否决。对桥拥有的每次调用都弹出提示,会将请求策略硬编码进 UI 插件,无法服务第二个发起方(沙箱升级发生在执行开始之后,没有 pre-execute 时刻),且钩子产生的 `ask` 决策没有共享机制。
+- **在 ACP 桥中内联 `tools/pre-execute` 权限门禁**:否决。对桥拥有的每次调用都弹出提示,会将请求策略硬编码进传输层,无法服务第二个发起方(沙箱升级发生在执行开始之后,没有 pre-execute 时刻),且钩子产生的 `ask` 决策没有共享机制。
- **通用用户交互 seam(`ctx.userInteraction`)**:否决作为审批机制。二者骨架相似(按 agent 路由、阻塞等待人类、处理缺失),但审批的契约在每个关键维度上都更窄:封闭的结果词汇而非自由文本、附着在工具调用上的协议原生提示而非通用表单、强制的缺失时失败关闭、以及审计事件。因此审批不走已交付的 `packages/ui/user-interaction` / `ask_user_question` 引出路径——引出表单不是权限提示,自由文本应答不是封闭结果;如果二者将来趋同,共享提供方管道仍然开放。
- **`dsh-tools` 中的静态可选注入**:否决。vendor 的 Cordis `Inject` 类型没有 optional 标志——对象形式将服务名映射到拦截配置,声明的 inject 会阻塞 fiber。`ctx.get('approval')` 是文档化的机会性消费模式(`tool-bash` 的 owner-token 查找、loop 的持久化探测),按调用读取存在性,跨 HMR 正确降级,无需额外机制。
- **能力 seam 的三包拆分**:否决。接口/实现/消费方适合实现可替换的 seam(bash-local vs bash-sandbox)。此处服务体是固定机制,可变部分是留在各自通道拥有者插件中的监听器——拆分只会制造一个空的实现包(「不要预防性拆分」)。
@@ -107,13 +107,13 @@ ACP 桥只应答其正向会话映射所拥有的精确 agent 对象。它把 `s
- `allowed-once` 派发一次操作;其他所有结果都以不同原因拒绝,而 `'never'` 会在提示前拒绝。
- 缺失、外部、无 agent、抛异常、无效或断开连接的应答路径都会失败关闭。
- 成功的请求按精确 agent 归属路由,并追加一对可回放、对模型不可见的审计事件;空闲时和提交前失败的请求会拒绝。
-- ACP 归属把提示限制在其会话内,而没有该服务的部署不产生提示或审计事件。
+- ACP 归属把决策限制在其会话内,而没有该服务的部署不产生请求或审计事件。
代价与已接受的局限:
- **两个急于决策的应答者竞争同一槽位。** 兄弟插件的监听器顺序不确定,seam 无法仲裁竞争的终端应答者。通过约定缓解(每个部署一个终端应答者;仅对「先决策或委派」门禁使用 `prepend`),而非事件总线不具备的优先级机制。
- **生产环境验证依赖单一组合。** `ask` 有两个生产者家族——钩子桥通过 `tools/pre-execute`,沙箱升级通过自己的门禁——协议格式录制在沙箱示例的快照套件中;因此在更多部署组合它之前,seam 的真实覆盖面就是这一种组合。
-- **归属以 `Agent` 对象标识为键。** 应答者先在 `agent.session.id` 处解析正向会话映射记录,再要求该记录拥有精确的 agent 对象;当前所有路径在 loop 和各 seam 之间传递同一对象,但未来如果某个边界克隆或代理了 agent,桥会委派并失败关闭——安全,但静默无 UI——届时需要另一种归属契约。
+- **归属以 `Agent` 对象标识为键。** 应答者先在 `agent.session.id` 处解析会话映射记录,再要求该记录拥有精确的 agent 对象;当前所有路径在 loop 和各 seam 之间传递同一对象,但未来如果某个边界克隆或代理了 agent,桥会委派并失败关闭,届时需要另一种归属契约。
## FAQ
@@ -123,10 +123,10 @@ ACP 桥只应答其正向会话映射所拥有的精确 agent 对象。它把 `s
- **谁决定一次调用是否需要 ask?** 策略生产者:返回 `permissionDecision: ask` 的钩子、任何 `tools/pre-execute` 监听器、或沙箱升级门禁。seam 和桥只负责路由和应答;二者都不注入自己对「什么值得弹出提示」的判断。
- **用户关闭提示或轮次在 ask 进行中中止时会发生什么?** 关闭映射为 `cancelled` 并携带自己的拒绝文本。已中止的 signal 直接结算为 `cancelled` 而不派发;ask 进行中的中止丢弃迟到的应答。当两个审计追加都提交时,任一路径都记录恰好一对事件,绝不会两对。
- **如果客户端以 harness 从未提供的选项应答呢?** 除已提供的 `allow_once` 之外的任何选项都映射为 `rejected`——来自不合规客户端的未知 optionId 永远不能授权。
-- **subagent 的审批如何路由?** 没有应答者拥有的 agent 穿过整个 waterfall 委派并失败关闭——进程内 subagent 被刻意设计为不可应答。`subagent-acp` 的子侧自动应答是独立的;将子 agent 的 ask 路由到父会话的编辑器已延后(§ 延后)。
+- **subagent 的审批如何路由?** 没有应答者拥有的 agent 穿过整个 waterfall 委派并失败关闭——进程内 subagent 被刻意设计为不可应答。`subagent-acp` 的子侧自动应答是独立的;将子 agent 的 ask 路由到父控制器已延后(§ 延后)。
- **`policy: 'never'` 在运行时实际改变了什么?** 服务在派发任何应答者之前,将该会话的每次 ask 解析为 `rejected`(在服务内部,因此没有注册顺序能绕过它);系统提示词声明该策略;切换在边界处被叙述;每次成功的自动拒绝都会记录审计对。
-- **热重载或 UI 插件在会话中途卸载时会发生什么?** 应答者随其拥有的 fiber 一起 dispose,因此下一次 ask 降级为 `unavailable` 而非挂在死通道上;重新挂载会重新注册应答者,无需追赶状态。
-- **用户在哪里看到自己在批准什么?** 在工具调用本身:提示通过 `callId` 附着在已流式输出的调用上(包含参数),并添加发起方的人类可读 `reason`;请求本身不携带参数副本。
+- **热重载或应答者在会话中途卸载时会发生什么?** 应答者随其拥有的 fiber 一起 dispose,因此下一次 ask 降级为 `unavailable` 而非挂在死通道上;重新挂载会重新注册应答者,无需追赶状态。
+- **客户端从哪里获得审批上下文?** 请求携带精确的 `callId` 和发起方的人类可读 `reason`;通道适配器可自行关联更丰富的工具调用状态,而无需在审批 seam 中重复携带参数。
## 先例
@@ -135,5 +135,5 @@ ACP 桥只应答其正向会话映射所拥有的精确 agent 对象。它把 `s
- `fs/write-intent` 门禁(`packages/fs/fs/`)——文档化的单占用决策槽 waterfall 语义(先到先得,通过 `next()` 委派),应答者契约复用了它。
- `hook/invoked`/`hook/result`——仅日志审计对先例,`approval/asked`/`approval/decided` 沿用了它;[钩子桥 Agent Note](2026-06-30-hook-bridges.md) 交付了 `permissionDecision: ask`,即第一个生产者。
- [拦截 seam Agent Note](2026-06-30-interception-seams.md)——`tools/pre-execute` 的 `allow`/`deny`/`ask` 词汇,本 seam 服务其中的 `ask`。
-- [ACP 支持 Agent Note](2026-06-14-acp-agent-client-protocol.md)——应答者路由时对正向会话映射执行的精确 agent 归属检查;[多会话 Agent Note](2026-06-14-acp-multi-session.md)——本设计实现的每会话权限归属阻塞项。
+- [仅面向自动化的 ACP Agent Note](../simplification/2026-07-23-acp-automation-only-protocol.md)——应答者路由时对会话映射执行的精确 agent 归属检查;[多会话 Agent Note](2026-06-14-acp-multi-session.md)——本设计实现的每会话权限归属阻塞项。
- 机会性 `ctx.get()` 消费模式(`tool-bash` 的 owner-token 查找、loop 的持久化探测)——`dsh-tools` 消费该 seam 而不阻塞其 fiber 的方式。
diff --git a/.agents/notes/implemented/feature/2026-07-06-sandbox.i18n.yaml b/.agents/notes/implemented/feature/2026-07-06-sandbox.i18n.yaml
index 7cb853744e..6437d7813d 100644
--- a/.agents/notes/implemented/feature/2026-07-06-sandbox.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-06-sandbox.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
-2026-07-06-sandbox.md: a63779d1848b5a77bfceeaea51f5103084b9fce3
-2026-07-06-sandbox.zh.md: 2511278b1424f01c8a54f8a9a9b1874888bb43b3
+2026-07-06-sandbox.md: 723ef170188dc11da24e049a1e2838fb240d0a17
+2026-07-06-sandbox.zh.md: a8c7743bb3d499fb58f507ea2c202b44efe2311d
diff --git a/.agents/notes/implemented/feature/2026-07-06-sandbox.md b/.agents/notes/implemented/feature/2026-07-06-sandbox.md
index a63779d184..1200b26606 100644
--- a/.agents/notes/implemented/feature/2026-07-06-sandbox.md
+++ b/.agents/notes/implemented/feature/2026-07-06-sandbox.md
@@ -10,7 +10,7 @@ A coding agent needs this product path: bash subprocesses — and the hook comma
The harness is an SDK, so confinement must be a capability developers COMPOSE: whether to sandbox, and which backend per platform, belongs in the leaf `cordis.yml` as a first-class entry — not inside one executor's private machinery. And the first-choice runner, `bwrap`, is unusable on exactly the hosts a sandbox matters most (minimal containers, disabled unprivileged userns, LSMs that deny `mount`), so a fallback runner has to ship with the SDK rather than be assumed on the host.
-Confinement alone leaves two gaps. A denial with no escalation path is terminal — the model can only give up, which pressure-cooks operators into configuring `workspace-write` or `danger-full-access` globally and defeats the sandbox. And the model-visible knobs (the sandbox mode, the approval policy) change over an agent's lifetime — an ACP user flips a per-session setting, an operator edits `cordis.yml` while the process is down — while the model must never act on a stale belief about them: what IS the state on every request, what changed while the agent lives, and what changed while nobody was watching all need answers.
+Confinement alone leaves two gaps. A denial with no escalation path is terminal — the model can only give up, which pressure-cooks operators into configuring `workspace-write` or `danger-full-access` globally and defeats the sandbox. The sandbox mode and approval policy can also change over an agent's lifetime through deployment config or an optional UI policy control; execution and model-visible policy must derive from the same logged state.
## Decision
@@ -40,13 +40,13 @@ The swap is invisible to every consumer of `ctx.bash`: the bash tools, hook comm
Misconfiguration fails loud: `mode` outside the closed vocabulary is rejected at plugin load, and a host with no usable backend throws the structured `SANDBOX_UNAVAILABLE` — at `confine()` before the command ever spawns — rather than degrading to unconfined execution. `runnerCommand` on `dsh-sandbox-local` is the operator's explicit assertion of a bwrap-compatible runner (chain and probes skipped); it doubles as the deterministic fake-runner seam for keyless tests.
-Denied file effects return a `[sandbox: file access denied under mode]` marker and instructions not to work around the denial. A confining executor adds paired `sandbox_permissions` and `justification` fields for one approved retry that must be strictly wider than the session's effective mode. A grant widens only that retry; rejection executes nothing, returns `the user rejected escalating this command to ""`, and permits no re-ask. The prompt does not announce sandbox mode, avoiding preemptive refusal. When `dsh-permission` is composed, ACP exposes one `Permissions` select whose presets write both knob events; unmatched knobs appear as switch-away-only `custom`. Only a switch to the deterministic `'never'` approval policy is stated in the prompt and narrated.
+Denied file effects return a `[sandbox: file access denied under mode]` marker and instructions not to work around the denial. A confining executor adds paired `sandbox_permissions` and `justification` fields for one approved retry that must be strictly wider than the session's effective mode. A grant widens only that retry; rejection executes nothing, returns `the user rejected escalating this command to ""`, and permits no re-ask. The prompt does not announce sandbox mode, avoiding preemptive refusal. When `dsh-permission` is composed with a UI adapter, one preset selects both knob values; unmatched values fold to `custom`. The [ACP automation composition](../../../../examples/acp-agent/README.md) does not mount that UI service and selects its deployment mode explicitly.
### Design detail
#### Scope grounding
-OS subprocess confinement applies to the bash executor, including hook commands, and later to ACP subagent children. Filesystem, web, and other tools execute in-process and require policy at their own seams; an argv wrapper cannot confine a function closing over `ctx`. The existing bash request/spec split carries per-call overrides, while `tools/pre-execute` and the approval seam own the human decision.
+OS subprocess confinement applies to the bash executor, including hook commands, and later to ACP subagent children. Filesystem, web, and other tools execute in-process and require policy at their own seams; an argv wrapper cannot confine a function closing over `ctx`. The existing bash request/spec split carries per-call overrides, while `tools/pre-execute` and the approval seam own the one-shot policy decision.
#### The seam: `ctx.sandbox`
@@ -92,7 +92,7 @@ Left open: what a durable grant's scope identity is beyond the sandbox mode —
effective(session) = findLast(the session's own knob events)?.value ?? the composition-config default
```
-The default is composition config (`cordis.yml`) — operator-owned, process-wide. A runtime switch is a SESSION-SCOPED override recorded as one log-only event in that session's own log. Restart immunity (resuming a session replays its log, so overrides come back with zero catch-up machinery) and multi-session isolation (one editor tab's `workspace-write` cannot disturb another's `read-only`) both fall out by construction, and no external config store exists anywhere.
+The default is composition config (`cordis.yml`) — operator-owned, process-wide. A runtime switch is a session-scoped override recorded as one log-only event in that session's own log. Restart immunity (resuming a session replays its log, so overrides come back with zero catch-up machinery) and multi-session isolation both fall out by construction, and no external config store exists anywhere.
**One event per knob, owned by its domain** — the merge-extensible `SessionEventMap` idiom every existing event family already follows (`approval/*` in `dsh-user-approval`, `hook/*` in the hooks packages):
@@ -107,9 +107,9 @@ Each owner exports the same three-piece kit: the event declaration, a pure fold
Sandbox mode is not narrated in the prompt; denial results report the mode when it matters, avoiding preemptive refusal based on a standing label. Approval policy is different: only `'never'` is stated because automatic rejection otherwise looks like a user decision. Policy-change notices are coalesced and delivered by the next pre-step, with log-derived fallback after restart. The notice source is inferred from event position: a knob event after the last request header is user-driven; unlogged drift is operator or config driven.
-**The editor surface** is protocol-native [Session Config Options](https://agentclientprotocol.com/protocol/session-config-options) — the spec's replacement for session modes (slated for removal in ACP v2), already SDK-typed. When `ctx.permission` is composed, the bridge advertises one `permission` select (category `mode`) in `session/new` and `session/load`; its options are the deployment's preset table, and its `currentValue` is `PermissionService.current()` over the session log plus composition defaults. The shipped `workspace-write` and `danger-full-access` presets each bundle a sandbox mode with an approval policy and write through to both domain setters; a knob combination outside the table is reported as switch-away-only `custom`. `session/set_config_option` validates and switches through the permission service, then returns the complete refreshed state (the spec contract).
+**The optional UI surface** is `PermissionService`: a deployment-defined preset table whose entries bundle one sandbox mode with one approval policy. The shipped `workspace-write` and `danger-full-access` presets write through to both domain setters; a knob combination outside the table is reported as `custom`. UI adapters may expose that table as a selector. The automation-only ACP transport advertises no configuration selector and mounts no permission-preset service.
-**Turn enclosure is the commit boundary.** A switch during an open turn appends immediately. An idle switch remains pending on the bridge record and is appended at the next prompt submission, before assembly or execution; last write wins per knob. Openness comes from log boundaries rather than `agent.status`, and setters do not append from inside a `session/event` listener because that would reorder later observers. Until anchoring, responses overlay the pending value. A crash discards it, and reload returns the durable fold.
+**The committed event is the commit boundary.** A runtime switch records its preset and changed knob events on the target session, and every later capability resolution folds the last values. Adapters own choosing a valid session append boundary; the ACP transport has no runtime switch path. (The former ACP idle-switch anchoring — holding a pending idle selection until the next prompt submission — left with that bridge.)
#### In-process tools
@@ -117,10 +117,10 @@ fs/web/todo execute in-process, so their sandbox semantics are policy at their s
### Testing
-- **Unit:** pin platform selection and profiles, fail-closed runner classification, per-call mode/root resolution, per-process facts, escalation validation and outcomes, permission preset folding and write-through, narrator coalescing, ACP advertisement and validation, and turn-enclosed config writes.
-- **Keyless real-runner:** exercise bwrap, Landlock, and Seatbelt against real filesystem effects at provider and bash-consumer layers; one real Cordis context concurrently drives two project sessions through shipped bash and fs tools, proving own-root success and sibling-root denial. Packed-install coverage proves the registry launcher remains executable. The real ACP composition pins permission switching and rejects unknown presets. CI rejects a silent all-skip.
+- **Unit:** pin platform selection and profiles, fail-closed runner classification, per-call mode/root resolution, per-process facts, escalation validation and outcomes, permission preset folding and write-through, and narrator coalescing.
+- **Keyless real-runner:** exercise bwrap, Landlock, and Seatbelt against real filesystem effects at provider and bash-consumer layers; one real Cordis context concurrently drives two project sessions through shipped bash and fs tools, proving own-root success and sibling-root denial. Packed-install coverage proves the registry launcher remains executable. CI rejects a silent all-skip.
- **With-key:** start the real ACP composition in read-only mode, let a model-driven bash write hit the runner's denial marker, then drive the bridge answerer and disk effect through granted and rejected workspace-write retries; unavailable credentials or runners self-skip.
-- **Snapshot:** pin the permission config-option wire, preset and knob events, prompt deltas and notices, and both scripted approval branches. A real ACP example scenario places its session under the user home while the deployment fallback points at `/tmp`, then pins a successful workspace-write mutation; this distinguishes session-root resolution from the process fallback without depending on runner-specific denial text. Other snapshots start unconfined so unrelated fixtures remain platform-independent, and policy scenarios switch explicitly.
+- **Snapshot:** pin prompt deltas and notices plus both scripted approval branches. A real ACP example scenario places its session under the user home while the deployment fallback points at `/tmp`, then pins a successful deployment-selected workspace-write mutation; this distinguishes session-root resolution from the process fallback without depending on runner-specific denial text. Other snapshots start unconfined so unrelated fixtures remain platform-independent.
## Deferred phases
@@ -152,7 +152,7 @@ Each phase gets its full design when picked up, validated against the code at th
- **Narrate via `agent/user-message` + a bus event** — rejected: it presupposes a turn-entry seam that does not exist (the real seam is `agent/prompt-submit`), and pre-step's position serves both the coalesced turn-entry notice and the mid-turn immediacy bound with one listener.
- **A standing prompt statement of the sandbox mode (+ a switch narrator)** — shipped first, then removed on live evidence: with `Bash commands run under the "read-only" file sandbox.` in every request, the model refused to ATTEMPT denied-then-escalatable work (five of twelve turns in the first manual session ended with zero tool calls), turning the sandbox into a soft lockout. The denial marker names the mode at the moment it matters and the escalation fields carry the recovery; the approval knob keeps its statement because an auto-rejection is behaviorally indistinguishable from a human "no".
- **Track "last told" with its own bookkeeping events** — rejected: the `request/header` fold already records the exact prompt the model saw; parsing the closed candidate sentences back replaces a second bookkeeping stream — events are needed only where they ARE the store.
-- **ACP session modes instead of config options** — rejected: the preset is already one deployment-defined config-option select, and modes are slated for removal in ACP v2.
+- **Independent sandbox and approval selectors** — rejected: one deployment-defined permission preset keeps the two policy knobs coherent for UI clients that expose runtime switching.
## Consequences
@@ -161,11 +161,11 @@ What shipped pins — the tiers in Testing hold each:
- A denied command retried with `sandbox_permissions` + `justification` prompts the user through the composed answerer chain; a grant runs THAT call under the wider mode (result facts say so) while every other call keeps its own effective mode; every non-grant outcome produces its distinct error text and executes nothing.
- The escalation fields exist exactly when the mounted executor confines; a request that is not strictly wider than the call's effective mode fails closed with its own text and prompts no one; a deployment with no ApprovalService fails escalating calls closed and leaves plain calls untouched.
- The system prompt never states the sandbox mode (an approval `'never'` policy is the one stated knob), and the whole exchange — headers, knob events, notices, approvals, results — reconstructs from the session log alone, with no event types beyond the two knob events.
-- N idle-time flips produce at most one anchored event per knob (a net-zero sequence anchors none — a no-op push from a client echoing current selections records nothing); an approval-policy switch is narrated in at most one coalesced notice; a mid-turn sandbox switch is honored by the next call's stamp.
-- A resumed session's overrides apply and are reported to the editor with no special-casing; a default changed while the process was down is narrated before the session's first new request, attributed to the operator.
-- Two concurrent sessions never see each other's state, notices, or config options.
+- One preset selection records only changed knob values, while a no-op selection records nothing; an approval-policy switch is narrated in at most one coalesced notice, and a committed sandbox switch is honored by the next call's stamp.
+- A resumed session's overrides apply with no catch-up state; a default changed while the process was down is narrated before the session's first new request, attributed to the operator.
+- Two concurrent sessions never see each other's state or notices.
- Two concurrent project sessions in one Cordis context resolve independent workspace roots; bash and fs writes succeed inside the calling session's cwd and fail against its neighbor's cwd.
-- `agent-loop` is untouched — everything rides `systemPrompt.section`, `SessionEventMap` merging, `agent.inject()`, `agent/step`, `agent/prompt-submit`, and the ACP handler surface.
+- `agent-loop` is untouched — everything rides `systemPrompt.section`, `SessionEventMap` merging, `agent.inject()`, `agent/step`, `agent/prompt-submit`, and capability-owned policy resolution.
Costs and accepted limits:
@@ -178,7 +178,6 @@ Costs and accepted limits:
- **The model may over-ask.** Escalating without denial grounding, or picking `danger-full-access` where `workspace-write` suffices: the description steers and the enum forces the ladder, but the human prompt is the actual gate; the `approval/asked` reasons make over-asking auditable, and a `prepend` policy answerer can auto-reject patterns a deployment never wants.
- **The advertised target set is static while the effective mode is per-session** (schemas are registry-global) — a session already at the widest mode is still offered the fields. Harmless by construction: the strict-wider check at execution, not the enum, is the safety boundary — a non-widening request fails with its own text and never prompts anyone.
- **A granted escalation is not a working sandbox.** An unavailable backend still fails closed even for a granted escalation to a confining mode — at `confine()` when the platform has no chain or every probe fails, at execution when an unprobed sole runner refuses (classified as a sandbox failure, not a command failure) — while a granted `danger-full-access` run never touches the provider at all: there the grant, not the probe, is the authority.
-- **An idle switch lives in bridge memory until the next prompt submission anchors it.** A crash in that window reverts it (reported on `session/load`), and a session that never submits another prompt never persists it — accepted, with a loop-owned idle commit turn left as future work if durability becomes required.
- **The approval narrator's restart baseline parses prompt prose.** The closed candidate sentence is owned by the writing module itself, so a wording change is a coordinated writer+parser edit in one file; a session whose headers predate the section silently adopts the current policy without a notice.
- **The approval section is still a dynamic prompt surface** (a `'never'` switch breaks provider prompt-prefix caching for that session). Accepted: policy switches are rare, and a model acting on a stale `'never'` is worse. The sandbox knob no longer touches the prompt at all.
- **The model may hold a stale belief about the sandbox mode** (nothing announces a switch). Accepted deliberately: the next attempt's marker or success corrects it, and the observed failure mode of announcing — preemptive refusal — is worse than one wasted retry.
@@ -192,7 +191,7 @@ Costs and accepted limits:
- **Does the sandbox restrict network or process visibility?** No — `SandboxMode` claims FILE effects only; the bwrap profile deliberately does not unshare pid, and no backend claims network. Whether network restriction becomes its own knob is left open in § The seam.
- **Which tools actually run confined?** OS subprocesses through `ctx.bash` — the bash tools, and hook commands transitively — plus the filesystem tools (`read`/`write`/`edit`) through the sandboxed `ctx.fs` provider (the [cross-family fs sandbox RFC](2026-07-14-cross-family-fs-sandbox.md)): bash confines via the OS runner, fs via an in-process path fence, both keying off the same `ctx.sandboxPolicy` mode. web/todo stay in-process and unfenced (web's only effect is network, outside the file-effect mode vocabulary).
- **Does a granted escalation persist?** No. The grant is consumed by the exact foreground or background call that asked; every neighboring call keeps its own effective mode. A later background denial surfaces through `task_output` and may ground a new exact-command retry.
-- **When does an editor's mode switch take effect?** Mid-turn: appended immediately, honored by the very next call's stamp. Idle: held on the bridge's session record, anchored at the next `agent/step` inside its open turn, with N flips coalescing to at most one event (none if net-zero); a crash before anchoring reverts it and `session/load` reports the truth. The model is not told — its next command simply behaves under the new mode.
+- **When does a runtime mode switch take effect?** Once its session event commits, the very next capability resolution folds and stamps the new mode. The model is not told a standing mode; its next command simply behaves under the new policy, and any denial names that policy at the point of use.
- **What survives a restart — and what if the operator changed the config default while the process was down?** Overrides replay from the session log (`effective = fold ?? config`), so a resumed session keeps its modes with zero catch-up machinery; a default that drifted offline changes behavior the same way a switch does (the approval policy, being stated, is additionally narrated with operator/config attribution).
- **What does `enforcement: 'partial'` on a result mean?** The selected backend enforces the subset its kernel ABI governs — e.g. Landlock before ABI v3 does not govern path truncate — and says so structurally instead of refusing the host; the probe's report line distinguishes the cases. The bwrap and Seatbelt profiles govern every promised file effect by construction, so they always report `full`.
diff --git a/.agents/notes/implemented/feature/2026-07-06-sandbox.zh.md b/.agents/notes/implemented/feature/2026-07-06-sandbox.zh.md
index 2511278b14..b6a9b831d1 100644
--- a/.agents/notes/implemented/feature/2026-07-06-sandbox.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-06-sandbox.zh.md
@@ -10,7 +10,7 @@ Status: implemented
harness 是一个 SDK,因此约束必须是开发者可组合的能力:是否启用沙箱、每个平台使用哪个后端,都应作为一等条目写在叶子 `cordis.yml` 中,而非藏在某个执行器的私有机制里。而首选 runner `bwrap` 恰恰在沙箱最重要的主机上不可用(精简容器、禁用了非特权 userns、LSM 拒绝 `mount`),因此备选 runner 必须随 SDK 一起交付,而不能假设主机已有。
-仅有约束还留下两个缺口。拒绝后没有升级路径就是死路:模型只能放弃,这会迫使运维人员全局配置 `workspace-write` 或 `danger-full-access`,从而使沙箱形同虚设。而模型可见的旋钮(沙箱模式、批准策略)在 agent 生命周期内会变化——ACP 用户切换按会话设置、运维人员在进程停止期间编辑 `cordis.yml`——模型绝不能基于过时的信念行动:每次请求时的实际状态是什么、agent 存活期间发生了什么变化、无人看管时又发生了什么变化,都需要有明确答案。
+仅有约束还留下两个缺口。拒绝后没有升级路径就是死路:模型只能放弃,这会迫使运维人员全局配置 `workspace-write` 或 `danger-full-access`,从而使沙箱形同虚设。而沙箱模式和批准策略也会通过部署配置或可选的 UI 策略控件在 agent 生命周期内变化;执行与模型可见的策略必须派生自同一份已记录的状态。
## 决策
@@ -40,13 +40,13 @@ harness 是一个 SDK,因此约束必须是开发者可组合的能力:是
配置错误大声失败:`mode` 不在封闭词汇中时在插件加载时被拒绝;主机上没有可用后端时在 `confine()` 阶段(命令 spawn 之前)抛出结构化的 `SANDBOX_UNAVAILABLE`,而非降级为无约束执行。`dsh-sandbox-local` 上的 `runnerCommand` 是运维人员对一个 bwrap 兼容 runner 的显式断言(跳过链和探测);它同时充当 keyless 测试的确定性 fake-runner seam。
-被拒绝的文件操作返回 `[sandbox: file access denied under mode]` 标记,并附带不要绕过拒绝的指令。约束执行器添加配对的 `sandbox_permissions` 和 `justification` 字段,用于一次经批准的重试,该重试必须严格宽于会话的有效模式。授权仅放宽该次重试;拒绝则不执行任何内容,返回 `the user rejected escalating this command to ""`,且不允许再次请求。提示词不声明沙箱模式,以避免基于常驻标签的预防性拒绝。当 `dsh-permission` 被组合时,ACP 暴露一个 `Permissions` 选择器,其 preset 同时写入两个旋钮事件;不匹配的旋钮组合显示为仅可切换离开的 `custom`。只有切换到确定性的 `'never'` 批准策略才会在提示词中声明并叙述。
+被拒绝的文件操作返回 `[sandbox: file access denied under mode]` 标记,并附带不要绕过拒绝的指令。约束执行器添加配对的 `sandbox_permissions` 和 `justification` 字段,用于一次经批准的重试,该重试必须严格宽于会话的有效模式。授权仅放宽该次重试;拒绝则不执行任何内容,返回 `the user rejected escalating this command to ""`,且不允许再次请求。提示词不声明沙箱模式,以避免基于常驻标签的预防性拒绝。当 `dsh-permission` 与某个 UI 适配器一起组合时,一个 preset 同时选定两个旋钮值;不匹配的组合折叠为 `custom`。[ACP 自动化组合](../../../../examples/acp-agent/README.md)不挂载该 UI 服务,而是显式选定其部署模式。
### 设计细节
#### 范围界定
-OS 子进程约束适用于 bash 执行器(包括钩子命令),后续还将适用于 ACP subagent 子进程。文件系统、web 和其他工具在进程内执行,需要在各自的 seam 层面实施策略;argv 包装无法约束一个闭包了 `ctx` 的函数。既有的 bash request/spec 拆分承载按调用的覆盖,而 `tools/pre-execute` 和 approval seam 负责人的决策。
+OS 子进程约束适用于 bash 执行器(包括钩子命令),后续还将适用于 ACP subagent 子进程。文件系统、web 和其他工具在进程内执行,需要在各自的 seam 层面实施策略;argv 包装无法约束一个闭包了 `ctx` 的函数。既有的 bash request/spec 拆分承载按调用的覆盖,而 `tools/pre-execute` 和 approval seam 负责一次性策略决策。
#### seam:`ctx.sandbox`
@@ -92,7 +92,7 @@ Landlock launcher 源码和包工作区位于 `native/landlock-run`,与 harnes
effective(session) = findLast(the session's own knob events)?.value ?? the composition-config default
```
-默认值是组合配置(`cordis.yml`)——运维人员拥有,进程范围。运行时切换是会话范围的覆盖,记录为该会话自身日志中的一条仅日志事件。重启免疫(恢复会话时回放其日志,覆盖自然恢复,无需追赶机制)和多会话隔离(一个编辑器标签页的 `workspace-write` 不会干扰另一个的 `read-only`)都是构造性的自然结果,且不存在任何外部配置存储。
+默认值是组合配置(`cordis.yml`)——运维人员拥有,进程范围。运行时切换是会话范围的覆盖,记录为该会话自身日志中的一条仅日志事件。重启免疫(恢复会话时回放其日志,覆盖自然恢复,无需追赶机制)和多会话隔离都是构造性的自然结果,且不存在任何外部配置存储。
**每个旋钮一种事件,由其领域拥有**——这是每个既有事件族已遵循的可合并扩展 `SessionEventMap` 惯用法(`dsh-user-approval` 中的 `approval/*`、hooks 包中的 `hook/*`):
@@ -107,9 +107,9 @@ interface SessionEventMap {
沙箱模式不在提示词中叙述;拒绝结果在需要时报告模式,避免基于常驻标签的预防性拒绝。批准策略不同:只有 `'never'` 被声明,因为自动拒绝在行为上与用户的「不」无法区分。策略变更通知被合并,由下一个步骤前检查点递送,重启后有基于日志的回退。通知来源从事件位置推断:最后一个 request header 之后的旋钮事件是用户驱动的;未记录的漂移是运维人员或配置驱动的。
-**编辑器界面**是协议原生的[会话配置选项](https://agentclientprotocol.com/protocol/session-config-options)——该规范对 session modes 的替代(计划在 ACP v2 中移除),已有 SDK 类型。当 `ctx.permission` 被组合时,bridge 在 `session/new` 和 `session/load` 中公布一个 `permission` 选择器(category `mode`);其选项是部署的 preset 表,其 `currentValue` 是 `PermissionService.current()` 对会话日志加组合默认值的结果。随附的 `workspace-write` 和 `danger-full-access` preset 各自捆绑一个沙箱模式与一个批准策略,并写入两个领域 setter;preset 表之外的旋钮组合报告为仅可切换离开的 `custom`。`session/set_config_option` 通过权限服务验证并切换,然后返回完整的刷新状态(规范契约)。
+**可选的 UI 界面**是 `PermissionService`:一张部署定义的 preset 表,每个条目捆绑一个沙箱模式与一个批准策略。随附的 `workspace-write` 和 `danger-full-access` preset 写入两个领域 setter;preset 表之外的旋钮组合报告为 `custom`。UI 适配器可以把该表暴露为选择器。仅面向自动化的 ACP 传输层不公布任何配置选择器,也不挂载权限 preset 服务。
-**轮次封闭是提交边界。** 开放轮次中的切换立即追加。空闲切换保持在 bridge 记录上待定,在下一次提示词提交时、assembly 或执行之前追加到开放轮次中;每个旋钮以最后写入为准。开放性来自日志边界而非 `agent.status`,setter 不从 `session/event` 监听器内追加,因为那会重排后续观察者。锚定之前,响应叠加待定值。崩溃丢弃它,重新加载返回持久 fold。
+**已提交的事件是提交边界。** 运行时切换在目标会话上记录其 preset 和发生变化的旋钮事件,之后每次能力解析都折叠最后的值。选择有效的会话追加边界由适配器负责;ACP 传输层没有运行时切换路径。(原先 ACP 的空闲切换锚定——将待定的空闲选择保留到下一次提示词提交——已随该桥一并移除。)
#### 进程内工具
@@ -117,10 +117,10 @@ fs/web/todo 在进程内执行,因此它们的沙箱语义是各自 seam 层
### 测试
-- **单元测试:** 固定平台选择和 profile、失败关闭的 runner 分类、按调用的模式/根目录解析、按进程事实、升级验证和结果、权限 preset fold 和写入透传、叙述器合并、ACP 公布和验证、轮次封闭的配置写入。
-- **Keyless 真实 runner:** 在提供方和 bash 消费方层面对 bwrap、Landlock 和 Seatbelt 执行真实文件系统效果测试;一个真实 Cordis 上下文通过已交付的 bash 和 fs 工具并发驱动两个项目会话,证明在自身根目录写入成功、在兄弟根目录写入被拒绝。Packed-install 覆盖率证明注册表 launcher 保持可执行。真实 ACP 组合固定权限切换并拒绝未知 preset。CI 拒绝静默全跳过。
+- **单元测试:** 固定平台选择和 profile、失败关闭的 runner 分类、按调用的模式/根目录解析、按进程事实、升级验证和结果、权限 preset fold 和写入透传、以及叙述器合并。
+- **Keyless 真实 runner:** 在提供方和 bash 消费方层面对 bwrap、Landlock 和 Seatbelt 执行真实文件系统效果测试;一个真实 Cordis 上下文通过已交付的 bash 和 fs 工具并发驱动两个项目会话,证明在自身根目录写入成功、在兄弟根目录写入被拒绝。Packed-install 覆盖率证明注册表 launcher 保持可执行。CI 拒绝静默全跳过。
- **With-key:** 以只读模式启动真实 ACP 组合,让模型驱动的 bash 写入命中 runner 的拒绝标记,再通过已授权与被拒绝的 workspace-write 重试驱动 bridge 应答器和磁盘效果;不可用的凭证或 runner 自动跳过。
-- **快照:** 固定权限 config-option 协议格式(wire format)、preset 和旋钮事件、提示词 delta 和通知、以及两个脚本化的 approval 分支。一个真实 ACP 示例场景把会话放在用户主目录下,同时让部署后备根目录指向 `/tmp`,然后固定一次成功的 workspace-write 变更;这能区分会话根目录解析与进程后备值,而不依赖 runner 特定的拒绝文本。其他快照以无约束启动,使无关 fixture(测试前置数据)保持平台无关;策略场景显式切换。
+- **快照:** 固定提示词 delta 和通知,以及两个脚本化的 approval 分支。一个真实 ACP 示例场景把会话放在用户主目录下,同时让部署后备根目录指向 `/tmp`,然后固定一次成功的、由部署选定的 workspace-write 变更;这能区分会话根目录解析与进程后备值,而不依赖 runner 特定的拒绝文本。其他快照以无约束启动,使无关 fixture(测试前置数据)保持平台无关。
## 延迟阶段
@@ -152,7 +152,7 @@ fs/web/todo 在进程内执行,因此它们的沙箱语义是各自 seam 层
- **通过 `agent/user-message` + 总线事件叙述**:否决。它预设了一个不存在的轮次入口 seam(真正的 seam 是 `agent/prompt-submit`),而步骤前检查点的位置使一个监听器能够同时服务合并的轮次入口通知和轮中即时性约束。
- **提示词中常驻声明沙箱模式(+ 切换叙述器)**:先交付后移除,基于实际证据:当每个请求中都有 `Bash commands run under the "read-only" file sandbox.` 时,模型拒绝尝试被拒绝后可升级的工作(首次手动会话中十二个轮次有五个以零工具调用结束),将沙箱变成了软锁定。拒绝标记在需要时命名模式,升级字段承载恢复路径;批准旋钮保留其声明,因为自动拒绝在行为上与人的「不」无法区分。
- **用专门的簿记事件追踪「上次告知」**:否决。`request/header` fold 已记录模型看到的确切提示词;将封闭的候选句子解析回来替代了第二条簿记流——事件仅在它们本身即为存储时才需要。
-- **ACP session modes 而非 config options**:否决。preset 已经是一个部署定义的 config-option 选择器,且 modes 计划在 ACP v2 中移除。
+- **相互独立的沙箱与批准选择器**:否决。一个部署定义的权限 preset 让两个策略旋钮对暴露运行时切换的 UI 客户端保持一致。
## 后果
@@ -161,11 +161,11 @@ fs/web/todo 在进程内执行,因此它们的沙箱语义是各自 seam 层
- 被拒绝的命令以 `sandbox_permissions` + `justification` 重试时,通过组合的应答器链提示用户;授权使该次调用在更宽模式下运行(结果事实如此报告),而其他所有调用保持各自的有效模式;每种非授权结果产生各自不同的错误文本且不执行任何内容。
- 升级字段恰好在已挂载的执行器约束时存在;不严格宽于调用有效模式的请求以自身文本失败关闭且不提示任何人;没有 ApprovalService 的部署对升级调用失败关闭,对普通调用不影响。
- 系统提示词从不声明沙箱模式(批准 `'never'` 策略是唯一被声明的旋钮),且整个交互——header、旋钮事件、通知、批准、结果——仅从会话日志即可重建,除两个旋钮事件外无额外事件类型。
-- N 次空闲切换每个旋钮最多产生一个锚定事件(净零序列不锚定任何事件——客户端回显当前选择的无操作推送不记录任何内容);批准策略切换最多以一条合并通知叙述;轮中沙箱切换由下一次调用的盖章兑现。
-- 恢复的会话的覆盖生效并报告给编辑器,无需特殊处理;进程停止期间变更的默认值在会话的首个新请求前被叙述,归因于运维人员。
-- 两个并发会话永远看不到彼此的状态、通知或配置选项。
+- 一次 preset 选择只记录发生变化的旋钮值,而无操作的选择不记录任何内容;批准策略切换最多以一条合并通知叙述,已提交的沙箱切换由下一次调用的盖章兑现。
+- 恢复的会话的覆盖直接生效,无需追赶状态;进程停止期间变更的默认值在会话的首个新请求前被叙述,归因于运维人员。
+- 两个并发会话永远看不到彼此的状态或通知。
- 同一个 Cordis 上下文中的两个并发项目会话解析各自独立的工作区根目录;bash 和 fs 写入在调用方会话的 cwd 内成功,对其相邻会话的 cwd 则失败。
-- `agent-loop` 未被触及——一切搭载 `systemPrompt.section`、`SessionEventMap` 合并、`agent.inject()`、`agent/step`、`agent/prompt-submit` 和 ACP handler 表面。
+- `agent-loop` 未被触及——一切搭载 `systemPrompt.section`、`SessionEventMap` 合并、`agent.inject()`、`agent/step`、`agent/prompt-submit` 和由能力拥有的策略解析。
代价与已接受的限制:
@@ -178,7 +178,6 @@ fs/web/todo 在进程内执行,因此它们的沙箱语义是各自 seam 层
- **模型可能过度请求。** 在没有拒绝依据的情况下升级,或在 `workspace-write` 足够时选择 `danger-full-access`:描述引导且枚举强制阶梯,但人的提示词是实际门控;`approval/asked` 原因使过度请求可审计,且 `prepend` 策略应答器可以自动拒绝部署永远不想要的模式。
- **公布的目标集是静态的,而有效模式是按会话的**(schema 是注册表全局的)——已处于最宽模式的会话仍被提供这些字段。构造上无害:执行时的严格放宽检查(而非枚举)是安全边界——非放宽请求以自身文本失败且不提示任何人。
- **授权的升级不等于可工作的沙箱。** 不可用的后端即使对授权升级到约束模式也仍然失败关闭——在平台没有链或所有探测失败时于 `confine()` 阶段,在未探测的唯一 runner 拒绝时于执行阶段(归类为沙箱失败而非命令失败)——而授权的 `danger-full-access` 运行根本不触及提供方:此时授权(而非探测)是权威。
-- **空闲切换存在于 bridge 内存中,直到下一次提示词提交锚定它。** 该窗口内的崩溃将其回退(在 `session/load` 时报告),且永不再提交提示词的会话永不持久化它——已接受,loop 拥有的空闲提交轮次留作未来工作(如果持久性成为需求)。
- **批准叙述器的重启基线解析提示词文本。** 封闭的候选句子由写入模块本身拥有,因此措辞变更是同一文件中写入器+解析器的协调编辑;header 早于该段落的会话静默采用当前策略而不发通知。
- **批准段落仍是动态提示词表面**(`'never'` 切换会破坏该会话的提供方提示词前缀缓存)。已接受:策略切换罕见,且模型基于过时的 `'never'` 行动更糟。沙箱旋钮不再触及提示词。
- **模型可能持有关于沙箱模式的过时信念**(没有任何东西宣布切换)。有意接受:下一次尝试的标记或成功会纠正它,而宣布的观察到的失败模式——预防性拒绝——比一次浪费的重试更糟。
@@ -192,7 +191,7 @@ fs/web/todo 在进程内执行,因此它们的沙箱语义是各自 seam 层
- **沙箱限制网络或进程可见性吗?** 不——`SandboxMode` 仅声称文件操作;bwrap profile 刻意不 unshare pid,没有后端声称网络。网络限制是否成为自己的旋钮留在 § seam 中开放。
- **哪些工具实际在约束下运行?** 通过 `ctx.bash` 的 OS 子进程——bash 工具及传递性的钩子命令——再加上通过沙箱化 `ctx.fs` 提供方运行的文件系统工具(`read`/`write`/`edit`,见[跨工具族 fs 沙箱 Agent Note](2026-07-14-cross-family-fs-sandbox.md)):bash 通过 OS runner 约束,fs 通过进程内路径围栏约束,二者都以同一个 `ctx.sandboxPolicy` 模式为键。web/todo 仍在进程内且不受限制(web 的唯一效果是网络,不在文件效果模式词汇内)。
- **授权的升级会持久化吗?** 不会。授权由发起请求的确切前台或后台调用消费;每个相邻调用保留自己的有效模式。后续的后台拒绝通过 `task_output` 呈现,并且可以作为一次新的精确命令重试的依据。
-- **编辑器的模式切换何时生效?** 轮中:立即追加,由下一次调用的盖章兑现。空闲:保持在 bridge 的会话记录上,在下一次 `agent/step` 时锚定到其开放轮次中,N 次切换合并为最多一个事件(净零则无);锚定前崩溃回退它,`session/load` 报告真实状态。模型不被告知——其下一个命令直接在新模式下运行。
+- **运行时模式切换何时生效?** 一旦其会话事件提交,紧接着的下一次能力解析就会折叠并盖章新模式。模型不被告知常驻模式;其下一个命令直接在新策略下运行,任何拒绝都会在使用点命名该策略。
- **重启后什么存活——如果运维人员在进程停止期间改了配置默认值呢?** 覆盖从会话日志回放(`effective = fold ?? config`),因此恢复的会话以零追赶机制保持其模式;离线漂移的默认值以与切换相同的方式改变行为(批准策略因被声明,还额外以运维人员/配置归因叙述)。
- **结果上的 `enforcement: 'partial'` 是什么意思?** 所选后端强制其内核 ABI 管控的子集——例如 ABI v3 之前的 Landlock 不管控路径 truncate——并以结构化方式如此声明而非拒绝主机;探测的报告行区分各种情况。bwrap 和 Seatbelt profile 构造上管控所有承诺的文件操作,因此始终报告 `full`。
diff --git a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.i18n.yaml b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.i18n.yaml
index f199fdb731..940908457f 100644
--- a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.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
-2026-07-07-mcp-client-plugin.md: 90c43383882034954f33b79459fac273e60b9e0e
-2026-07-07-mcp-client-plugin.zh.md: a60aa537a03dec5add7e91fd78b91fece4e4a259
+2026-07-07-mcp-client-plugin.md: 2a8ad62e3bad1f2ae47100294f5dba25ceb47d12
+2026-07-07-mcp-client-plugin.zh.md: 9860d5ec805b394eb0c5f59f54349264593a5599
diff --git a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.md b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.md
index 90c4338388..2a8ad62e3b 100644
--- a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.md
+++ b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.md
@@ -99,7 +99,7 @@ This server-qualified shape is the de-facto standard among multi-server agent cl
1. On connect: drain `client.listTools()` pagination, derive every tool's `publicName`, then register each as a raw `ToolDefinition` via `ctx.tools.register()`. The MCP JSON Schema and description pass through unchanged (no `defineTool` DSL conversion); only the model-facing `name` is replaced.
2. Listen for `notifications/tools/list_changed` → re-run the same sync (dispose previous generation, register new). Deterministic names mean unchanged tools keep their names across re-syncs.
3. The executor closes over `rawName`; the public name is never sent to the server and never parsed to recover the raw name.
-4. No `presentCall`/`presentResult` — the ACP bridge's generic-card fallback handles rendering.
+4. No `presentCall`/`presentResult` — UI consumers use the provider-neutral generic-card fallback.
5. Tools are transparent in the system prompt — no "[via MCP]" annotation beyond the name itself.
### Public name normalization
@@ -201,7 +201,7 @@ Coverage is named per tier; each behavior lives at the cheapest tier that can ex
- **Unit** (`tests/mcp-client.spec.ts`, `tests/apply.spec.ts`, mocked MCP SDK): the `publicToolName` algorithm (clean, normalize, truncate-and-hash, determinism, distinct-identity separation), raw-vs-public wire discipline, cross-server and native-tool coexistence, duplicate-`serverName` load failure and reservation release, invalid-tool-list rejection, generation swap/rollback, failed-re-sync retention, result mapping, cancellation, config schema validation. 100% per-file coverage gates the package.
- **E2E** (`tests/mcp-client.e2e.ts`, keyless): the real MCP protocol against the in-repo fixture server, `@modelcontextprotocol/server-everything`, and `@modelcontextprotocol/server-filesystem` over stdio, and against an in-process `StreamableHTTPServerTransport` server over Streamable HTTP — discovery under the namespace, dotted-name normalization end to end, execution round-trips, duplicate-`serverName` rejection, disposal.
-- **Snapshot**: deliberately none. MCP tools introduce no new transcript surface — they register as raw `ToolDefinition`s and render through the ACP bridge's generic-card fallback, which the bridge's unit suite already pins (`packages/ui/acp/tests/stream-update.spec.ts`). Adding an MCP server to the snapshot example's `cordis.yml` would mutate the pinned `text-turn` system-prompt fixture (forcing a with-key re-record of every recorded expected output) and make every replay depend on spawning an external MCP server process — for zero new rendering behavior. If a later change gives MCP tools their own render intent, that change names its snapshot coverage then.
+- **Snapshot**: deliberately none. MCP tools introduce no new presentation shape — they register as raw `ToolDefinition`s and UI consumers use the generic-card fallback already pinned by their presentation suites. Adding an MCP server to a runnable snapshot composition would mutate its pinned system-prompt fixture and make every replay depend on spawning an external MCP server process for no new behavior. If a later change gives MCP tools their own render intent, that change names its snapshot coverage then.
## Consequences
diff --git a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.zh.md b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.zh.md
index a60aa537a0..9860d5ec80 100644
--- a/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-07-mcp-client-plugin.zh.md
@@ -99,7 +99,7 @@ type Config = StdioConfig | StreamableHttpConfig
1. 连接时:遍历 `client.listTools()` 的分页结果,推导每个工具的 `publicName`,然后通过 `ctx.tools.register()` 将其注册为原始 `ToolDefinition`。MCP 的 JSON Schema 和描述原样透传(不做 `defineTool` DSL 转换);仅替换模型可见的 `name`。
2. 监听 `notifications/tools/list_changed` → 重新执行同步(dispose 上一代、注册新一代)。确定性命名意味着未变化的工具在重新同步后保持原名。
3. 执行器闭包持有 `rawName`;公开名称永远不发送给服务器,也永远不被解析以还原原始名称。
-4. 无 `presentCall`/`presentResult`——ACP 桥接的通用卡片兜底负责渲染。
+4. 无 `presentCall`/`presentResult`——UI 消费方使用提供方无关的通用卡片兜底。
5. 工具在系统提示词中是透明的——除名称本身外不附加「[via MCP]」标注。
### 公开名称规范化
@@ -201,7 +201,7 @@ v1 否决。它能防止跨服务器冲突,但无法将 MCP 注册与原生 ha
- **单元测试**(`tests/mcp-client.spec.ts`、`tests/apply.spec.ts`,mock MCP SDK):`publicToolName` 算法(干净名称、规范化、截断加 hash、确定性、不同标识的分离)、raw 与 public 的协议纪律、跨服务器与原生工具共存、重复 `serverName` 加载失败与预留释放、无效工具列表拒绝、代切换/回滚、重新同步失败时的保留、结果映射、取消、配置 schema 校验。100% 逐文件覆盖率门禁约束该包。
- **E2E**(`tests/mcp-client.e2e.ts`,无需密钥):使用真实 MCP 协议对接仓库内的 fixture(测试前置数据)服务器、`@modelcontextprotocol/server-everything` 和 `@modelcontextprotocol/server-filesystem`(stdio 传输),以及进程内 `StreamableHTTPServerTransport` 服务器(Streamable HTTP 传输)——命名空间下的发现、带点号名称的端到端规范化、执行往返、重复 `serverName` 拒绝、dispose。
-- **快照**:刻意不做。MCP 工具不引入新的 transcript(文本记录)呈现面——它们以原始 `ToolDefinition` 注册,通过 ACP 桥接的通用卡片兜底渲染,该兜底已由桥接的单元测试套件固定(`packages/ui/acp/tests/stream-update.spec.ts`)。将 MCP 服务器添加到快照示例的 `cordis.yml` 会改变已固定的 `text-turn` 系统提示词 fixture(迫使每条录制的预期输出都需要带密钥重新录制),且使每次回放依赖于 spawn 外部 MCP 服务器进程——而新增渲染行为为零。如果后续变更为 MCP 工具引入专属渲染意图,该变更届时自行声明快照覆盖。
+- **快照**:刻意不做。MCP 工具不引入新的展示形态——它们以原始 `ToolDefinition` 注册,UI 消费方使用各自展示测试套件已固定的通用卡片兜底。将 MCP 服务器添加到某个可运行的快照组合会改变其已固定的系统提示词 fixture,且使每次回放依赖于 spawn 外部 MCP 服务器进程,而新增行为为零。如果后续变更为 MCP 工具引入专属渲染意图,该变更届时自行声明快照覆盖。
## 后果
diff --git a/.agents/notes/implemented/feature/2026-07-07-plan-mode.md b/.agents/notes/implemented/feature/2026-07-07-plan-mode.md
deleted file mode 100644
index 06fd50e1f0..0000000000
--- a/.agents/notes/implemented/feature/2026-07-07-plan-mode.md
+++ /dev/null
@@ -1,192 +0,0 @@
-# Agent Note: Plan mode — a logged per-agent session mode
-
-Status: implemented
-
-> **Superseded vocabulary (2026-07-22):** [Collapse named session modes into plan mode](../simplification/2026-07-22-plan-specific-collaboration-state.md) replaces this note's generic `dsh-mode`, `mode/set`, definition map, and `ctx.modes` design with the current plan-specific `dsh-plan-mode`, `plan/mode`, `{ section }`, and `ctx.planMode` contract. The review, boundary, reconstructability, and sandbox-orthogonality decisions below remain in force; generic API examples are retained as the historical design this simplification removed.
-
-## Problem
-
-Before this change, the harness had no durable way to put one agent into a distinct working stance. Plan mode needs the agent to explore and design under planning guidance, produce a reviewable artifact, cross an explicit approval boundary, and restore that state across resume and fork without making the model-visible request diverge from the session log.
-
-The extension seams already supplied the surrounding pieces: [`system-prompt/assemble`](../../../../packages/core/system-prompt/README.md) shapes guidance per step and the shipped request is logged in `request/header*` events ([reconstructability](../../implemented/architecture/2026-07-05-reconstructable-requests.md)); [`ctx.userInteraction`](../../../../packages/ui/user-interaction/README.md) carries the approval question and corrective feedback ([ask-user precedent](../../implemented/feature/2026-06-25-ask-user-question.md)); `SessionEventMap` carries durable per-agent facts ([the `todo/write` precedent](../../implemented/feature/2026-06-29-todo-write-tool.md)). The missing piece was the named session state that joins those seams while leaving execution enforcement on the independent sandbox and approval axes.
-
-## Decision
-
-The deliverable is **plan mode**. It ships as the first **session mode** — a named, logged, per-agent COLLABORATION state: a mode definition is deployment-configured guidance the model sees, while the mode IN FORCE for an agent is session state folded from its log. Modes are one axis and the enforcement knobs — the sandbox mode, the approval policy — are others: they never read or write each other, matching how Codex keeps its Plan/Default collaboration presets separate from its sandbox and approval settings. One new product package, `@deepseek-ai/dsh-mode` at `packages/mode/mode/`, owns the event vocabulary, a thin `ctx.modes` service, and every listener; the loop does not change. `plan` is the only required definition — the mode-shaped vocabulary exists so a second mode never renames durable event types, not because more modes ship now.
-
-The state is one `SessionEventMap` member: **`mode/set`**, a log-only, non-surface event carrying `{ mode: string }` with whole-value-replace semantics, plus a pure `foldMode(events)` that returns the mode in force — the last `mode/set`, or the default mode when none exists. Because [the log is the fact channel](../../implemented/architecture/2026-06-30-event-domain-semantics.md), resume, fork, and compaction restore the mode with no extra machinery, and UIs read flips off `session/event`. The default mode is the absence of mode guidance — no section, filtering, or gate. Loading `dsh-mode` still contributes one stable `exit_plan_mode` schema in every mode; that fixed cost avoids tool-catalog churn at mode boundaries.
-
-A mode's whole surface is soft: a `mode:policy` prompt section renders the active definition's guidance, while `exit_plan_mode` remains in the registered tool catalog across every mode and rejects at execution unless the folded mode is `plan`. A transition therefore changes only the system-prompt portion of the attributable `request/header` on the next step, keeping [reconstructability](../../implemented/architecture/2026-07-05-reconstructable-requests.md) green without changing native schemas or Code Mode's SDK. A mode deliberately enforces NOTHING: no execution gate, no tool filtering, no reach into the sandbox or approval knobs — a user who wants a hard read-only floor while planning switches the sandbox-mode option beside the mode picker, in either order, and neither axis disturbs the other. There is likewise NO per-mode tool allow/deny list — which tools a mode admits is an effects question, parked until tool definitions declare their effects ([Deferred](#deferred)); a mode's restraint is its section's guidance plus the exit review.
-
-The model leaves plan mode through the **`exit_plan_mode`** tool: its single argument is the plan text, which makes the plan reconstructable from the log, and the tool conducts the review itself through the user-interaction seam — a question whose supporting detail carries the exact plan, with options and a free-text channel, not a bare permission — so an approval flips the logged mode back to the default, and a rejection becomes the corrective error carrying the user's feedback verbatim, which keeps the model planning with direction. A user flips the mode from any surface through `ctx.modes.set()`; the flip is applied at the next turn boundary (session events are turn-enclosed) and narrated to the model once, only when the model-visible state actually changed.
-
-## High-level API
-
-### A plan-mode session end to end
-
-The user switches the session to plan mode through the ACP mode picker or `/plan [message]` in a terminal front door, and from the next step every request ships the configured plan guidance section. When the optional message is present, that same command submits it into the affected step. The `exit_plan_mode` schema was already present in default and remains byte-identical.
-
-The model explores and designs; the section's guidance is what defers changes into the plan. The sandbox and approval knobs keep whatever the user set them to — a deployment (or user) that wants kernel-enforced read-only during planning pairs plan mode with the independent sandbox-mode option.
-
-When ready, the model calls `exit_plan_mode` with the plan markdown as its argument; the review question carries that exact markdown as supporting detail — approve, or keep planning, with free-text feedback welcome. A native call also renders the plan card; a Code Mode nested dispatch has no native card, so the review detail is the common presentation surface.
-
-On approve, the tool flips the logged mode back to the default: the next step drops the plan section while retaining the same tool catalog (the changed header is in the log), and execution tracking from there is already `todo_write`'s job. On keep-planning, the model receives a corrective error carrying the user's feedback text, revises, and re-presents.
-
-### Deployment configuration
-
-Mode definitions are validated plugin Config — per repo convention, changeable from `cordis.yml` with no code edit. The deployment must provide the complete `plan` section; the package embeds no model instructions. Additional modes use the same config map:
-
-```yaml
-- id: mode
- name: '@deepseek-ai/dsh-mode'
- config:
- modes:
- plan:
- section: |
- You are in plan mode: explore and design, then present the
- plan for approval through exit_plan_mode.
-```
-
-A definition is exactly `{ section }` — there is deliberately no per-mode tool list and no enforcement field ([FAQ](#faq)). Definition names use the lowercase slash-command subset `/^[a-z][a-z0-9_-]*$/u`; `default` is reserved (the absence of policy) and rejected as a key. An invalid name or unknown definition key — a `tools` list or an `access` cap included — fails validation at load; an unknown mode name fails loudly at `set()` time.
-
-### In the terminal
-
-Terminal front doors get one entry command per configured definition through the plugin-owned command registry (`@deepseek-ai/dsh-commands`): `dsh-mode` registers `/plan [message]` for the required definition and, for example, `/review [message]` when `review` is configured. Each command records its named switch; a non-empty optional message is trimmed and passed to `agent.steer()`, which places it in a running agent's next step or delegates to `send()` for a new idle turn. The command name and result stay out of model history, while that explicit message is logged as an ordinary user message under the selected mode. The synthetic `default` entry contributes no command. The exit review prompts right in the terminal with no new machinery: it is an ordinary user-interaction question, so it rides the composed user-interaction provider's prompt queue that `ask_user_question` already uses.
-
-### Over ACP
-
-The mode PICKER is this package's surface: `session/new`/`session/load` advertise `availableModes`/`currentModeId` from `ctx.modes` (consumed opportunistically via `ctx.get`, the `tool-bash` pattern), `session/set_mode` calls `set()` and notifies `current_mode_update` optimistically (the pending mode IS the user's selection; the logged `mode/set` follows at the boundary), and a `session/event` listener re-notifies on each logged flip that differs from the last sent. The exit tool reuses the user-interaction ACP provider's elicitation flow; its ACP mapping carries the review `detail` because Code Mode nested dispatches have no native plan card, while native calls may additionally stream the plan card. Individual environment knobs — sandbox mode, approval policy, the model — are NOT modes and belong to `session/set_config_option` ([FAQ](#faq)).
-
-### For agent creators
-
-`ctx.modes` is the whole programmatic surface: `list()` returns the configured definitions plus the synthetic `default` entry (for pickers), `get(agent)` returns the folded mode plus any pending intent, and `set(agent, mode)` validates the name against `list()`'s vocabulary and records the boundary-applied intent — `default` is always a valid target, so exiting a mode is the same call as entering one. There is no creation-time mode option — a caller selects through `set()` before the first turn, which flushes identically. There is no live `agent/*` mirror to subscribe: UIs read `mode/set` off `session/event`, per [event-domain semantics](../../implemented/architecture/2026-06-30-event-domain-semantics.md).
-
-## Detailed design
-
-### Vocabulary
-
-```text
-'mode/set': { mode: string } // SessionEventMap merge in dsh-mode: log-only, non-surface,
- // whole-value replace — the last one in the log wins
-DEFAULT_MODE = 'default' // the fold of a log with no mode/set; reserved, not definable
-```
-
-The payload carries no reason/provenance field: a tool-driven flip sits next to its `tool/call` in the log and a user flip sits at its turn boundary, so the cause is log-adjacent — the same "narrative fields are derivable" call the [reconstructability Agent Note](../architecture/2026-07-05-reconstructable-requests.md) made for request-header facts (the in-flight `env/state` event carries a `source` precisely because its drift variant has NO log-adjacent cause — a contrast, not a conflict). Mode names are config-declared vocabulary, not opaque cross-boundary ids, so they stay bare strings (no `Branded`).
-
-### Config and the resolve step
-
-```text
-interface ModeDefinition { section: string } // prompt text — a mode's whole vocabulary
-interface ModeConfig { modes: Record } // plan is required and owns its complete prompt
-resolveConfig(config): ResolvedModes // explicit resolve (the dsh-bash template), fail-loud:
- // missing plan, 'default', blank sections, and unknown keys rejected
-```
-
-The one-field shape is deliberate minimalism, not the final vocabulary: a per-tool policy dimension returns as effects metadata on tool definitions ([Deferred](#deferred)), read here rather than re-declared per mode — the config shape must not need a migration when it arrives.
-
-### The fold, the service, and the flush
-
-`foldMode(events)` is pure (exported for reconstructors and tests) and folds the append-only session log directly; `mode/set` is not a surface node, so compaction cannot shadow it. `set(agent, mode)` validates the name against `list()`'s vocabulary — the configured definitions plus the reserved `default`, which is rejected as a config KEY but always accepted as a `set()` TARGET — drops a no-op (target equals pending, else current), and otherwise records `{ mode, narrate }` in a `WeakMap` pending-intent slot. It cannot append immediately because [every session event is turn-enclosed](../../implemented/architecture/2026-06-15-turn-enclosure-invariant.md) and an idle agent has no open turn.
-
-Contained listeners on the loop's interception seams ([defensive patterns](../../../../docs/defensive-patterns.md): a policy plugin must not block a prompt or a turn) flush the pending intent as a `mode/set` append — `agent/prompt-submit` fires inside the just-opened turn before its first assembly, and `agent/turn-continuation` fires after an ordinary step closes before its successor. Automatic request recovery bypasses continuation, so a prepended `agent/request-error` wrapper delegates through the composed policy and asynchronous backoff, then flushes only a `retry` decision before the waterfall returns to the loop; an effect-scoped lifetime guard suppresses a captured wrapper that resumes after plugin disposal. All three paths sit outside tool execution and log publication (post-commit `session/event` observers are observe-only), so every step runs under the mode its assembly folded. When the flushed mode differs from the fold at the last `request/header`, the flush appends one coalesced `context/message` notice in the same frame ("The user switched this session to plan mode."); the user-visible narration cases are enumerated in the [FAQ](#faq).
-
-### The soft layer: a computed section and a stable exit schema
-
-The registered prompt section reads the calling agent's mode from `AssembleContext.agent` and resolves to the active definition's guidance or `''`. The loop renders per step and logs a complete `request/header` whenever the rendered header changes, so entering or leaving a mode is attributable. The section is static per mode and the plan itself stays in the conversation as messages and tool arguments; re-injecting separate plan state on every request ([Prior art](#prior-art)'s compaction-survival hack) is unnecessary prompt churn.
-
-The guidance contribution is `{ name: 'mode:policy', order: 50, text: context => … }`: after persona (0), before tool guidance (100–199), and empty for default or agent-less assembly. `exit_plan_mode` is registered once through `ctx.tools` and never filtered, so native schemas and Code Mode's generated SDK remain byte-identical across mode switches; a deployment without `dsh-mode` lacks that one binding. There is NO `tools/pre-execute` listener: a mode gates nothing, while the exit tool's own folded-mode check rejects out-of-plan calls. The exit review is a question with options and feedback, not a permission, so it lives inside the tool's execution over the user-interaction seam.
-
-### `exit_plan_mode`
-
-`defineTool` has one required `plan: string` argument. Native execution records it in the ordinary `tool/call`; Code Mode records the outer `run_code` source before execution and appends the normalized nested arguments in `tool/code-dispatch` after the dispatch settles. `execute` rejects an agent-less call (the [`todo_write` precedent](../../implemented/feature/2026-06-29-todo-write-tool.md)), rejects any folded mode other than `plan`, rejects an empty or heading-less plan before asking the reviewer, then conducts one single-select `ctx.userInteraction.ask()` review whose `detail` is the exact plan — approve or keep planning — with free-text feedback open. Only exactly one `Approve` selection consents; every other shape fails closed. Approval records a SILENT boundary-applied intent to switch to `default` and returns a short confirmation. The deployment guidance tells the model to make this the only and final tool call in its response; if a model violates that rule, the runtime still holds plan guidance for the rest of the batch, and the next step logs a changed header with the guidance removed and tool schemas unchanged. Every non-approval outcome returns a corrective `isError` and leaves the mode in `plan`.
-
-Its [render intent](../../implemented/architecture/2026-07-02-tool-render-intent-union.md), decided up front: `presentCall` is a `generic` card titled by the plan's first heading with the plan markdown as content, plus a `generic` result card. Native front doors show that card before the question; Code Mode nested dispatches do not produce native call-card events, so the user-interaction `detail` independently carries the same plan on every provider. The seam is consumed opportunistically (`ctx.get('userInteraction')`), so `dsh-mode` composes without it and degrades to the manual exit pinned in the [FAQ](#faq).
-
-### Dependencies and surfaces
-
-`dsh-mode` is one product package, not a capability-seam trio ([Alternatives considered](#alternatives-considered)): it peers on `cordis`, `dsh-session`, `dsh-agent`, `dsh-tools`, and `dsh-system-prompt`, injects `['tools', 'systemPrompt']`, and reads `ctx.userInteraction` opportunistically at execute time (a type-only peer edge on `dsh-user-interaction`); its only UI-facing edges are optional type-only peers (`dsh-commands` for the per-definition entry commands). Beyond the `ctx.modes` call surface everything participates through listeners, so dropping the package gracefully removes modes rather than breaking a consumer. Terminal front doors need no mode-specific code: `dsh-mode` itself registers each definition's command on the command registry when one is composed (an optional type-only peer edge on `dsh-commands`), and the exit review rides the composed user-interaction provider's prompt queue. The ACP wire mapping is pinned in [High-level API](#over-acp); package-wise the bridge takes a type-only peer edge on `dsh-mode` and reads the service opportunistically, so a bridge without the plugin behaves exactly as today.
-
-### The recorded scenario and the harness op
-
-`input.json` gains one step op, `{ "op": "setMode", "modeId": "plan" }`, driven through the real `session/set_mode` RPC, and a scripted `elicitationAnswers` queue. The `plan-mode` scenario enters plan before turn 1, runs a real `cat` under the independently configured sandbox, presents a plan through `exit_plan_mode`, receives scripted approval, then edits on the next step. The first `request/header` contains the full stable toolset plus the configured mode section; the post-approval changed header retains byte-identical tool schemas and removes only that section. `plan-mode-reject` pins corrective free-text feedback and the unchanged plan state. Both recordings replay host commands under Seatbelt or bwrap; backend-specific sandbox denial stays at the bash-tool unit tier.
-
-### The mechanical tail
-
-No new cordis event is declared (`mode/set` rides `session/event`; the listeners attach to existing waterfalls), so the events catalog is untouched. Regenerated in the same change: the persistence log catalog (`mode/set`), the services catalog (`ctx.modes`, JSDoc-complete), the config catalog (`ModeConfig`), the tool catalog (`exit_plan_mode`), the producer/consumer map and doc graphs, and the module graph. Repo plumbing: a root tsconfig `paths` entry, the new group's README plus a [packages map](../../../../packages/README.md) row (a new top-level group is the deliberate act that table names), an `architecture.md` capability-services row for `ctx.modes` (budget-checked), and the cookbook row upgrade.
-
-## Deferred
-
-Each behind its own decision: subagent mode inheritance via a forwarded creation-time mode option (removed as unconsumed; it returns with its first consumer), preset modes beyond `plan` (read-only, accept-edits), the idle-record primitive if pending-intent loss proves real, and — the big one — **effects self-declaration on tool definitions**: a per-tool read-only/mutating classification (the MCP `ToolAnnotations` vocabulary — `readOnlyHint`/`destructiveHint` — is the natural template, with its untrusted-hint caveat implying trust tiers). That item is what a general per-mode tool policy waits on: this Agent Note first shipped an interim per-mode name allowlist and removed it before release — a hand-maintained list mislabels the effects question, must track every tool a deployment composes, and rots silently as tools arrive — so mode-scoped tool availability (and per-tool `ask` policies) returns as a CONSUMER of declared effects, which is its restart trigger.
-
-The canonical [`examples/acp-agent`](../../../../examples/acp-agent/) composition mounts the mode and question-tool plugins on the full ACP coding server; plan mode is an additive session feature, not a second server profile. Its snapshot suite pins the plan-shaped initial header, a real read, scripted approval, stable tool schemas across the pure-removal header delta, a subsequent edit, rejection feedback, and the keyless mode wire. A self-skipping real-API smoke boots that same leaf, verifies the file before approving the review, and verifies the approved implementation afterward.
-
-## FAQ
-
-Behavioral clarifications of the chosen design; rejected designs live in [Alternatives considered](#alternatives-considered), accepted costs in [Consequences](#consequences).
-
-**When does a user's mode flip take effect?** At the next pre-assembly boundary: `agent/prompt-submit` covers the first step, `agent/turn-continuation` covers a normal successor, and the post-composed `agent/request-error` retry decision covers automatic recovery. A mode selected while a request or retry backoff is in flight therefore shapes the following model request. This is the "applies to subsequent requests" semantics every product in [Prior art](#prior-art) ships.
-
-**When is a mode change narrated to the model?** Only when the model-visible state actually changed: the flush compares the flushed mode against the fold at the last `request/header` and narrates once, coalesced. A net-zero flip sequence (plan then back, all before the boundary) narrates nothing; a tool-driven exit narrates through its own tool result instead; a mode set before the first turn narrates nothing — the section is the state statement. The principle is the in-flight env-state proposal's boundary narration: a silently flipped prompt surface leaves the transcript arguing from a state the header no longer has.
-
-**What happens on resume when the config no longer defines the folded mode?** A folded mode name the current config no longer defines behaves as the default mode without a notice, so the session neither gains a substitute restriction nor becomes unusable. `set()`'s loud validation covers only the write path; a resumed log answers to the config it finds.
-
-**What if a deployment composes no user-interaction provider?** Plan mode stays safe but manual: `ctx.userInteraction.ask()` throws `NO_PROVIDER` (and an absent seam never resolves at all), the tool returns the corrective `isError`, and the exit degrades to the user toggling modes — never to an unreviewed exit. The mode section tells the model to present its plan through `exit_plan_mode` — and to ask the user in prose if that fails — so it keeps presenting instead of stalling.
-
-**Why is there no per-mode tool allowlist?** Because "which tools are safe in a planning mode" is a property of each TOOL (its effects), not of the mode — a per-mode name list re-declares that fact in the wrong home, must enumerate every tool the deployment composes (MCP servers included), and rots silently as tools arrive. Until tool definitions declare their effects ([Deferred](#deferred), where the removed interim allowlist is archived with its restart trigger), a mode restrains by its section and the exit review; the exposure is an accepted cost ([Consequences](#consequences)).
-
-**Do subagents inherit the parent's mode?** A fork child inherits for free — the parent's `mode/set` is inside the seeded prefix. A spawn child starts in the default mode; a creation-time mode option and automatic forwarding by subagent providers are deferred together ([Deferred](#deferred)).
-
-**How does plan mode relate to the sandbox's read-only mode?** They are separate axes that never touch: the mode is the collaboration stance (a `mode/set` fold), the sandbox mode is an enforcement knob (a `bash/sandbox-mode` fold, [the sandbox Agent Note](2026-07-06-sandbox.md)) — plan mode neither reads nor caps it, exactly as Codex keeps its Plan/Default presets separate from its sandbox and approval settings. A user who wants kernel-enforced read-only while planning sets both: flip the mode picker AND the sandbox-mode option, in either order; each switch changes only its own fold, so there is no interference and no restore step to crash out of. The log attributes each axis to its own event — the stance to `mode/set`, the confinement to `bash/sandbox-mode`.
-
-**Why aren't sandbox mode, approval policy, or the model themselves modes?** They are individual environment knobs and belong to ACP's `session/set_config_option`; the division this proposal pins is picker-to-modes / knobs-to-config-options, recorded in [the feature matrix](../../../../packages/ui/acp/acp-feature-support.md) now that both this stack's picker and the sandbox stack's config options are landed. A mode definition may later bundle env facts (applied through `ctx.envState` where mounted) so a Codex-style preset stays a single mode; fusing approval policy into the mode CONCEPT itself is rejected in [Alternatives considered](#alternatives-considered).
-
-## Prior art
-
-A survey of shipped plan modes (Claude Code, Cursor, Copilot, OpenCode, Gemini CLI, Cline, Windsurf, Codex) shows the same five parts everywhere — the low-authority tool policy, plan artifact, approval moment, execution-state switch, and durable state that [Problem](#problem) builds on.
-
-The mode surface is a LIST everywhere it is advertised, never a boolean: Claude Code's picker offers `plan` beside `acceptEdits` (plus an auto-mode entry into plan), and Codex exposes `Plan` beside `Default` as collaboration-mode presets while keeping approval and sandbox settings separate. This is the surface [the ACP feature matrix](../../../../packages/ui/acp/acp-feature-support.md) records as the gap, and what sizes the vocabulary as named modes rather than a flag.
-
-The deployment-owned example prompt borrows the instrumental behavior, not product-specific mechanics. From Codex: remain in plan mode despite imperative implementation language, explore before asking, distinguish repository facts from user-owned choices, and make the plan decision-complete across APIs, data flow, failures, tests, and assumptions. From Claude Code: prohibit mutations and commits, prefer existing patterns, use questions only for requirements or approach choices, and finish through the exit tool rather than a prose approval request. It deliberately omits Codex protocol tags and Claude's plan-file or phased-subagent machinery because those belong to their runtimes, not this plugin contract.
-
-The ecosystems that leave modes to convention show the failure shapes to avoid. Pi-style mode extensions fight over a last-wins global active-tool list, enforce "read-only" by prompt text alone (a hallucinated call to a still-registered tool executes), and re-inject plan state into every request to survive compaction. The contested global list and the re-injection hack close structurally here — per-agent folded state, and a log-only non-surface event compaction cannot shadow. The prompt-only shape, by contrast, is deliberately KEPT — it is what Codex ships for Plan, and it is why the mode axis composes freely with the enforcement axes: a deployment that wants a hard floor pairs the mode with the independent sandbox knob instead of the mode carrying its own enforcement ([FAQ](#faq)).
-
-## Alternatives considered
-
-**Permission modes as the concept (the Claude Code shape).** One `permissionMode` fusing approval policy and tool policy. Here those are two axes with two owners: the approval seam owns "who answers this question", modes own "what surface does the model get". ACP models them as related but distinct (a mode may select an approval policy later — a mode definition gains a field, not a merger).
-
-**A capability-seam trio.** Interface/implementation/consumer fits a swappable backend; a mode's variable parts are config values, not implementations. Splitting would manufacture an empty implementation package — the same "don't split preemptively" call the approval seam and [`todo/`](../../implemented/feature/2026-06-29-todo-write-tool.md) made.
-
-**Loop-owned mode state.** Rejected on the standing rule (plugins, not loop changes): every hook the feature needs — assemble, pre-execute, turn boundaries, session events — is already a documented seam, so a loop edit would buy nothing but coupling.
-
-**A per-mode tool allowlist with a deny-by-default gate (the first shipped shape).** Removed before release. A hand-maintained name list re-declares a per-TOOL fact (its effects) per MODE: it must enumerate every tool the deployment composes — MCP servers and future registrations included — and it rots silently as tools arrive (a new read-only tool is blocked until someone edits every mode; the author burden lands on whoever knows the mode, not whoever knows the tool). It also over-promises: the list looks like a security boundary while the real boundary for anything non-shell does not exist. The general dimension is parked on effects self-declaration ([Deferred](#deferred)); the consequence — plan mode is guidance-only, the very Pi hole the gate once closed — is accepted deliberately, priced in [Consequences](#consequences).
-
-**An `access` sandbox cap on the mode (the second shipped shape).** Also removed before release. `ModeDefinition.access` clamped the bash seam's per-call sandbox resolution to a mode-declared ceiling (a `bash/resolve-mode` waterfall + ladder-min listener, with guards withholding bash under an unconfinable executor and denying escalation mid-mode). The state stayed orthogonal — the clamp never wrote the sandbox knob — but the AXES did not: entering plan changed what the sandbox enforced, fusing the collaboration stance with an enforcement level and contradicting the Codex-shaped separation the review converged on (Plan/Default presets never touch sandbox or approval settings). One user-visible symptom of the fusion: flipping the sandbox option to `workspace-write` while planning silently did nothing. The cap, the waterfall, and the mode→bash dependency edge were removed together; a deployment gets kernel-enforced read-only planning by pairing the mode with the independent sandbox-mode option, and a mode-triggered PRESET (a mode definition bundling suggested knob values, applied as ordinary knob switches) can return later without re-fusing the axes.
-
-**Runtime-only mode (UI- or bridge-local, unlogged).** Resume and fork would silently drop the mode, and the header deltas a mode causes would have no attributable cause in the log. Logged state is what makes the mode auditable and restorable for free.
-
-**Mode flips as `context/message` via `agent.inject()`.** Reuses an existing turn-enclosure path, but puts policy state into the model transcript — the model does not need to be told twice (the section already tells it), and a log-only fact should not occupy surface.
-
-**A plan-file store (`.plans/` directory).** A second durable home for what the log already carries replayably; a deployment wanting files can add a tool that writes them. One home per fact.
-
-**A boolean `planMode` instead of named modes.** Too narrow for the surface the repo already tracks: ACP advertises a mode LIST and the shipped pickers fill it with more than plan ([Prior art](#prior-art)); generalizing later would rename durable event vocabulary. The string-shaped mechanism costs nothing extra now; only `plan` ships as a definition.
-
-**A tool-policy-stack service (the Pi-critique remedy).** A dedicated composition service for tool policies is premature: this implementation performs no mode-scoped tool filtering, and future effect policies can compose through the existing guarded execution seams. Formalize only when declared tool effects create a concrete composition requirement.
-
-**Exit approval through the approval seam (a `{ kind: 'ask' }` gate decision).** The original sketch, natural while the approval seam was the only asking machinery in flight — but it seats a review in a permission chair: the seam's outcome vocabulary is deliberately closed and one-shot (`allowed-once`/`rejected`), so a rejection carries no feedback and an approval can never grow options (approve-and-accept-edits). The exit moment is a question, not a permission — the user-interaction seam gives it options plus the free-text channel, and the rejection feedback reaches the model verbatim. The approval seam remains the right seat for genuine permission gates (the sandbox escalation), and the registry's `ask` vocabulary stays available to deployments that want one there.
-
-**Exit by prose or steering instead of a tool.** No artifact and no approval moment — the tool's argument IS the reviewable plan, and its review question is what gives the human a structured yes/no attached to the exact transition.
-
-## Consequences
-
-What holds now, pinned by the unit, protocol, snapshot, and real-API tiers:
-
-- The mode in force is a pure function of the session log: resume and fork restore it with no extra machinery, and a `mode/set` is followed by a matching complete `request/header` on the next changed step.
-- A user-driven flip narrates exactly once at the next boundary and a net-zero flip sequence narrates nothing; a tool-driven exit narrates only through its tool result.
-- In default mode the plugin contributes no mode section but does contribute the stable `exit_plan_mode` schema; a deployment without `dsh-mode` lacks that binding.
-- Native tool schemas and Code Mode's SDK stay byte-identical across default, plan, and custom-mode transitions; only the configured guidance section changes.
-- Plan mode changes nothing on the enforcement axes: the toolset, the sandbox mode, escalation, and the approval policy behave identically in plan and default — pairing the mode with the independent sandbox/approval knobs is how a deployment hardens planning.
-- Mode definitions are changeable from `cordis.yml` with no code edit; the complete plan instructions are required there, while missing plan config, malformed definitions, and unknown keys fail at load and unknown mode names fail at `set()`.
-- `exit_plan_mode` is always advertised, rejects outside plan, drops only plan guidance after approval, and carries keep-planning feedback in a corrective `isError`; ACP mode updates and each surface's user-interaction provider carry the human side.
-- The docs tail shipped with the landing: READMEs, regenerated catalogs (persistence log, config, cordis services, tools), the packages map and architecture rows, and the cookbook row.
-
-The accepted costs: a pending user flip set while idle is lost if the process dies before the next turn (the UI re-applies; the idle-record primitive is the escape hatch if this bites in practice). A mode transition changes the system prompt at order 50, so the cache path from that point onward changes, but the tool schemas and Code Mode SDK no longer churn. **A mode restrains by guidance alone**: a model that ignores the section CAN mutate during plan — the review moment, the session log, and independent sandbox, approval, and filesystem policies are the containment surface. Hardening planning means setting those knobs, not widening the mode; the removed enforcement shapes and their effects-declaration restart trigger remain in [Alternatives considered](#alternatives-considered) and [Deferred](#deferred). The ACP mode surface carries the picker while sandbox, approval, and model selectors remain config options under the division pinned in the [FAQ](#faq) and [feature matrix](../../../../packages/ui/acp/acp-feature-support.md). If ACP removes session modes in favor of config options, the picker mapping can migrate without changing the logged mode state or model surface.
diff --git a/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.i18n.yaml b/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.i18n.yaml
index da1ff70e09..d4b48d4f41 100644
--- a/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.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
-2026-07-10-sqlite-session-query-provider.md: be8795b609b52eeb03268c4986b52004eef0dba9
+2026-07-10-sqlite-session-query-provider.md: 98618a7eb572ce59c5fa5984675c9dc57b3f4289
2026-07-10-sqlite-session-query-provider.zh.md: bb3650da907cf86a853f748fa0ee40d5c2168709
diff --git a/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.md b/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.md
index be8795b609..98618a7eb5 100644
--- a/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.md
+++ b/.agents/notes/implemented/feature/2026-07-10-sqlite-session-query-provider.md
@@ -34,13 +34,13 @@ Both persistent and live FTS5 tables use `unicode61`. The implementation experim
The shared extractor includes message text, reasoning, nested tool-call/result content, tool names and arguments, blocked-prompt reasons, todo status/content, and error or terminal status detail. Structural boundaries, stream chunks, request headers, successful completion markers, and unknown declaration-merged event/content variants produce no document. Surface classification reuses `foldSurface()` so search agrees with model-history derivation.
-One serialized operation reads the provider-neutral `SessionPersistence` snapshot listing, compares each source-qualified opaque revision with the revision stored beside the indexed session, loads only new or changed logs, reconciles rows in one transaction, and executes the query. It never calls the backend's mutating `load()` for an id currently owned by `ctx.sessions`; the TEMP overlay records persisted availability, and the durable base refreshes after the live owner detaches. A revision identifies its backing persistence store as well as the backend-local log revision, so reopening against the same store reuses indexed rows while switching to an independent store cannot collide on a session id and local counter. Observation repeats when listing changes during a load; this incorporates a mutating load repair's refreshed revision before commit. Repeated queries and unchanged reopen load no full persisted logs. New, changed, and deleted sessions update on the next stable search. A source or extraction failure cannot mark a row current, and a transaction failure rolls back so a later search retries.
+One serialized operation reads the provider-neutral `SessionPersistence` snapshot listing, compares each source-qualified opaque revision with the revision stored beside the indexed session, loads only new or changed logs, reconciles rows in one transaction, and executes the query. It passes the caller's exact abort signal into snapshot listing and non-mutating inspection, directly awaits every started backend operation, and checks cancellation after each await and before starting more work. Cancellation therefore rejects only after active backend work is quiescent, starts no subsequent observation or reconciliation step, and keeps a following search serialized behind cleanup even if a backend ignores the signal. The operation never calls the backend's mutating `load()` for an id currently owned by `ctx.sessions`; the TEMP overlay records persisted availability, and the durable base refreshes after the live owner detaches. A revision identifies its backing persistence store as well as the backend-local log revision, so reopening against the same store reuses indexed rows while switching to an independent store cannot collide on a session id and local counter. Observation repeats when listing changes during a load; this incorporates a mutating load repair's refreshed revision before commit. Repeated queries and unchanged reopen load no full persisted logs. New, changed, and deleted sessions update on the next stable search. A source or extraction failure cannot mark a row current, and a transaction failure rolls back so a later search retries.
Persisted documents survive restarts. Live sessions use connection-local TEMP tables, shadow the persisted base for the same id, and reveal that base on detach. Closing the database drops live rows. Unmounting persistence hides durable rows without treating absence as authoritative deletion; remounting observes and reconciles the backend again. Conflicting immutable live and durable headers fail rather than combining sources.
The derived schema has its own application id and monotonic schema version. Persistent and TEMP session metadata store the integer `SessionHeader.createdAt` contract in strict `INTEGER` columns. A recognized incompatible version resets only this derived database. A database with a foreign application id or unrecognized user tables is refused before journal-mode mutation, which prevents an accidentally configured canonical session database from being changed. On POSIX filesystems, missing directories and database files are created owner-only so new SQLite sidecars inherit that mode; existing modes are preserved. One service in one process exclusively owns a derived-index path; cross-process writers are unsupported because generations and live TEMP shadow state are connection-owned.
-Cancellation rejects queued operations and caller waits around asynchronous source observation without committing an aborted observation. Node's synchronous `DatabaseSync` MATCH call cannot be interrupted once it is executing on the JavaScript thread, so the service checks the signal at serialized boundaries but does not promise mid-statement preemption.
+Cancellation rejects queued operations promptly. Once asynchronous source observation starts, the caller waits for that backend promise to settle before rejection, without committing an aborted observation or starting more source/index work. Node's synchronous `DatabaseSync` metadata and MATCH calls cannot be interrupted once executing on the JavaScript thread, so the service checks the signal around those calls but does not promise mid-statement preemption.
## Alternatives considered
@@ -54,6 +54,6 @@ Cancellation rejects queued operations and caller waits around asynchronous sour
Search has a small provider-neutral API while its only backend owns every derived-index state transition. The separate database adds configuration and a lightweight snapshot read before queries, but index corruption, reset, and tokenizer changes cannot endanger canonical logs. Durable revisions avoid full-log reads and rewrites for unchanged sessions; TEMP live overlays preserve current-session truth without making uncheckpointed events durable.
-The chosen tokenizer supports short tokens with a smaller index but does not promise substring recall. Literal phrases make query syntax safe and predictable at the cost of excluding boolean/full MATCH expressions. Cancellation is effective while queued or awaiting sources, but synchronous SQLite execution remains a non-preemptible section.
+The chosen tokenizer supports short tokens with a smaller index but does not promise substring recall. Literal phrases make query syntax safe and predictable at the cost of excluding boolean/full MATCH expressions. Cancellation is prompt while queued and quiescent while awaiting sources; synchronous SQLite execution remains a non-preemptible section bracketed by signal checks.
Unit coverage pins extraction, filters, both search scopes, all default surfaces, metadata-before-ranking, snippets, literal escaping, deterministic ties, complete pagination, scoped cursor invalidation, dynamic persistence mount/unmount, restart reconciliation, live shadow/reveal/reopen, schema safety, rollback retry, and queued/in-flight source-wait cancellation. A keyless real-Loader-path test combines the package with the real SQLite persistence backend.
diff --git a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.md b/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.md
deleted file mode 100644
index 189f75fc12..0000000000
--- a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.md
+++ /dev/null
@@ -1,59 +0,0 @@
-# Agent Note: Optional time-context plugin
-
-Status: implemented
-
-English | [中文](2026-07-14-time-context-plugin.zh.md)
-
-## Problem
-
-The dynamic system-prompt storage and refresh decision in this record is superseded by [Durable per-step time context](2026-07-16-durable-per-step-time-context.md). The opt-in package, zoned formatting, and validation remain; the follow-up owns the current model-visible and durability contract.
-
-An agent request has no live clock unless a deployment puts one in prompt text or gives the model a query tool. Static text becomes stale, while a tool call adds overhead to ordinary reasoning about dates, deadlines, or idle time. Without elapsed time, the model cannot distinguish an immediate follow-up from one sent hours after the preceding message.
-
-Prompt assembly can derive both facts per step from durable session timestamps, and request-header logging can record the exact rendered value. Accumulating stale readings in conversation history or waking idle agents would violate the existing request lifecycle.
-
-## Decision
-
-`@deepseek-ai/dsh-time-context` is an opt-in function plugin at `packages/context/time-context/`. The `context/` product group holds bounded request-context enrichments that define neither a tool nor a service. `dsh-agent-spine-demo` and shipped examples do not load the package; deployments mount it explicitly when its token and disclosure costs are acceptable.
-
-The plugin registers the global `context:time` system-prompt section at order 10, after the deployment persona and before tool guidance. For an active turn it emits an ISO-shaped timestamp with numeric UTC offset and IANA zone, plus a compact whole-second duration since the last model-visible message before the turn opened. Bare and idle assemblies receive an empty section.
-
-### Previous-message baseline
-
-At a turn's first assembly, the provider scans before `turn/start` for the latest `user/message`, `assistant/message`, `tool/result`, `context/message`, or `steering/message`. It excludes the current prompt so the duration expresses the inter-turn gap instead of approximately zero. Every refresh in that turn keeps the same baseline, and the first turn reports `unavailable (no earlier message in this session)`.
-
-The baseline is the session event's append time, not an unlogged client timestamp. Resume and fork behavior are therefore deterministic from the durable log, and the model-visible value remains reconstructable without a new event. A backward wall-clock adjustment clamps the duration to zero.
-
-### Refresh policy
-
-`refreshIntervalMs` defaults to 60,000 and must be a non-negative safe integer. Every turn's first request refreshes. Later assemblies in that turn reuse the block until its age reaches the interval; `0` refreshes every step. No timer creates work during model calls, tools, or idle time because refresh is request-bound.
-
-When `timeZone` is omitted, `Intl.DateTimeFormat` resolves the Node process's system zone once at plugin load. Node honors `TZ`; without that override, the host or container supplies the zone. An explicit value must be an IANA identifier and is validated at load. The captured zone remains stable until plugin reload, and the ISO-shaped local timestamp includes its current numeric offset so daylight-saving changes stay explicit. This is the deployment process's zone, not a remote user's zone.
-
-### Logging and token shape
-
-The loop records the temporal block in full `request/header` snapshots before transmission, satisfying the [reconstructable-requests contract](../architecture/2026-07-05-reconstructable-requests.md). Each request carries one current block; earlier readings do not remain in conversation history. The plugin owns the fact and contributes it through the prompt registry, following the [prompt-variables Agent Note](../architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md) without a loop special case.
-
-## Testing
-
-Unit tests pin formatting, baselines, refresh policy, validation, per-agent state, disposal, and load-time system-zone capture. A real agent-loop test pins the transmitted prompt and full `request/header` snapshots. A keyless subprocess e2e boots a test-only `cordis.yml` through the real Loader and stdio app, omits `timeZone` under a controlled `TZ`, drives two turns, and verifies the persisted request headers externally. Default snapshot compositions omit the plugin, so their transcript fixtures contain no temporal block.
-
-## Alternatives considered
-
-- **Append a `context/message` on every turn or refresh** — rejected because readings and token cost would accumulate in history. Replacing a prior surface node would preserve its old position, while replacing the tail would hide intervening conversation.
-- **Use `agent/session-prefix`** — rejected because the session-stable prefix cannot represent a per-turn or per-step clock.
-- **Mutate requests in `agent/request`** — rejected because that seam shapes call config after the message boundary; inserted model content would bypass prompt-pressure accounting and request-header logging.
-- **Register separate `{{current_time}}` and `{{elapsed}}` variables** — rejected because independent providers can sample different instants and require shared caching. One section records the pair atomically without a deployment-authored template.
-- **Refresh from a background timer** — rejected because a new value has no consumer outside request assembly. Timer-driven `agent.inject()` would create turns and wake idle sessions merely to report time passing.
-- **Keep UTC as the omitted default** — rejected because an explicitly enabled clock should follow its deployment environment unless the operator chooses UTC. `timeZone: UTC` remains available when a deployment requires it.
-- **Add a time-zone detection library** — rejected because Node's `Intl` runtime already exposes the process's IANA zone. Another dependency cannot infer a remote user's zone either.
-- **Mount the plugin in `dsh-agent-spine-demo`** — rejected because time zone, disclosure, token budget, and freshness are deployment policy. Opt-in keeps default context stable.
-- **Place the package in `core/`** — rejected because `core/` owns the product API spine, while this plugin is an optional leaf with no service key.
-
-## Consequences
-
-- Opted-in models receive a zoned clock and inter-turn duration without a tool call. The system-prompt cost is fixed per request instead of growing with the session.
-- An omitted `timeZone` follows the process's `TZ`, host, or container zone as observed at plugin load. Operators must configure an explicit zone when the deployment environment does not represent the intended user.
-- A refresh changes the request header and can add a full `request/header` snapshot with reason `change`. `refreshIntervalMs` trades freshness against the number and size of durable full snapshots; `0` records a new value on every step whose whole-second rendering changes.
-- No request exists solely to refresh time. A long-running tool leaves the prior reading until the next step assembles.
-- Duration reflects harness processing time at durable append boundaries, not client-network latency before logging. Preserving a client-origin timestamp requires a separate durable input contract.
diff --git a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.zh.md b/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.zh.md
deleted file mode 100644
index 12671cb891..0000000000
--- a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.zh.md
+++ /dev/null
@@ -1,59 +0,0 @@
-# Agent Note:可选时间上下文插件
-
-Status: implemented
-
-[English](2026-07-14-time-context-plugin.md) | 中文
-
-## 问题
-
-本记录中的动态系统提示词存储和刷新决策已由[持久的逐步骤时间上下文](2026-07-16-durable-per-step-time-context.md)取代。需要显式启用的包(package)、分区时间格式和校验仍然保留;后续 Agent Note 负责当前的模型可见与持久性契约。
-
-如果部署方既未在提示词中提供时钟,也未给模型提供查询工具,agent(智能体)请求就无法获得实时准确的时间。静态文本会变得陈旧,而对于日期、截止时间或闲置时长等常规推理,调用工具会增加开销。缺少已经过去的时长时,模型无法区分紧接着发送的消息与上一条消息几小时后才发送的消息。
-
-提示词组装流程可以在每个步骤中根据持久会话时间戳派生这两项信息,请求头日志则可以记录实际渲染的确切值。在会话历史中累积陈旧读数或唤醒空闲 agent 都会违反现有请求生命周期。
-
-## 决策
-
-`@deepseek-ai/dsh-time-context` 是位于 `packages/context/time-context/`、需要显式启用的函数插件。`context/` 产品分组用于容纳既不定义工具、也不定义服务的有界请求上下文增强。`dsh-agent-spine-demo` 和仓库提供的示例都不会加载该包;只有当 token 与信息披露成本可接受时,部署方才显式挂载它。
-
-该插件注册顺序值为 10 的全局系统提示词区段 `context:time`,位置在部署方角色设定之后、工具指导之前。对于活跃轮次,它会输出带数字 UTC 偏移和 IANA 时区、形似 ISO 的时间戳,以及从轮次开始前最后一条模型可见消息起算的紧凑整秒时长。未绑定 agent 或 agent 处于空闲状态时,该区段为空。
-
-### 上一条消息基线
-
-在轮次首次组装时,提供方会在 `turn/start` 之前查找最近的 `user/message`、`assistant/message`、`tool/result`、`context/message` 或 `steering/message`。它会排除当前提示词,使时长表达轮次间隔,而不是接近零。同一轮次中的每次刷新都保留这条基线;首个轮次报告 `unavailable (no earlier message in this session)`。
-
-基线采用会话事件的追加时间,而不是日志中不存在的客户端时间戳。因此,恢复和 fork 行为可以从持久日志中确定性重现,模型可见值也无需新增事件即可重建。系统挂钟向后调整时,插件会将时长钳制为零。
-
-### 刷新策略
-
-`refreshIntervalMs` 默认值为 60,000,并且必须是非负安全整数。每个轮次的首次请求都会刷新。同一轮次中的后续组装会复用该区块,直至其存在时间达到该间隔;设为 `0` 时每个步骤都刷新。刷新仅由请求驱动,因此在模型调用、工具运行或空闲期间,计时器不会创建任务。
-
-省略 `timeZone` 时,`Intl.DateTimeFormat` 会在插件加载时解析一次 Node 进程的系统时区。Node 会遵循 `TZ`;没有该覆盖值时,时区由主机或容器提供。显式值必须是 IANA 标识符,并在加载时接受校验。捕获的时区在插件重新加载前保持稳定,形似 ISO 的本地时间戳包含其当前数字偏移,使夏令时变化保持显式可见。该默认值代表部署进程的时区,而不是远程用户的时区。
-
-### 日志与 token 形态
-
-agent loop(智能体循环)会在发送前通过完整的 `request/header` 快照记录时间区块,从而满足[可重建请求契约](../architecture/2026-07-05-reconstructable-requests.md)。每个请求只携带一个当前区块;先前的读数不会保留在会话历史中。该插件拥有时间信息,并按照[提示词变量 Agent Note](../architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md)通过提示词注册表贡献该信息,无需为循环添加特殊分支。
-
-## 测试
-
-单元测试固定格式化、基线、刷新策略、校验、逐 agent 状态、资源释放行为,以及系统时区在加载时的捕获行为。使用真实 agent loop 的测试固定实际发送的提示词和完整的 `request/header` 快照。无密钥子进程端到端测试通过真实 Loader 和 stdio 应用启动测试专用 `cordis.yml`,在受控 `TZ` 下省略 `timeZone`,驱动两个轮次,并从外部校验持久请求头。默认快照组合不包含该插件,因此其中的 transcript(文本记录)fixture(测试前置数据)不包含时间区块。
-
-## 考虑过的替代方案
-
-- **每个轮次或每次刷新都追加一条 `context/message`**——不予采纳,因为读数和 token 成本会在历史中累积。替换先前的表层节点会保留其旧位置,而替换尾部节点会隐藏中间的会话内容。
-- **使用 `agent/session-prefix`**——不予采纳,因为会话期间保持稳定的前缀无法表示逐轮次或逐步骤变化的时钟。
-- **在 `agent/request` 中修改请求**——不予采纳,因为该边界在消息边界之后塑造调用配置;插入模型可见内容会绕过提示词压力核算和请求头日志。
-- **注册独立的 `{{current_time}}` 和 `{{elapsed}}` 变量**——不予采纳,因为独立提供方可能在不同时间点采样,并且需要共享缓存。单个区段会以原子方式记录两项信息,也不需要部署方编写时间模板。
-- **通过后台计时器刷新**——不予采纳,因为请求组装之外没有消费新值的对象。由计时器驱动 `agent.inject()` 会创建轮次,并且只为报告时间流逝就唤醒空闲会话。
-- **省略配置时仍默认使用 UTC**——不予采纳,因为显式启用的时钟应跟随部署环境,除非运维方选择 UTC。需要 UTC 的部署仍可配置 `timeZone: UTC`。
-- **引入时区探测库**——不予采纳,因为 Node 的 `Intl` 运行时已经能够提供进程的 IANA 时区,而且额外依赖同样无法推断远程用户的时区。
-- **在 `dsh-agent-spine-demo` 中挂载插件**——不予采纳,因为时区、信息披露、token 预算和新鲜度都属于部署策略。选择加入能保持默认上下文稳定。
-- **将包放入 `core/`**——不予采纳,因为 `core/` 负责产品 API 主干,而该插件是没有服务键的可选叶节点。
-
-## 后果
-
-- 选择加入的模型无需调用工具,即可获得分区时钟和轮次间隔时长。每个请求的系统提示词成本固定,不会随会话增长。
-- 省略 `timeZone` 时,插件采用加载时观察到的进程 `TZ`、主机或容器时区。当部署环境不能代表目标用户时,运维方必须显式配置时区。
-- 刷新会改变请求头,并可能新增一份 reason 为 `change` 的完整 `request/header` 快照。`refreshIntervalMs` 用新鲜度换取完整持久快照的数量与大小;设为 `0` 时,每个整秒渲染结果发生变化的步骤都会记录新值。
-- 系统不会仅为刷新时间而创建请求。长时间运行的工具会保留先前读数,直至下一步骤开始组装。
-- 时长反映持久追加边界处的 harness 处理时间,不包含消息进入日志之前的客户端网络延迟。若要保留客户端来源时间戳,需要单独的持久输入契约。
diff --git a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.i18n.yaml b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.i18n.yaml
index f037660761..910872881a 100644
--- a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-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
-2026-07-16-durable-per-step-time-context.md: 2d7076d51dbe1a64e5042230bddc6844141ff265
-2026-07-16-durable-per-step-time-context.zh.md: 432e0305cf44dcce1053c6580c9f0039309a7af4
+2026-07-16-durable-per-step-time-context.md: 4bc17b3c08707fcaa4f0f431e71ddbe567a03c9e
+2026-07-16-durable-per-step-time-context.zh.md: 836c0f83fbe9d6120741a261cf25ce7d8c227bdf
diff --git a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.md b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.md
index 2d7076d51d..4bc17b3c08 100644
--- a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.md
+++ b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.md
@@ -12,13 +12,13 @@ A process-local refresh cache makes displayed time depend on state that cannot s
## Decision
-`@deepseek-ai/dsh-time-context` is an opt-in function plugin in `packages/context/time-context/`. It registers a prepended `agent/pre-step` listener and, when an injection is due, calls `agent.inject()` for a pre-step attempt whose signal is not already aborted. The injected `context/message` carries source `{ kind: 'plugin', plugin: 'time-context' }` and append surface metadata; a suppressed attempt appends nothing.
+`@deepseek-ai/dsh-time-context` is an opt-in function plugin in `packages/context/time-context/`. The `context/` group holds bounded request-context enrichments that define neither a tool nor a service, and shipped examples do not mount this plugin because its time-zone disclosure and token cost are deployment policy. It registers a prepended `agent/pre-step` listener and, when an injection is due, calls `agent.inject()` for a pre-step attempt whose signal is not already aborted. The injected `user/message` carries source `{ kind: 'plugin', plugin: 'time-context' }` and append surface metadata; a suppressed attempt appends nothing.
The listener records preparation context before a possible `step/start`. Its prepended registration runs before ordinary automatic compaction listeners, so pressure estimation and any resulting surface rewrite observe a newly appended reading. A later pre-step listener can cancel or fail the attempt before the step opens; the reading remains because the durable log is append-only and this plugin performs no rollback.
The optional `timeZone` config resolves the Node process's IANA zone once at plugin load when omitted; an explicit value is validated by `Intl.DateTimeFormat`. The timestamp includes the numeric UTC offset and resolved IANA zone.
-The optional `refreshIntervalMs` config is manually validated at plugin load as a non-negative safe integer. Omission or `0` injects on every eligible preparation attempt. A positive value scans the raw session events for the most recent `context/message` with this plugin's source and injects when none exists, wall time moved backward, or the event is at least the configured age. The raw event timestamp governs even after compaction shadows the message, so scheduling persists across turns and process resume without a timer or process-local cache.
+The optional `refreshIntervalMs` config is manually validated at plugin load as a non-negative safe integer. Omission or `0` injects on every eligible preparation attempt. A positive value scans the raw session events for the most recent `user/message` with this plugin's source and injects when none exists, wall time moved backward, or the event is at least the configured age. The raw event timestamp governs even after compaction shadows the message, so scheduling persists across turns and process resume without a timer or process-local cache.
### Text and elapsed baselines
@@ -29,7 +29,7 @@ Time sampled while preparing turn , step 1:
Elapsed since the preceding model-visible message: .
```
-The baseline is the latest preceding user, assistant, tool-result, context, or steering message. This includes the accepted prompt that opened an ordinary message turn. If no model-visible message exists, the duration is `unavailable`.
+The baseline is the latest preceding user, assistant, tool-result, or steering message. This includes the accepted prompt that opened an ordinary message turn. If no model-visible message exists, the duration is `unavailable`.
An injected later-step reading is:
@@ -48,11 +48,7 @@ The plugin contributes nothing to system-prompt assembly. `request/header` conta
## Testing
-Unit and real-loop tests pin formatting, both elapsed baselines, interval omission and zero, threshold boundaries, cross-turn and per-session scheduling, backward-clock behavior, invalid config, resumed raw-event lookup after compaction, aborted-signal behavior, later-listener cancellation and failure, listener disposal, source and surface metadata, cumulative multi-step visibility, and absence from request headers. A keyless subprocess e2e boots the real Loader and stdio app, drives two turns, and verifies the persisted context events externally.
-
-## Supersedes
-
-This decision supersedes the dynamic system-prompt storage and refresh policy in [Optional time-context plugin](2026-07-14-time-context-plugin.md). It keeps the package location, opt-in deployment stance, timestamp formatting, process-zone default, and load-time validation. Durable history replaces the `context:time` prompt section, process-local refresh cache, and request-header deltas; `refreshIntervalMs` controls durable append frequency instead of prompt replacement.
+Unit and real-loop tests pin formatting, both elapsed baselines, interval omission and zero, threshold boundaries, cross-turn and per-session scheduling, backward-clock behavior, invalid config, resumed raw-event lookup after compaction, aborted-signal behavior, later-listener cancellation and failure, listener disposal, source and surface metadata, cumulative multi-step visibility, and absence from request headers. A keyless subprocess e2e boots the real Loader with the Headless composition, drives two ordered one-shot turns, and verifies the persisted plugin-attributed messages externally.
## Alternatives considered
@@ -61,10 +57,13 @@ This decision supersedes the dynamic system-prompt storage and refresh policy in
- **Inject from a background timer** — rejected because idle time has no pending request to consume the value, and timer-driven injection would create durable turns solely to report time passing.
- **Expose time only through a tool** — rejected because ordinary temporal reasoning would require an avoidable tool round trip and would not guarantee a reading before every step.
- **Use `agent/session-prefix`** — rejected because one loop-instance prefix cannot represent distinct step timestamps and does not accumulate historically attributable readings.
+- **Mutate assembled requests or register independent prompt variables** — rejected because request-local insertion bypasses the durable surface and separate providers can sample different instants. One attributed context message records the timestamp and elapsed baseline atomically.
+- **Default to UTC or add a time-zone detection dependency** — rejected because an explicitly mounted plugin follows its process environment unless the operator selects an IANA zone, while no server-side library can infer a remote user's zone.
+- **Mount the plugin in shipped compositions or place it in `core/`** — rejected because disclosure, time zone, freshness, and history cost are deployment choices for an optional context leaf, not product-spine policy.
## Consequences
- Omission or `0` records every eligible preparation attempt; a positive interval reduces append frequency and history growth while preserving durable scheduling across resume.
- Timing context remains append-only until compaction shadows older surface nodes, including a preparation reading left by a later cancellation or failure.
- The first-step duration normally measures from the prompt that opened the turn, while later-step durations measure model and tool processing since the preceding step context.
-- An omitted `timeZone` still reflects the deployment process rather than a remote user, and elapsed time still uses durable harness append boundaries rather than client-origin timestamps.
+- An omitted `timeZone` still reflects the deployment process rather than a remote user, and elapsed time still uses durable harness append boundaries rather than client-origin timestamps. Supporting client-origin time requires a separate durable input contract.
diff --git a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.zh.md b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.zh.md
index 432e0305cf..836c0f83fb 100644
--- a/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-16-durable-per-step-time-context.zh.md
@@ -12,13 +12,13 @@ Status: implemented
## 决策
-`@deepseek-ai/dsh-time-context` 是位于 `packages/context/time-context/`、需要显式启用的函数插件。它注册一个前置的 `agent/pre-step` 监听器,并在需要注入时,为信号尚未取消的预步骤尝试调用 `agent.inject()`。注入的 `context/message` 携带来源 `{ kind: 'plugin', plugin: 'time-context' }` 和追加表层元数据;受间隔抑制的尝试不会追加任何内容。
+`@deepseek-ai/dsh-time-context` 是位于 `packages/context/time-context/`、需要显式启用的函数插件。`context/` 分组容纳有界的请求上下文增强,这些增强既不定义工具也不定义服务;已交付示例不挂载此插件,因为时区披露与 token 成本属于部署策略。它注册一个前置的 `agent/pre-step` 监听器,并在需要注入时,为信号尚未取消的预步骤尝试调用 `agent.inject()`。注入的 `user/message` 携带来源 `{ kind: 'plugin', plugin: 'time-context' }` 和追加表层元数据;受间隔抑制的尝试不会追加任何内容。
监听器在可能出现的 `step/start` 之前记录准备上下文。它采用前置注册,因此先于普通自动压缩监听器运行,使压力估算和由此产生的表层重写都能观察到新追加的读数。后续预步骤监听器可能在步骤开启前取消尝试或使其失败;持久日志仅追加,且本插件不执行回滚,因此该读数会保留下来。
省略可选配置 `timeZone` 时,插件在加载时解析一次 Node 进程的 IANA 时区;显式值由 `Intl.DateTimeFormat` 校验。时间戳包含数字 UTC 偏移和解析后的 IANA 时区。
-插件在加载时手动校验可选配置 `refreshIntervalMs`,其值必须为非负安全整数。省略或设为 `0` 时,每次符合条件的准备尝试都会注入。设为正数时,插件扫描原始会话事件,查找来源属于本插件的最新 `context/message`;不存在此类事件、系统挂钟向后移动,或该事件已达到配置时长时,插件执行注入。即使压缩已隐藏消息,调度仍以原始事件时间戳为准,因此该机制无需计时器或进程本地缓存,也能跨轮次和进程恢复持续生效。
+插件在加载时手动校验可选配置 `refreshIntervalMs`,其值必须为非负安全整数。省略或设为 `0` 时,每次符合条件的准备尝试都会注入。设为正数时,插件扫描原始会话事件,查找来源属于本插件的最新 `user/message`;不存在此类事件、系统挂钟向后移动,或该事件已达到配置时长时,插件执行注入。即使压缩已隐藏消息,调度仍以原始事件时间戳为准,因此该机制无需计时器或进程本地缓存,也能跨轮次和进程恢复持续生效。
### 文本与时长基线
@@ -29,7 +29,7 @@ Time sampled while preparing turn , step 1:
Elapsed since the preceding model-visible message: .
```
-基线是前一条用户消息、助手消息、工具结果、上下文消息或 steering(中途引导)消息。对于普通消息轮次,这包括开启轮次的已接受提示词。如果不存在模型可见消息,时长为 `unavailable`。
+基线是前一条用户消息、助手消息、工具结果或 steering(中途引导)消息。对于普通消息轮次,这包括开启轮次的已接受提示词。如果不存在模型可见消息,时长为 `unavailable`。
后续步骤的注入读数为:
@@ -48,11 +48,7 @@ Elapsed since the preceding step context: .
## 测试
-单元测试和真实 agent loop(智能体循环)测试固定格式化、两种时长基线、间隔省略和零值、阈值边界、跨轮次和各会话独立调度、挂钟后退行为、无效配置、压缩后基于恢复会话的原始事件查找、已取消信号行为、后续监听器取消和失败、监听器 dispose(资源释放)、来源与表层元数据、多步骤累计可见性,以及请求头中不存在时间上下文。无密钥子进程 e2e 测试通过真实 Loader 和 stdio 应用启动,驱动两个轮次,并从外部校验持久化的上下文事件。
-
-## 取代的决策
-
-本决策取代[可选时间上下文插件](2026-07-14-time-context-plugin.md)中的动态系统提示词存储和刷新策略。它保留包位置、选择加入式部署、时间戳格式、进程时区默认值和加载时校验。持久历史取代 `context:time` 提示词区段、进程本地刷新缓存和请求头增量;`refreshIntervalMs` 用于控制持久追加频率,而非提示词替换。
+单元测试和真实 agent loop(智能体循环)测试固定格式化、两种时长基线、间隔省略和零值、阈值边界、跨轮次和各会话独立调度、挂钟后退行为、无效配置、压缩后基于恢复会话的原始事件查找、已取消信号行为、后续监听器取消和失败、监听器 dispose(资源释放)、来源与表层元数据、多步骤累计可见性,以及请求头中不存在时间上下文。无密钥子进程 e2e 测试使用 Headless 组合启动真实 Loader,依次驱动两个单次任务轮次,并从外部校验持久化且来源归属于插件的消息。
## 考虑过的替代方案
@@ -61,10 +57,13 @@ Elapsed since the preceding step context: .
- **通过后台计时器注入**——不予采纳,因为空闲期间没有待处理请求消费该值,而且计时器驱动的注入会仅为报告时间流逝而创建持久轮次。
- **只通过工具提供时间**——不予采纳,因为普通时间推理会产生本可避免的工具往返,也不能保证每个步骤之前都有读数。
- **使用 `agent/session-prefix`**——不予采纳,因为一个 loop 实例前缀无法表示不同的步骤时间戳,也不会累计具有历史归属的读数。
+- **修改已组装的请求或注册独立提示词变量**——不予采纳,因为请求内插入会绕过持久表层,不同提供方也可能在不同时间采样。一条带来源归属的上下文消息会原子地记录时间戳和时长基线。
+- **默认使用 UTC 或增加时区检测依赖**——不予采纳,因为显式挂载的插件默认遵循其进程环境,除非操作方选择 IANA 时区,而任何服务端库都无法推断远程用户的时区。
+- **在已交付组合中挂载插件,或把它放进 `core/`**——不予采纳,因为披露内容、时区、新鲜度和历史成本是可选上下文叶节点的部署选择,不是产品主干策略。
## 后果
- 省略 `refreshIntervalMs` 或设为 `0` 时,每次符合条件的准备尝试都会留下记录;正数间隔会减少追加频率和历史增长,同时使持久调度在恢复后继续生效。
- 时间上下文仅追加并保留到压缩隐藏旧表层节点为止,其中也包括后续取消或失败所留下的准备读数。
- 第一个步骤的时长通常从开启轮次的提示词起算,后续步骤的时长则反映自上一条步骤上下文以来的模型与工具处理时间。
-- 省略 `timeZone` 时仍采用部署进程而非远程用户的时区,时长仍采用 harness 的持久追加边界而非客户端来源时间戳。
+- 省略 `timeZone` 时仍采用部署进程而非远程用户的时区,时长仍采用 harness 的持久追加边界而非客户端来源时间戳。若要支持客户端来源的时间,需要另行建立持久输入契约。
diff --git a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.i18n.yaml b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.i18n.yaml
index 1a8c03cb52..ad0592ae57 100644
--- a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.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
-2026-07-16-harness-level-loop.md: 9a9511b9dcea1b5fdc90f4fc716c4399f2346967
-2026-07-16-harness-level-loop.zh.md: 284e73051eaaa4633b9f56367de9096dadc8184e
+2026-07-16-harness-level-loop.md: 15b5ce7e20b7afc429f6ff7b8a4d2d69150c22a0
+2026-07-16-harness-level-loop.zh.md: a3fabe40d36c45c715f613ce8def35faa427d3bd
diff --git a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.md b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.md
index 9a9511b9dc..15b5ce7e20 100644
--- a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.md
+++ b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.md
@@ -39,7 +39,7 @@ Time-based `/loop` or scheduled execution is a third policy and is not implement
| `@deepseek-ai/dsh-tool-goal` | `packages/goal/tool-goal/`, model-facing consumer | Registers exclusive `get_goal`, `create_goal`, and `update_goal`; authenticates live turn provenance and narrows autonomous-round authority to completion or blocking reports with machine-routable reason codes. |
| `@deepseek-ai/dsh-goal-session` | `packages/goal/goal-session/`, continuation policy | Reserves, fences, admits, attributes, settles, cancels, and quiescently drains same-session goal rounds without importing the concrete loop. |
| `@deepseek-ai/dsh-commands` | `packages/ui/commands/`, UI registry | Owns `CommandDefinition`, discovery, scoped registration, direct dispatch, `CommandResult`, and request cancellation for human-only commands. |
-| `@deepseek-ai/dsh-command-goal` | `packages/goal/command-goal/`, human-command producer | Registers `/goal` status, creation, edit, pause, resume, and clear over the goal domain for TUI and ACP. |
+| `@deepseek-ai/dsh-command-goal` | `packages/goal/command-goal/`, human-command producer | Registers `/goal` status, creation, edit, pause, resume, and clear over the goal domain for TUI. |
| `@deepseek-ai/dsh-tool-ralph` | `packages/workflow/tool-ralph/`, fixed workflow consumer | Registers `ralph({ objective, maxRounds? })`, validates the fresh structured provider and bounded `RalphRoundReport`, and returns `complete`, `blocked`, or `budget-limited`. |
The detailed contracts live in the [goal-domain](2026-07-19-persisted-same-session-goal-domain.md), [model goal-tools](2026-07-19-model-facing-goal-tools.md), [goal-round driver](2026-07-19-same-session-goal-round-driver.md), [command registry](2026-07-19-plugin-command-registration.md), [human goal-command](2026-07-19-human-goal-command.md), and [Ralph workflow-tool](2026-07-19-fresh-agent-ralph-workflow-tool.md) Agent Notes.
@@ -70,7 +70,7 @@ The human UX follows the compact Codex shape in the [public OpenAI Codex TUI dis
The model receives only `get_goal`, `create_goal`, and `update_goal`. It may create a goal when a direct human request clearly asks for substantial multi-round work, and it may infer that intent in any language. It must not turn routine one-turn work into a goal. Direct-human provenance is enforced in code; semantic interpretation remains model judgment. An autonomous goal round may report `complete` or `blocked` for the exact current goal round but cannot edit, pause, resume, or replace the human objective.
-TUI and ACP mount the shared command registry and complete goal stack by default and expose `/goal` through one producer. Every effective registered command is discoverable and invocable through every composed command adapter; a plugin incompatible with an application omits its command producer from that composition rather than relying on registry-level surface masks. The UI-less agent spine is opt-in so one-shot callers do not silently become multi-round operations. The headless CLI and JSON-RPC front doors do not consume the command plane; ordinary human text can still authorize model goal tools when that stack is composed.
+TUI mounts the shared command registry and complete goal stack by default and exposes `/goal` through one producer. ACP mounts the goal domain, model tools, and same-session driver but deliberately omits the human command plane. Every effective registered command is discoverable and invocable through every composed command adapter; a plugin incompatible with an application omits its command producer from that composition rather than relying on registry-level surface masks. The UI-less agent spine is opt-in so one-shot callers do not silently become multi-round operations. The headless CLI and JSON-RPC front doors do not consume the command plane; ordinary human text can still authorize model goal tools when that stack is composed.
### Fresh-agent Ralph execution
@@ -94,7 +94,7 @@ External products are comparators, not compatibility targets. The local source s
### Verification
-The six owning Agent Notes record unit, integration, process, snapshot, cancellation, replay, and built-runtime coverage. The stack exercises strict goal-record folding, compare-and-set races, session fork inheritance, disarmed restoration, natural-language direct-human authority, configurable caps and blocked thresholds, exact goal-round attribution, adapter-wide command discovery, and transcript isolation. Shipped keyless snapshots cover model goal creation/inspection through the headless app, multi-round same-session lifecycle and cancellation through ACP, direct `/goal` status without a model turn, and two real Ralph rounds through the headless app. The Ralph snapshot boots the worker-thread engine, spawn provider, structured-output runtime, and agent loop, then inspects distinct unseeded child logs and exact one-way bounded handoff while pinning the parent stream. Focused real-stack tests additionally cover completion, blocker and round-limit outcomes, malformed and oversized reports, ordinary child failure with the last good handoff, one phase event, and cancellation to child quiescence. Package sources remain under the repository's per-file 100% coverage gate, and built-binary tests cover installed-artifact resolution. The implementation experience is recorded in the root testing policy: every non-trivial model- or human-visible change must carry a real-example keyless snapshot in the same PR rather than relying on package-only or mock-only fixture coverage.
+The six owning Agent Notes record unit, integration, process, snapshot, cancellation, replay, and built-runtime coverage. The stack exercises strict goal-record folding, compare-and-set races, session fork inheritance, disarmed restoration, natural-language direct-human authority, configurable caps and blocked thresholds, exact goal-round attribution, adapter-wide command discovery, and transcript isolation. Shipped keyless snapshots cover model goal creation/inspection through the headless app, multi-round same-session lifecycle and cancellation through ACP, and two real Ralph rounds through the headless app; focused command tests pin direct `/goal` status without a model turn. The Ralph snapshot boots the worker-thread engine, spawn provider, structured-output runtime, and agent loop, then inspects distinct unseeded child logs and exact one-way bounded handoff while pinning the parent stream. Focused real-stack tests additionally cover completion, blocker and round-limit outcomes, malformed and oversized reports, ordinary child failure with the last good handoff, one phase event, and cancellation to child quiescence. Package sources remain under the repository's per-file 100% coverage gate, and built-binary tests cover installed-artifact resolution. The implementation experience is recorded in the root testing policy: every non-trivial model- or human-visible change must carry a real-example keyless snapshot in the same PR rather than relying on package-only or mock-only fixture coverage.
## Alternatives considered
@@ -126,4 +126,4 @@ The six owning Agent Notes record unit, integration, process, snapshot, cancella
- **No goal reflector** — concern events, automatic no-progress heuristics, goal revision by an independent reflector, stuck-pattern detection, and `loop_split` are not implemented. Humans can edit, pause, clear, or resume the goal directly.
- **Ralph policy remains narrow** — one round creates one fresh child; within-round fan-out, evaluator/worker role separation, dynamic provider/model selection, and structural recursive-Ralph tool denial need separate policy surfaces. Prompt guidance is not enforcement.
- **Ralph does not retry a failed child** — an ordinary failure preserves the failed round and last good handoff, while fatal workflow infrastructure failures can end before that state is available. Retry count, backoff, and richer failure transport need separate policy and seam design.
-- **Portable UI remains modest** — TUI and ACP render plain-text goal status and generic Ralph cards. There is no continuous status widget, reconnectable command output, modal goal editor, or command plane in the headless CLI or JSON-RPC front doors.
+- **Portable UI remains modest** — TUI renders plain-text goal status and generic Ralph cards. ACP carries only committed assistant text; there is no continuous status widget, reconnectable command output, modal goal editor, or command plane in ACP, the headless CLI, or JSON-RPC.
diff --git a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.zh.md b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.zh.md
index 284e73051e..a3fabe40d3 100644
--- a/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-16-harness-level-loop.zh.md
@@ -39,7 +39,7 @@ Status: implemented
| `@deepseek-ai/dsh-tool-goal` | `packages/goal/tool-goal/`,面向模型消费者 | 注册互斥的 `get_goal`、`create_goal` 与 `update_goal`;认证实时 Turn 来源,并把自治 Round 权限收窄到带机器可路由原因代码的完成或阻塞报告。 |
| `@deepseek-ai/dsh-goal-session` | `packages/goal/goal-session/`,续行策略 | 在不导入具体 loop 的情况下,预留、设围栏、接纳、归属、结算、取消并静止排空同会话目标回合。 |
| `@deepseek-ai/dsh-commands` | `packages/ui/commands/`,UI 注册表 | 拥有面向人类专用命令的 `CommandDefinition`、发现、作用域注册、直接分发、`CommandResult` 与请求取消。 |
-| `@deepseek-ai/dsh-command-goal` | `packages/goal/command-goal/`,人类命令生产方 | 为 TUI 和 ACP 注册构建在目标领域之上的 `/goal` 状态、创建、编辑、暂停、恢复与清除。 |
+| `@deepseek-ai/dsh-command-goal` | `packages/goal/command-goal/`,人类命令生产方 | 为 TUI 注册构建在目标领域之上的 `/goal` 状态、创建、编辑、暂停、恢复与清除。 |
| `@deepseek-ai/dsh-tool-ralph` | `packages/workflow/tool-ralph/`,固定工作流消费者 | 注册 `ralph({ objective, maxRounds? })`,验证全新结构化 provider 与有界 `RalphRoundReport`,并返回 `complete`、`blocked` 或 `budget-limited`。 |
详细契约分别由[目标领域](2026-07-19-persisted-same-session-goal-domain.md)、[模型目标工具](2026-07-19-model-facing-goal-tools.md)、[目标回合驱动器](2026-07-19-same-session-goal-round-driver.md)、[命令注册表](2026-07-19-plugin-command-registration.md)、[人类目标命令](2026-07-19-human-goal-command.md)与 [Ralph 工作流工具](2026-07-19-fresh-agent-ralph-workflow-tool.md) Agent Note 拥有。
@@ -70,7 +70,7 @@ fork 会话会继承持久目标前缀,因为这是自然的重放结果。for
模型只接收 `get_goal`、`create_goal` 和 `update_goal`。当直接人类请求清楚要求大量多 Round 工作时,模型可以创建目标,并且可以从任何语言推断该意图。它不得把日常单 Turn 工作变成目标。直接人类来源由代码强制执行;语义解释仍是模型判断。自治目标 Round 可以为准确当前目标 Round 报告 `complete` 或 `blocked`,但不能编辑、暂停、恢复或替换人类目标。
-TUI 与 ACP 默认挂载共享命令注册表和完整目标栈,并通过同一个生产方暴露 `/goal`。每条有效已注册命令都能被每个已组合的命令适配器发现和调用;若插件与某应用不兼容,该应用组合会省略其命令生产方,而不是依赖注册表层面的表面掩码。无 UI agent spine 要求显式选择加入,以免单次调用方静默变成多 Round 操作。无头 CLI 与 JSON-RPC 前端不消费命令平面;挂载目标栈后,普通人类文本仍可授权模型目标工具。
+TUI 默认挂载共享命令注册表和完整目标栈,并通过一个生产方暴露 `/goal`。ACP 挂载目标领域、模型工具和同会话驱动器,但有意省略人类命令平面。每条有效已注册命令都能被每个已组合的命令适配器发现和调用;若插件与某应用不兼容,该应用组合会省略其命令生产方,而不是依赖注册表层面的表面掩码。无 UI agent spine 要求显式选择加入,以免单次调用方静默变成多 Round 操作。无头 CLI 与 JSON-RPC 前端不消费命令平面;挂载目标栈后,普通人类文本仍可授权模型目标工具。
### 全新 agent Ralph 执行
@@ -94,7 +94,7 @@ Codex 提供了这里采用的最小可观察目标 UX:一个附着于聊天
### 验证
-六份所属 Agent Note 记录了单元、集成、进程、快照、取消、重放与构建后运行时覆盖。该栈验证严格目标记录折叠、比较并交换竞争、会话 fork 继承、恢复后未激活、自然语言直接人类权限、可配置上限与阻塞阈值、准确目标回合归属、适配器范围的命令发现与转录隔离。已发布的无密钥快照覆盖通过无头应用创建/检查模型目标、通过 ACP 执行多 Round 同会话生命周期与取消、无需模型 Turn 的直接 `/goal` 状态,以及通过无头应用执行两个真实 Ralph Round。Ralph 快照会启动工作线程引擎、spawn provider、结构化输出运行时与 agent loop,随后检查互不相同且无种子的子日志和准确单向有界交接,同时固定父级事件流。聚焦的真实栈测试还覆盖完成、阻塞与 Round 上限结果、畸形及过大报告、保留上一份有效交接的普通子 agent 失败、单个阶段事件,以及取消后达到子 agent 静止状态。包源码继续受仓库逐文件 100% 覆盖率门禁约束,构建后二进制测试覆盖已安装产物解析。实现经验已记录进根测试策略:每项非平凡的模型或人类可见变更都必须在同一 PR 中携带真实示例无密钥快照,而不能依赖仅包级或仅模拟夹具的覆盖。
+六份所属 Agent Note 记录了单元、集成、进程、快照、取消、重放与构建后运行时覆盖。该栈验证严格目标记录折叠、比较并交换竞争、会话 fork 继承、恢复后未激活、自然语言直接人类权限、可配置上限与阻塞阈值、准确目标回合归属、适配器范围的命令发现与转录隔离。已发布的无密钥快照覆盖通过无头应用创建/检查模型目标、通过 ACP 执行多 Round 同会话生命周期与取消,以及通过无头应用执行两个真实 Ralph Round;聚焦的命令测试固定了无需模型 Turn 的直接 `/goal` 状态。Ralph 快照会启动工作线程引擎、spawn provider、结构化输出运行时与 agent loop,随后检查互不相同且无种子的子日志和准确单向有界交接,同时固定父级事件流。聚焦的真实栈测试还覆盖完成、阻塞与 Round 上限结果、畸形及过大报告、保留上一份有效交接的普通子 agent 失败、单个阶段事件,以及取消后达到子 agent 静止状态。包源码继续受仓库逐文件 100% 覆盖率门禁约束,构建后二进制测试覆盖已安装产物解析。实现经验已记录进根测试策略:每项非平凡的模型或人类可见变更都必须在同一 PR 中携带真实示例无密钥快照,而不能依赖仅包级或仅模拟夹具的覆盖。
## 考虑过的替代方案
@@ -126,4 +126,4 @@ Codex 提供了这里采用的最小可观察目标 UX:一个附着于聊天
- **没有目标反思器**——concern 事件、自动无进展启发式、由独立反思器执行的目标修订、卡住模式检测与 `loop_split` 均未实现。人类可以直接编辑、暂停、清除或恢复目标。
- **Ralph 策略仍然狭窄**——一个 Round 创建一个全新子 agent;Round 内扇出、评估器/工作者角色分离、动态 provider/模型选择与结构化递归 Ralph 工具禁止都需要独立策略表面。提示词指导不是强制执行。
- **Ralph 不会重试失败的子 agent**——普通失败会保留失败 Round 与上一份有效交接,而致命工作流基础设施错误可能在该状态可用前结束。重试次数、退避与更丰富的失败传输需要独立的策略与接缝设计。
-- **可移植 UI 仍较朴素**——TUI 与 ACP 渲染纯文本目标状态和通用 Ralph 卡片。系统没有持续状态组件、可重连命令输出、模态目标编辑器,无头 CLI 与 JSON-RPC 前端也没有命令平面。
+- **可移植 UI 仍较朴素**——TUI 渲染纯文本目标状态和通用 Ralph 卡片。ACP 只承载已提交的助手文本;系统没有持续状态组件、可重连命令输出、模态目标编辑器,ACP、无头 CLI 与 JSON-RPC 也没有命令平面。
diff --git a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.i18n.yaml b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.i18n.yaml
index be66b14fbf..f7e242b78c 100644
--- a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.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
-2026-07-16-persistent-pty-sessions.md: b33993d36753d3195ec52d3b38dda62746a47bf3
-2026-07-16-persistent-pty-sessions.zh.md: 6ed330d75824a4e6fca9de0d82db61a7c6543322
+2026-07-16-persistent-pty-sessions.md: 148d4a2f47689e38a3ec83a7a41e4f75c4b73d95
+2026-07-16-persistent-pty-sessions.zh.md: 9a9d9cd4b0f61e8abaf011996ecd8739d13851f8
diff --git a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md
index b33993d367..148d4a2f47 100644
--- a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md
+++ b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md
@@ -24,7 +24,7 @@ The implementation supports interactive shells and line-oriented REPLs on Linux
|---|---|---|
| `dsh-pty` | `PtyService`, branded `PtySessionId`, backend registry, owner-scoped session contract, and result types | `ctx.pty` |
| `dsh-pty-local` | [`node-pty`](https://github.com/microsoft/node-pty)-based local backend, platform process inspection, bounded terminal buffer, sandbox resolution, and process-tree supervision | registers a backend on `ctx.pty` |
-| `dsh-tool-pty` | Six model-facing tools, task-runtime integration for background sends, guidance, and ACP render intents | registers on `ctx.tools` |
+| `dsh-tool-pty` | Six model-facing tools, task-runtime integration for background sends, guidance, and UI render intents | registers on `ctx.tools` |
Idle detection is backend behavior, not a second public seam. A remote or container backend may have authoritative readiness signals that do not resemble local `/proc` inspection; every `PtyBackend` therefore returns the common send result while owning its detection mechanism internally.
@@ -58,7 +58,7 @@ The implementation uses only public `node-pty` capabilities: child PID, `data` a
| `terminal_close` | Close one session and await process-tree quiescence | `{ killed }` |
| `terminal_list` | List the caller's live sessions | owner-scoped session summaries |
-The ACP render contract is exact and location-free. `terminal_send` uses terminal call/result cards only for foreground sends; its background form is generic `execute`. `terminal_open`, `terminal_read`, `terminal_signal`, `terminal_close`, and `terminal_list` use generic `execute`, `read`, `execute`, `delete`, and `read` cards respectively. No PTY tool emits `locations`.
+The UI render contract is exact and location-free. `terminal_send` uses terminal call/result cards only for foreground sends; its background form is generic `execute`. `terminal_open`, `terminal_read`, `terminal_signal`, `terminal_close`, and `terminal_list` use generic `execute`, `read`, `execute`, `delete`, and `read` cards respectively. No PTY tool emits `locations`.
`terminal_send({ sessionId, text, submit?, run_in_background? })` treats `text` as UTF-8 bytes and resolves `submit` to `true` in the tool implementation. When `submit` is true it writes the platform Enter sequence after the text; when false it writes only the text, allowing control characters and REPL fragments without hidden content heuristics. `enableRunInBackground` defaults to true; false removes `run_in_background` from the schema and rejects the same undeclared argument if a caller forces it through execution.
@@ -155,7 +155,7 @@ The package ships concise tool guidance explaining persistent state, owner isola
- Per-file coverage pins owner fencing, concurrent reservations, unpublished-spawn cancellation and awaited teardown, sandbox-mode change rejection, retriable lifecycle cleanup, readiness tiers, sanitizer carry state, complete UTF-8 bounds, task integration, schemas, and exact render intents.
- Linux process fixtures cover non-leader and non-main-thread stdin waits, zombie quiescence, unreadable process state, supported syscall tables, unsupported architectures, and false-positive rejection; macOS inspector logic is injected into the same unit suite.
- Real `node-pty` tests exercise shell state, shared sandbox policy, environment scrubbing, raw-mode foreground `SIGINT`, a TERM-ignoring descendant, and immediate post-disposal quiescence on supported hosts.
-- A Loader-driven `cordis.yml` test mounts the real three-package composition, while ACP and headless snapshots pin the six schemas, bounded results, error rendering, and terminal/generic cards through opt-in overlays.
+- A Loader-driven `cordis.yml` test mounts the real three-package composition. ACP and headless snapshots pin the six schemas, bounded results, and errors through opt-in overlays; TUI snapshots pin terminal and generic card presentation.
- Package contracts, the architecture map, core data structures, generated catalogs, and the website API describe the same shipped surface.
- The repository CI-equivalent sequence owns type, lint, coverage, snapshot, documentation, build, hygiene, demo, and built-entry verification.
diff --git a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.zh.md b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.zh.md
index 6ed330d758..9a9d9cd4b0 100644
--- a/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.zh.md
@@ -24,7 +24,7 @@ harness 可以运行前台与后台命令、编辑文件和委派工作,但无
|---|---|---|
| `dsh-pty` | `PtyService`、branded `PtySessionId`、后端注册表、按 owner 隔离的会话契约和结果类型 | `ctx.pty` |
| `dsh-pty-local` | 基于 [`node-pty`](https://github.com/microsoft/node-pty) 的本地后端、平台进程检查、有界终端缓冲、沙箱解析和进程树监管 | 在 `ctx.pty` 上注册后端 |
-| `dsh-tool-pty` | 6 个面向模型的工具、后台发送的 task 运行时集成、使用指引和 ACP render intent | 注册到 `ctx.tools` |
+| `dsh-tool-pty` | 6 个面向模型的工具、后台发送的 task 运行时集成、使用指引和 UI 渲染意图 | 注册到 `ctx.tools` |
idle 检测属于后端行为,不是第二条公共 seam。远程或容器后端可能拥有完全不同于本地 `/proc` 检查的权威就绪信号;因此每个 `PtyBackend` 都返回统一的发送结果,同时在内部拥有自己的检测机制。
@@ -58,7 +58,7 @@ agent scope dispose 时先关闭注册,再等待全部所属 PTY 静默退出
| `terminal_close` | 关闭一个会话并等待进程树静默退出 | `{ killed }` |
| `terminal_list` | 列出调用方的活会话 | 按 owner 隔离的会话摘要 |
-ACP 渲染契约精确且不携带位置信息。`terminal_send` 只为前台发送使用 terminal 调用卡片和结果卡片;后台形式使用通用 `execute` 卡片。`terminal_open`、`terminal_read`、`terminal_signal`、`terminal_close` 和 `terminal_list` 分别使用通用 `execute`、`read`、`execute`、`delete` 和 `read` 卡片。所有 PTY 工具都不发出 `locations`。
+UI 渲染契约精确且不携带位置信息。`terminal_send` 只为前台发送使用 terminal 调用卡片和结果卡片;后台形式使用通用 `execute` 卡片。`terminal_open`、`terminal_read`、`terminal_signal`、`terminal_close` 和 `terminal_list` 分别使用通用 `execute`、`read`、`execute`、`delete` 和 `read` 卡片。所有 PTY 工具都不发出 `locations`。
`terminal_send({ sessionId, text, submit?, run_in_background? })` 将 `text` 视为 UTF-8 字节,并由工具实现在解析阶段把 `submit` 默认成 `true`。`submit` 为 true 时先写入文本,再写入平台 Enter 序列;为 false 时只写文本,使控制字符和 REPL 片段无需隐藏的内容启发式即可发送。`enableRunInBackground` 默认为 true;设为 false 时,schema 中会移除 `run_in_background`,调用方即使强行把这个未声明参数传入执行流程,也会被拒绝。
@@ -122,7 +122,7 @@ plugins:
maxResultBytes: 262144
```
-包提供简洁的工具指引,说明持久状态、owner 隔离、不确定的 idle 结果、清理,以及无需交互时优先使用现有一次性工具。已发布的基础示例不挂载 PTY:PTY 仅通过专用组合 opt-in,ACP 与 headless 快照 overlay 覆盖该组合。`dsh-tool-pty` 实例一旦启用,6 个工具和 `run_in_background` 就会默认启用;部署可通过配置仅禁用后台参数。
+包提供简洁的工具指引,说明持久状态、owner 隔离、不确定的 idle 结果、清理,以及无需交互时优先使用现有一次性工具。已发布的基础示例不挂载 PTY:PTY 仅通过专用组合 opt-in,ACP(Agent Client Protocol)与 headless 快照 overlay 覆盖该组合。`dsh-tool-pty` 实例一旦启用,6 个工具和 `run_in_background` 就会默认启用;部署可通过配置仅禁用后台参数。
### 推迟的工作
@@ -155,7 +155,7 @@ plugins:
- 每文件覆盖率固定 owner 隔离、并发预留、未发布 spawn 的取消与等待式 teardown、沙箱模式变更拒绝、可重试的生命周期清理、就绪层级、sanitizer carry state、完整 UTF-8 结果上限、task 集成、schema 和精确 render intent。
- Linux 进程 fixture 覆盖非 leader 与非主线程的 stdin 等待、僵尸进程静止性、不可读进程状态、受支持的 syscall 表、不支持的架构和误报拒绝;同一单元测试套件通过注入覆盖 macOS 检查器逻辑。
- 真实 `node-pty` 测试在受支持宿主上覆盖 shell 状态、共享沙箱策略、环境清洗、raw mode 下的前台 `SIGINT`、忽略 `SIGTERM` 的子进程,以及 dispose 返回后立即静默。
-- Loader 驱动的 `cordis.yml` 测试挂载真实三包组合;ACP 与 headless 快照通过 opt-in overlay 固定 6 个 schema、有界结果、错误渲染和 terminal/generic card。
+- Loader 驱动的 `cordis.yml` 测试挂载真实三包组合。ACP 与 headless 快照通过 opt-in overlay 固定 6 个 schema、有界结果和错误;TUI 快照固定 terminal 与 generic 卡片展示。
- 包契约、架构图、核心数据结构、生成目录和 website API 描述同一个已发布接口。
- 仓库 CI 等价序列负责类型、lint、覆盖率、快照、文档、构建、hygiene、demo 和 built-entry 验证。
diff --git a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.i18n.yaml b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.i18n.yaml
index 8d6c7be831..79a4f5e7cc 100644
--- a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.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
-2026-07-17-dedicated-full-screen-tui-front-door.md: ecfda138593fc2b98ac42929acc586b11e437ee2
-2026-07-17-dedicated-full-screen-tui-front-door.zh.md: 6b8cc63f7657672a6da542e2033d765b54bd4f07
+2026-07-17-dedicated-full-screen-tui-front-door.md: aac67ffec89606d04d5abfd233d0469e2241b102
+2026-07-17-dedicated-full-screen-tui-front-door.zh.md: 0f64e2ce14b18d315f75913b82c731e77758e377
diff --git a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.md b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.md
index ecfda13859..aac67ffec8 100644
--- a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.md
+++ b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.md
@@ -14,7 +14,7 @@ The interactive channel must remain a Cordis plugin over the same agent, session
DeepSeek Harness ships [`@deepseek-ai/dsh-tui`](../../../../packages/ui/tui/README.md) as a dedicated Cordis plugin. It owns terminal input and presentation only; agent lifecycle, session persistence, tool execution, and the model-facing question tool remain separate composition entries. The plugin requires both stdin and stdout to be TTYs and fails instead of silently changing to line-oriented behavior.
-The app layer has one terminal front door. `@deepseek-ai/dsh-tui-demo` mounts the TUI before the configured agent, and `examples/tui-agent` owns the interactive coding composition and Code Mode overlay directly. Non-interactive tasks use `@deepseek-ai/dsh-cli-demo`; ACP remains a separate editor protocol.
+The app layer has one terminal front door. `@deepseek-ai/dsh-tui-demo` mounts the TUI before the configured agent, and `examples/tui-agent` owns the interactive coding composition and Code Mode overlay directly. Non-interactive tasks use `@deepseek-ai/dsh-cli-demo`; ACP remains a separate automation protocol.
The selected front door receives the exact generated or resumed `SessionId` used by the pre-created agent. It mounts before the agent composition, waits for the matching root agent, and enters full-screen mode only after that agent exists. A matching `agent-loop/config-start-failed` event is therefore reported before screen takeover and exits with status 1.
diff --git a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.zh.md b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.zh.md
index 6b8cc63f76..0f64e2ce14 100644
--- a/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-17-dedicated-full-screen-tui-front-door.zh.md
@@ -14,7 +14,7 @@ Status: implemented
DeepSeek Harness 将 [`@deepseek-ai/dsh-tui`](../../../../packages/ui/tui/README.md) 作为独立的 Cordis 插件交付。该插件只负责终端输入与呈现;agent 生命周期、会话持久化、工具执行以及模型可见的提问工具仍由不同组合项负责。插件要求 stdin 和 stdout 均为 TTY;条件不满足时会失败,不会静默切换为逐行输出。
-应用组合层只有一个终端入口。`@deepseek-ai/dsh-tui-demo` 在已配置 agent 之前挂载 TUI,`examples/tui-agent` 直接拥有交互式 coding 组装及其 Code Mode overlay。非交互任务使用 `@deepseek-ai/dsh-cli-demo`;ACP 仍是独立的编辑器协议。
+应用组合层只有一个终端入口。`@deepseek-ai/dsh-tui-demo` 在已配置 agent 之前挂载 TUI,`examples/tui-agent` 直接拥有交互式 coding 组装及其 Code Mode overlay。非交互任务使用 `@deepseek-ai/dsh-cli-demo`;ACP 仍是独立的自动化协议。
所选入口接收预创建 agent 使用的同一个新建或恢复 `SessionId`。入口先于 agent 组合挂载,等待相符的根 agent 出现,然后才进入全屏模式。因此,相符的 `agent-loop/config-start-failed` 事件会在接管屏幕前报告,并以状态码 1 退出。
diff --git a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.i18n.yaml b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.i18n.yaml
index 17e3cdf5e2..67b6b8da3d 100644
--- a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.i18n.yaml
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write
-2026-07-19-fresh-agent-ralph-workflow-tool.md: c2db4d7dd30c27a25adecdfc425db261cc3dfeb5
-2026-07-19-fresh-agent-ralph-workflow-tool.zh.md: e33e9848d71c98c8f83494ebe8bf171ef10b9305
+2026-07-19-fresh-agent-ralph-workflow-tool.md: 6fe96587c49ef0316d1618fda2ee26b015b1ce87
+2026-07-19-fresh-agent-ralph-workflow-tool.zh.md: c615c16ac020a0c905f6c8b52d8dc487fbcfe8be
diff --git a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.md b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.md
index c2db4d7dd3..6fe96587c4 100644
--- a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.md
+++ b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.md
@@ -40,7 +40,7 @@ The workflow language maps a normally settled but unsuccessful child to `null`.
The model may supply only `objective` and optional `maxRounds`; provider selection, report schema, handoff cap, and script are deployment-owned. A fixed prompt section says to use `ralph` only when the direct human explicitly asks for Ralph or fresh-agent iteration, and distinguishes it from same-session goals, bounded delegation, and general fan-out workflows. This is guidance rather than a new goal UX state machine.
-ACP and terminal presentation use a generic `ralph` card whose raw input is the objective. Successful completion and blocker envelopes say that a worker reported the outcome rather than presenting it as independent certification. The parent transcript retains the original tool call and one bounded successful terminal report or an error, not intermediate child messages. Shipped headless, TUI, and ACP compositions load the plugin beside the existing workflow engine; JSON-RPC remains unchanged because its default composition does not expose workflows.
+Human-facing presentation uses a generic `ralph` card whose raw input is the objective; ACP carries only the committed assistant text. Successful completion and blocker envelopes say that a worker reported the outcome rather than presenting it as independent certification. The parent transcript retains the original tool call and one bounded successful terminal report or an error, not intermediate child messages. Shipped headless, TUI, and ACP compositions load the plugin beside the existing workflow engine; JSON-RPC remains unchanged because its default composition does not expose workflows.
## Testing
diff --git a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.zh.md b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.zh.md
index e33e9848d7..c615c16ac0 100644
--- a/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-19-fresh-agent-ralph-workflow-tool.zh.md
@@ -40,7 +40,7 @@ Ralph 插件的 `subagentProvider` 默认为 `spawn`。每次调用前,它要
模型只能提供 `objective` 和可选的 `maxRounds`;provider 选择、报告 schema、交接上限和脚本都由部署拥有。固定提示区段说明,只有直接人类明确要求 Ralph 或全新 agent 迭代时才使用 `ralph`,并将其与同会话目标、有界委派和通用扇出工作流区分开。这是指导,而不是新的目标 UX 状态机。
-ACP 和终端展示使用通用 `ralph` 卡片,并把目标作为原始输入。成功完成与阻塞的外层文本会说明结果由工作者报告,而不会把它呈现为独立认证。父转录只保留原始工具调用,以及一份有界成功终止报告或一个错误,不包含中间子 agent 消息。发布的无头、TUI 与 ACP 组合会在现有工作流引擎旁加载该插件;JSON-RPC 保持不变,因为其默认组合不暴露工作流。
+面向人类的展示使用通用 `ralph` 卡片,并把目标作为原始输入;ACP 只承载已提交的助手文本。成功完成与阻塞的外层文本会说明结果由工作者报告,而不会把它呈现为独立认证。父转录只保留原始工具调用,以及一份有界成功终止报告或一个错误,不包含中间子 agent 消息。发布的无头、TUI 与 ACP 组合会在现有工作流引擎旁加载该插件;JSON-RPC 保持不变,因为其默认组合不暴露工作流。
## 测试
diff --git a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.i18n.yaml b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.i18n.yaml
index 5379f25772..44e1412cc4 100644
--- a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.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
-2026-07-19-human-goal-command.md: a272206a3bfad50a01ce871c56c7e7bcf924684e
-2026-07-19-human-goal-command.zh.md: 370c9bc24510320c70e3d789926c492e543968b1
+2026-07-19-human-goal-command.md: ce5c37fd28f9432d8c9a8797cac32c632617e317
+2026-07-19-human-goal-command.zh.md: d5ce36bd75a6f1070ee1eaeb1ac6ee97778c246b
diff --git a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.md b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.md
index a272206a3b..ce5c37fd28 100644
--- a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.md
+++ b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.md
@@ -6,7 +6,7 @@ English | [中文](2026-07-19-human-goal-command.zh.md)
## Problem
-The same-session goal domain and model tools provide the state machine and semantic natural-language path, but they are not a sufficient human UX. A user needs to inspect the exact current phase and round budget without asking the model, explicitly pause or clear work without spending a model turn, and rearm a restored active goal after the required post-resume human decision. Implementing those actions independently in TUI and ACP would duplicate parsing, let the surfaces drift, and risk routing an unknown or unavailable command into the model.
+The same-session goal domain and model tools provide the state machine and semantic natural-language path, but they are not a sufficient human UX. A user needs to inspect the exact current phase and round budget without asking the model, explicitly pause or clear work without spending a model turn, and rearm a restored active goal after the required post-resume human decision. Implementing those actions independently in each UI would duplicate parsing, let the surfaces drift, and risk routing an unknown or unavailable command into the model.
The command must also respect the goal design's two kinds of state. Durable phase, objective, revisions, and rounds come from the session log; process-local activation decides whether an active goal may continue automatically. Showing only “active” after a resume would be misleading when the restored goal is intentionally disarmed and waiting for human authorization.
@@ -22,7 +22,7 @@ The command follows the compact Codex shape in the [public OpenAI Codex TUI disp
`/goal ` creates an active armed goal. A completed goal may be replaced, which creates a fresh goal identity through the existing domain rule. Any unfinished goal makes the command fail directly with instructions to use inline edit or explicit clear. The generic command service deliberately has no modal confirmation API, so silently clearing and creating two durable records would manufacture destructive consent and expose a non-atomic failure window.
-`/goal edit ` edits the current non-complete goal without changing phase or activation. On a completed goal it creates a fresh active goal because the domain does not permit completed state to resume and a new completion objective is a new goal identity. Bare `edit` is an error rather than an editor launch because ACP's shared unstructured command contract has no portable modal editor.
+`/goal edit ` edits the current non-complete goal without changing phase or activation. On a completed goal it creates a fresh active goal because the domain does not permit completed state to resume and a new completion objective is a new goal identity. Bare `edit` is an error rather than an editor launch because the portable unstructured command contract has no modal editor.
`/goal pause`, `/goal resume`, and `/goal clear` call the matching compare-and-set domain verbs against the current view. Resume covers both stopped durable phases and an active-but-disarmed goal after session resume, fork, or driver replacement. Domain rules still reject exhausted round caps, redundant active/armed resume, invalid phase transitions, and stale identity. Clear removes the current pointer while the session log retains the revisioned tombstone and earlier snapshots.
@@ -40,16 +40,16 @@ Generic slash input, status text, and errors are not persisted. Successful goal
`agent-spine-demo` accepts an optional `goals` composition object containing the goal-domain and model-tool owner configs. Omission or `false` leaves the stack unmounted. This explicit opt-in is important for headless one-shot callers: their result API settles one correlated physical turn and must not silently become a long-running logical goal operation.
-The interactive app bundles make the opposite product choice. ACP and TUI default `goals` to the owner defaults and mount the goal domain, model tools, same-session driver, command registry, and this producer. Both apps accept `goals: false` as one coherent stack opt-out. The Python SDK runtime closure ships this producer alongside ACP, commands, and the goal stack so an external `cordis.yml` can compose the same command.
+The TUI app bundle makes the opposite product choice. It defaults `goals` to the owner defaults and mounts the goal domain, model tools, same-session driver, command registry, and this producer; `goals: false` removes the stack coherently. The [ACP automation app](../simplification/2026-07-23-acp-automation-only-protocol.md) also defaults the goal domain and model tools but deliberately omits command services. The Python SDK runtime closure ships this producer, commands, and the goal stack so an external `cordis.yml` can compose the same command.
## Testing
-The producer suite uses the real command registry, goal service, agent registry, and session log. It covers Loader-safe exports, registry discovery, disposal, empty status, objective parsing, unfinished replacement refusal, inline edit, completed replacement, all missing-state controls, pause/resume/clear, every durable phase, blocked code/explanation presentation, armed/disarmed presentation, sanitized domain errors, unexpected failures, and persisted mutation records. App composition tests cover explicit spine opt-in, TUI/ACP defaults, coherent opt-out, forwarded domain/tool config, command discovery, the packaged-runtime closure, and the expanded model-tool assembly. A keyless snapshot boots the shipped ACP application, observes its advertised `/goal` metadata, invokes `/goal` directly, and pins the no-model-turn result; the surrounding ACP snapshots also pin the goal tool schemas in that composition.
+The producer suite uses the real command registry, goal service, agent registry, and session log. It covers Loader-safe exports, registry discovery, disposal, empty status, objective parsing, unfinished replacement refusal, inline edit, completed replacement, all missing-state controls, pause/resume/clear, every durable phase, blocked code/explanation presentation, armed/disarmed presentation, sanitized domain errors, unexpected failures, and persisted mutation records. App composition tests cover explicit spine opt-in, TUI defaults, coherent opt-out, forwarded domain/tool config, command discovery, the packaged-runtime closure, and the expanded model-tool assembly. ACP backend snapshots continue to pin the goal tool schemas independently of this human command.
## Alternatives considered
-- **Let the model handle `/goal` as ordinary text** — rejected because status and direct lifecycle actions would cost a model turn, could be reinterpreted, and would not provide deterministic ACP discovery.
-- **Implement separate TUI and ACP handlers** — rejected because grammar, error behavior, and goal-state formatting would drift and optional deployments could not add or remove the capability as one effect.
+- **Let the model handle `/goal` as ordinary text** — rejected because status and direct lifecycle actions would cost a model turn, could be reinterpreted, and would not provide deterministic command discovery.
+- **Implement separate handlers in each UI** — rejected because grammar, error behavior, and goal-state formatting would drift and optional deployments could not add or remove the capability as one effect.
- **Add modal editing and replacement confirmation to `ctx.commands`** — rejected because the existing cross-surface contract is unstructured input plus direct output; a general interaction protocol needs more than this one producer.
- **Silently replace an unfinished goal** — rejected because it combines clear and create without atomicity or explicit destructive intent.
- **Expose goal id and revision in human status** — rejected because human actions always target the exact current view inside one synchronous handler; those fields add implementation noise without preventing another race.
@@ -57,7 +57,7 @@ The producer suite uses the real command registry, goal service, agent registry,
## Consequences
-- TUI and ACP expose one Codex-shaped `/goal` command supplied by a removable plugin.
+- TUI exposes one Codex-shaped `/goal` command supplied by a removable plugin.
- Human status distinguishes durable phase from live activation and reports the exact goal-round cap.
- Direct pause, resume, clear, creation, and edit consume no model turn while their accepted mutations remain reconstructable from the session log.
- Restored sessions wait for a human decision; `/goal resume` is the literal command path, while an ordinary prompt in any language may authorize the model tool path.
@@ -67,6 +67,6 @@ The producer suite uses the real command registry, goal service, agent registry,
- The portable command contract has no modal editor or confirmation interaction; inline edit and explicit clear are intentional until a general cross-surface interaction primitive exists.
- `/goal` does not accept a per-command round cap. Deployment config owns the default, and the authorized model tool can edit a cap after direct human instruction.
-- TUI and ACP render portable plain text rather than a continuously updated goal status widget. Reconnectable command output and adapter-specific status indicators are deferred.
-- The headless CLI and JSON-RPC front doors do not consume the command registry.
+- TUI renders portable plain text rather than a continuously updated goal status widget. Reconnectable command output and adapter-specific status indicators are deferred.
+- The ACP automation server, headless CLI, and JSON-RPC front doors do not consume the command registry.
- The command observes and mutates state but does not certify completion or blockers. Evaluator-backed certification remains deferred to a separate policy layer with an explicit authority and isolation contract.
diff --git a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.zh.md b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.zh.md
index 370c9bc245..d5ce36bd75 100644
--- a/.agents/notes/implemented/feature/2026-07-19-human-goal-command.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-19-human-goal-command.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-同会话目标领域和模型工具提供了状态机与自然语言语义路径,但尚不足以构成面向人类的 UX。用户需要在不询问模型的情况下检查准确的当前阶段与回合预算,在不消耗模型轮次的情况下明确暂停或清除工作,并在会话恢复后经过必要的人类决策重新激活已恢复的活跃目标。若在 TUI 与 ACP 中分别实现这些操作,就会重复解析逻辑、导致两个表面发生偏差,还可能把未知或不可用的命令交给模型处理。
+同会话目标领域和模型工具提供了状态机与自然语言语义路径,但尚不足以构成面向人类的 UX。用户需要在不询问模型的情况下检查准确的当前阶段与回合预算,在不消耗模型轮次的情况下明确暂停或清除工作,并在会话恢复后经过必要的人类决策重新激活已恢复的活跃目标。若在各 UI 中分别实现这些操作,就会重复解析逻辑、导致各界面发生偏差,还可能把未知或不可用的命令交给模型处理。
该命令还必须遵守目标设计中的两类状态。持久阶段、目标描述、修订号与回合来自会话日志;进程本地激活态决定活跃目标能否自动继续。恢复后若只显示“活跃”,就会掩盖目标已被有意设为未激活、正在等待人类授权这一事实。
@@ -22,7 +22,7 @@ Status: implemented
`/goal ` 创建活跃且已激活的目标。已完成目标可以被替换,此时通过现有领域规则创建新的目标身份。任何未完成目标都会让命令直接失败,并提示用户使用行内编辑或明确清除。通用命令服务有意不提供模态确认 API;若静默执行清除再创建两条持久记录,就等于凭空制造破坏性同意,并暴露一个非原子的失败窗口。
-`/goal edit ` 编辑当前未完成目标,但不改变其阶段或激活态。若目标已经完成,则创建一个新的活跃目标,因为领域不允许恢复已完成状态,而新的完成条件应拥有新的目标身份。单独使用 `edit` 会返回错误而不是启动编辑器,因为 ACP 共享的非结构化命令契约没有可移植的模态编辑器。
+`/goal edit ` 编辑当前未完成目标,但不改变其阶段或激活态。若目标已经完成,则创建一个新的活跃目标,因为领域不允许恢复已完成状态,而新的完成条件应拥有新的目标身份。单独使用 `edit` 会返回错误而不是启动编辑器,因为可移植的非结构化命令契约没有模态编辑器。
`/goal pause`、`/goal resume` 与 `/goal clear` 使用当前视图调用相应的比较并交换领域动词。恢复既适用于停止的持久阶段,也适用于会话恢复、fork 或驱动器替换后处于活跃但未激活状态的目标。领域规则仍会拒绝已耗尽的回合上限、对已活跃且已激活目标的重复恢复、非法阶段转换与陈旧身份。清除会移除当前指针,而会话日志保留带修订号的墓碑和此前快照。
@@ -40,16 +40,16 @@ Status: implemented
`agent-spine-demo` 接受可选的 `goals` 组合对象,其中包含目标领域与模型工具的所有者配置。省略或设为 `false` 时不会挂载该栈。对无头单次调用方而言,明确选择加入非常重要:它们的结果 API 会在一个相关物理轮次后结束,不能静默变成长时间运行的逻辑目标操作。
-交互式应用包作出相反的产品选择。ACP 与 TUI 默认让 `goals` 使用所有者默认值,并挂载目标领域、模型工具、同会话驱动器、命令注册表与本生产方。两个应用都接受 `goals: false` 作为一致的整体退出选项。Python SDK 运行时闭包把本生产方与 ACP、命令及目标栈一并交付,使外部 `cordis.yml` 能组合相同命令。
+TUI 应用包作出相反的产品选择。它默认让 `goals` 使用所有者默认值,并挂载目标领域、模型工具、同会话驱动器、命令注册表与本生产方;`goals: false` 会一致地移除整个栈。[ACP(Agent Client Protocol)自动化应用](../simplification/2026-07-23-acp-automation-only-protocol.md)也默认挂载目标领域与模型工具,但有意省略命令服务。Python SDK 运行时闭包交付本生产方、命令与目标栈,使外部 `cordis.yml` 能组合相同命令。
## 测试
-生产方测试套件使用真实命令注册表、目标服务、agent 注册表与会话日志。它覆盖 Loader 安全导出、注册表发现、资源释放、空状态、目标描述解析、拒绝未完成目标替换、行内编辑、已完成目标替换、所有缺失状态控制、暂停/恢复/清除、每个持久阶段、阻塞代码/说明展示、已激活/未激活展示、经净化的领域错误、意外失败与持久变更记录。应用组合测试覆盖显式主干选择加入、TUI/ACP 默认值、一致退出、转发的领域/工具配置、命令发现、打包运行时闭包与扩展后的模型工具组装。一个无密钥快照会启动交付的 ACP 应用,观察其公布的 `/goal` 元数据,直接调用 `/goal`,并固定不经过模型轮次的结果;周边 ACP 快照还会固定该组合中的目标工具 schema。
+生产方测试套件使用真实命令注册表、目标服务、agent 注册表与会话日志。它覆盖 Loader 安全导出、注册表发现、资源释放、空状态、目标描述解析、拒绝未完成目标替换、行内编辑、已完成目标替换、所有缺失状态控制、暂停/恢复/清除、每个持久阶段、阻塞代码/说明展示、已激活/未激活展示、经净化的领域错误、意外失败与持久变更记录。应用组合测试覆盖显式主干选择加入、TUI 默认值、一致退出、转发的领域/工具配置、命令发现、打包运行时闭包与扩展后的模型工具组装。ACP 后端快照继续固定目标工具 schema,与这项面向人类的命令无关。
## 考虑过的替代方案
-- **让模型把 `/goal` 当作普通文本处理**——不予采纳,因为状态与直接生命周期操作会消耗模型轮次、可能被重新解释,也无法提供确定性的 ACP 发现。
-- **分别实现 TUI 和 ACP 处理器**——不予采纳,因为语法、错误行为与目标状态格式会发生偏差,可选部署也无法把该功能作为一个 effect 统一增删。
+- **让模型把 `/goal` 当作普通文本处理**——不予采纳,因为状态与直接生命周期操作会消耗模型轮次、可能被重新解释,也无法提供确定性的命令发现。
+- **在各 UI 中分别实现处理器**——不予采纳,因为语法、错误行为与目标状态格式会发生偏差,可选部署也无法把该功能作为一个 effect 统一增删。
- **为 `ctx.commands` 添加模态编辑与替换确认**——不予采纳,因为现有跨表面契约是非结构化输入加直接输出;通用交互协议所需的设计远超这一个生产方。
- **静默替换未完成目标**——不予采纳,因为这会在没有原子性或明确破坏性意图的情况下组合清除与创建。
- **在人类状态中暴露目标 id 与修订号**——不予采纳,因为人类操作始终在一个同步处理器内针对准确当前视图;这些字段只会增加实现噪声,无法消除其他竞争。
@@ -57,7 +57,7 @@ Status: implemented
## 后果
-- TUI 与 ACP 暴露由可移除插件提供的同一个 Codex 形态 `/goal` 命令。
+- TUI 暴露由可移除插件提供的 Codex 形态 `/goal` 命令。
- 人类状态会区分持久阶段与实时激活态,并报告准确的目标回合上限。
- 直接暂停、恢复、清除、创建与编辑不消耗模型轮次,而其已接受变更仍可从会话日志重建。
- 恢复后的会话等待人类决策;`/goal resume` 是字面命令路径,任何语言的普通提示词则可以授权模型工具路径。
@@ -67,6 +67,6 @@ Status: implemented
- 可移植命令契约没有模态编辑器或确认交互;在出现通用跨表面交互原语之前,行内编辑与明确清除是有意选择。
- `/goal` 不接受逐命令回合上限。部署配置拥有默认值;得到直接人类指示后,已授权模型工具可以编辑上限。
-- TUI 与 ACP 渲染可移植纯文本,而不是持续更新的目标状态组件。可重连命令输出和适配器专用状态指示器予以延期。
-- 无头 CLI 与 JSON-RPC 前端不消费命令注册表。
+- TUI 渲染可移植纯文本,而不是持续更新的目标状态组件。可重连命令输出和适配器专用状态指示器予以延期。
+- ACP 自动化服务器、无头 CLI 与 JSON-RPC 前端不消费命令注册表。
- 该命令观察并改变状态,但不认证完成或阻塞。基于评估器的认证延期到具有明确权限与隔离契约的独立策略层。
diff --git a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.i18n.yaml b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.i18n.yaml
index c8cf32d77a..9beb453252 100644
--- a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.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
-2026-07-19-model-facing-goal-tools.md: 795fa4040d4d6f861b1b3161a1c37146551084ad
-2026-07-19-model-facing-goal-tools.zh.md: 399a38a8ef16def4525b361437767552a961b994
+2026-07-19-model-facing-goal-tools.md: 2c53a7658e213dee4fecc93709244f97b821aca0
+2026-07-19-model-facing-goal-tools.zh.md: aa7b5ea14b7afff88819f0efa35c5f2d5e2e933e
diff --git a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.md b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.md
index 795fa4040d..bc4305af80 100644
--- a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.md
+++ b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.md
@@ -20,7 +20,7 @@ The surface also needs to preserve the separation between durable state and live
The prompt tells the model that it may infer goal intent from a direct human request in any wording or language, but should not convert routine single-turn work into a goal. It must read the current goal before updating and copy the exact id and revision. On a restored or forked active-but-disarmed goal, a semantic human request to continue is grounds for `resume`. Completion is reserved for an achieved objective, and difficulty or uncertainty alone is not a blocker; a block report must name the concrete condition.
-All three tools use exclusive execution so a model-ordered batch observes prior mutations and their new revisions. Results are compact JSON. ACP presentation is a pure function of arguments and uses generic read or mutation cards; mutation cards select meaningful action values before the goal id, so accepted fillers cannot blank their input. Activation is reported only as live observation and is never written into replay state.
+All three tools use exclusive execution so a model-ordered batch observes prior mutations and their new revisions. Results are compact JSON. UI presentation is a pure function of arguments and uses generic read or mutation cards; mutation cards select meaningful action values before the goal id, so accepted fillers cannot blank their input. Activation is reported only as live observation and is never written into replay state.
An autonomous goal round that successfully reports completion or blocking marks its tool result as concluding the physical turn, preventing an unnecessary follow-up request. Direct-human mutations do not conclude the turn: the assistant can acknowledge the change, and concurrent human steering remains available to ordinary stopping checks.
diff --git a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.zh.md b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.zh.md
index 399a38a8ef..b07f62aa52 100644
--- a/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-19-model-facing-goal-tools.zh.md
@@ -20,7 +20,7 @@ Status: implemented
提示词告诉模型:它可以从任何措辞或语言的直接人类请求中推断目标意图,但不应把常规单轮工作转换为目标。更新前必须读取当前目标,并复制准确的 id 和修订号。对于恢复或派生后处于活跃但未激活状态的目标,人类在语义上要求继续即可成为执行 `resume` 的依据。只有目标已经实现时才能标记完成,困难或不确定性本身不构成阻塞;阻塞报告必须说明具体条件。
-三个工具都采用独占执行,使模型排序的批次可以观察此前变更及其新修订号。结果为紧凑 JSON。ACP 展示是参数的纯函数,使用通用读取或变更卡片;变更卡片选择输入时,先取有实际意义的操作值,再取目标 id,因此允许的占位值不会使卡片输入留空。激活态仅作为实时观察返回,绝不会写入回放状态。
+三个工具都采用独占执行,使模型排序的批次可以观察此前变更及其新修订号。结果为紧凑 JSON。UI 展示是参数的纯函数,使用通用读取或变更卡片;变更卡片选择输入时,先取有实际意义的操作值,再取目标 id,因此允许的占位值不会使卡片输入留空。激活态仅作为实时观察返回,绝不会写入回放状态。
自主目标回合成功报告完成或阻塞后,其工具结果会被标记为结束该物理轮次,避免再发起一次不必要的模型请求。直接人类发起的变更不会结束轮次:agent 可以确认该变更,并且并发的人类 steering(中途引导)仍可参与普通的停止检查。
diff --git a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.i18n.yaml b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.i18n.yaml
index 9789d902fc..8b2ddb5ac1 100644
--- a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.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
-2026-07-19-plugin-command-registration.md: bc3d33f9abf7cd87b78aac8f7d36ac9c021a7910
-2026-07-19-plugin-command-registration.zh.md: 054ab3a90eeecc8c5ddc2ff072b53112fd8e7845
+2026-07-19-plugin-command-registration.md: 343cb5d946dba9fb881adf12c197961dfd6a359b
+2026-07-19-plugin-command-registration.zh.md: 27757f05afe04d7cbd4ceaf9380b6f73441cc4b9
diff --git a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.md b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.md
index bc3d33f9ab..343cb5d946 100644
--- a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.md
+++ b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.md
@@ -6,17 +6,17 @@ English | [中文](2026-07-19-plugin-command-registration.zh.md)
## Problem
-The TUI owns seven slash commands, while ACP defines a standard command catalog and invocation shape. Keeping command names, help text, autocomplete, dispatch, and cancellation inside each adapter makes every new command an adapter edit, prevents optional plugins from contributing commands, and lets the two front doors drift. Treating slash input as an ordinary model prompt is also unsafe: a user-visible direct action can unexpectedly consume tokens or let the model reinterpret an unknown command.
+The TUI owns slash commands. Keeping command names, help text, autocomplete, dispatch, and cancellation inside the adapter makes every new command a TUI edit and prevents optional plugins from contributing commands. Treating slash input as an ordinary model prompt is also unsafe: a user-visible direct action can unexpectedly consume tokens or let the model reinterpret an unknown command.
-A shared mechanism must remain a UI concern rather than a model tool or agent-loop branch. It also needs exact per-agent visibility, HMR-safe removal, per-session ACP discovery, direct result rendering, and request-scoped cancellation without automatically adding command text or output to model history.
+A shared mechanism must remain a UI concern rather than a model tool or agent-loop branch. It also needs exact per-agent visibility, HMR-safe removal, direct result rendering, and request-scoped cancellation without automatically adding command text or output to model history.
## Decision
-`@deepseek-ai/dsh-commands` in `packages/ui/commands/` is the product command registry. The terminal and ACP app bundles mount it beside their consuming front door, and the SDK project helper emits the same service when scaffolding ACP directly; the executor-less, UI-less agent spine remains independent. TUI and ACP inject the service, while command producers depend only on the registry and any domain they operate.
+`@deepseek-ai/dsh-commands` in `packages/ui/commands/` is the product command registry. The TUI app bundle mounts it beside its consuming front door; the [automation-only ACP app](../simplification/2026-07-23-acp-automation-only-protocol.md) and the executor-less, UI-less agent spine omit it. TUI injects the service, while command producers depend only on the registry and any domain they operate.
### Registry contract
-A `CommandDefinition` contains a lowercase name without `/`, a non-empty description, an optional unstructured-input hint, and an abortable handler. Registration validates and detaches the metadata, freezes the effective definition, and returns the exact Cordis effect disposer. Duplicate names fail within one layer. Every adapter consuming the registry sees every effective definition; a command plugin that cannot operate in a deployment omits its registration there instead of encoding adapter identities in the shared domain.
+A `CommandDefinition` contains a lowercase name without `/`, a non-empty description, an optional unstructured-input hint, and an abortable handler. Registration validates and detaches the metadata, freezes the effective definition, and returns the exact Cordis effect disposer. Duplicate names fail within one layer. Every consumer sees every effective definition; a command plugin that cannot operate in a deployment omits its registration there instead of encoding consumer identities in the shared domain.
`list(agent)` returns immutable name-sorted descriptors after scoped shadowing. `find(agent, name)` resolves the effective definition. `execute(agent, line, signal)` parses and runs a known definition, returning a detached `success` or `error` result; invalid syntax and unknown names return `undefined` so the adapter owns its direct error text.
@@ -36,46 +36,36 @@ Expected handler failures return `CommandResult.error`. Thrown or malformed resu
### TUI mapping
-The TUI registers `help`, `clear`, `cancel`, `reasoning`, `tools`, `redraw`, and `exit` as agent-scoped command definitions instead of switching on strings. Its autocomplete and help view read the live catalog, so plugin commands appear and disappear with their effects. Any submitted line beginning with `/` stays in the command plane; unknown input produces a terminal warning rather than falling through to `Agent.send()` or `Agent.steer()`.
+The TUI registers its built-in slash commands as agent-scoped command definitions instead of switching on strings. Its autocomplete and help view read the live catalog, so plugin commands appear and disappear with their effects. Any submitted line beginning with `/` stays in the command plane; unknown input produces a terminal warning rather than falling through to `Agent.send()` or `Agent.steer()`.
Each submitted command owns an `AbortController`. TUI disposal aborts outstanding dispatches, removes the local definitions, and waits for the command-producing fiber before completing teardown.
-### ACP mapping
-
-The bridge follows the current [ACP v1 slash-command contract](https://agentclientprotocol.com/protocol/v1/slash-commands). `session/new` and `session/load` emit the exact agent's full `available_commands_update` snapshot; a new session's RPC response introduces its server-generated id before the snapshot is enqueued. Every registry change emits a replacement snapshot for each live session. Names, descriptions, and optional unstructured-input hints map directly to `AvailableCommand`.
-
-ACP permits a command prompt to contain additional supported content blocks. The bridge applies its ordinary lossless `text` and `resource_link` flattening, then enters the command plane when the result starts with `/`. Unsupported prompt blocks are rejected by the existing capability boundary. Known commands execute directly; unknown or malformed slash input returns a direct error and never reaches the model. Successful text, expected errors, and thrown-failure diagnostics stream as live `agent_message_chunk` output and settle `end_turn`.
-
-One model prompt or direct command may be in flight per ACP session, independently across sessions. `session/cancel` aborts the direct command when one owns the request; it calls `Agent.cancel()` only for an agent prompt, so cancelling a command cannot destroy unrelated queued or injected agent work. Connection teardown aborts commands and then disposes the owned agents.
-
## Testing
The registry suite covers syntax boundaries, immutable normalization, runtime metadata validation, deterministic sorting, global and scoped shadowing, duplicate rejection, exact disposal, contained change-notification failures, direct invocation, expected and malformed results, synchronous and asynchronous failure, and every abort timing edge at per-file 100% statement, branch, function, and line coverage.
-TUI tests exercise all migrated built-ins, live plugin discovery, help/autocomplete refresh, direct results, unknown-command rejection, raw-input delivery, definition removal, startup rollback, and disposal cancellation. ACP tests use the real SDK connection, agent factory, loop, and JSONL persistence to verify create/load snapshots, dynamic updates, scoped multi-session catalogs, supported-block flattening, direct success/error/failure, unknown-command isolation, cancellation, and the absence of model requests or session messages. The SDK helper suite pins direct-ACP composition. Keyless ACP and terminal snapshots pin the new protocol and rendered transcript shapes.
+TUI tests exercise all migrated built-ins, live plugin discovery, help/autocomplete refresh, direct results, unknown-command rejection, raw-input delivery, definition removal, startup rollback, and disposal cancellation. Keyless terminal snapshots pin the rendered help, error, and command-result shapes.
## Alternatives considered
-- **Keep adapter-local switches** — rejected because optional plugins cannot contribute discovery and behavior without editing every front door.
+- **Keep adapter-local switches** — rejected because optional plugins cannot contribute discovery and behavior without editing the TUI.
- **Represent human commands as model tools** — rejected because discovery and direct invocation are human UI behavior; routing through the model adds latency, token cost, and reinterpretation.
-- **Put the registry in the core agent spine** — rejected because headless and JSON-RPC agents do not consume it, while the two UI app bundles can compose it explicitly.
+- **Put the registry in the core agent spine** — rejected because UI-less front doors do not consume it, while TUI can compose it explicitly.
- **Make `dsh-agent-loop` inject commands** — rejected because the loop does not execute or discover human commands. Agent-scoped producers declare the UI dependency in a child plugin instead.
- **Attach adapter masks to each definition** — rejected because support is a composition fact, not command-domain state. Every composed adapter exposes a registered command; an incompatible plugin omits registration in that deployment.
- **Send unknown slash input to the model** — rejected because typoed or unavailable direct actions must fail predictably rather than change execution planes.
- **Persist generic command input and output** — rejected because adapter notices are not model-visible state. A handler that changes durable behavior calls the owning domain API, which records its own events.
-- **Restrict ACP commands to one text block** — rejected because ACP v1 permits accompanying content; the bridge already has a lossless accepted-block translation.
## Consequences
-- Command producers are ordinary removable plugins, and TUI/ACP share one validated catalog and dispatch contract.
+- Command producers are ordinary removable plugins, and TUI consumes their validated catalog and dispatch contract.
- Agent-specific definitions retain existing flat scope and shadow semantics without a core-to-UI dependency.
- Unknown slash input and command output are deterministic UI behavior with zero direct model tokens.
-- ACP clients receive current per-session snapshots after creation, load, registration, and HMR removal.
- Direct command cancellation is isolated from model-turn cancellation.
## Known limitations and deferred work
-- Input metadata is ACP's current unstructured text hint. Typed forms, argument schemas, and completion providers remain command-owned or require a later protocol extension.
-- Generic command output is live-only and is not reconstructed after TUI restart or ACP reconnect.
+- Input metadata is limited to an unstructured text hint. Typed forms, argument schemas, and completion providers remain command-owned or require a later registry or consumer extension.
+- Generic command output is live-only and is not reconstructed after TUI restart.
- Registry cancellation stops awaiting immediately, but external work stops only when a handler cooperates with its signal.
-- The headless CLI and JSON-RPC SDK front doors do not expose the command plane; only TUI and ACP consume it.
+- The ACP automation server, headless CLI, and JSON-RPC SDK front doors do not expose the command plane; only TUI consumes it.
diff --git a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.zh.md b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.zh.md
index 054ab3a90e..27757f05af 100644
--- a/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-19-plugin-command-registration.zh.md
@@ -6,17 +6,17 @@ Status: implemented
## 问题
-TUI 拥有七个斜杠命令,而 ACP 定义了标准命令目录与调用形态。如果命令名、帮助文本、自动补全、分派和取消都留在各适配器内部,每个新命令都需要修改适配器,可选插件无法贡献命令,两个前端也会逐渐偏离。把斜杠输入当作普通模型提示同样不安全:用户可见的直接操作可能意外消耗 token,或让模型重新解释未知命令。
+TUI 拥有斜杠命令。如果命令名、帮助文本、自动补全、分派和取消都留在适配器内部,每个新命令都需要修改 TUI,可选插件也无法贡献命令。把斜杠输入当作普通模型提示同样不安全:用户可见的直接操作可能意外消耗 token,或让模型重新解释未知命令。
-共享机制必须仍是 UI 关注点,而不是模型工具或智能体循环分支。它还需要精确的逐智能体可见性、可安全 HMR 移除、逐会话 ACP 发现、直接结果渲染和请求作用域取消,同时不会自动把命令文本或输出加入模型历史。
+共享机制必须仍是 UI 关注点,而不是模型工具或智能体循环分支。它还需要精确的逐智能体可见性、可安全 HMR 移除、直接结果渲染和请求作用域取消,同时不会自动把命令文本或输出加入模型历史。
## 决策
-位于 `packages/ui/commands/` 的 `@deepseek-ai/dsh-commands` 是产品命令注册表。终端与 ACP 应用 bundle(组合包)把它挂载在消费该服务的前端旁,SDK 项目 helper(辅助器)在直接搭建 ACP 时也会生成同一服务;无执行器、无 UI 的智能体 spine(主干)保持独立。TUI 与 ACP 注入该服务,命令生产者只依赖注册表及其操作的领域。
+位于 `packages/ui/commands/` 的 `@deepseek-ai/dsh-commands` 是产品命令注册表。TUI 应用 bundle(组合包)把它挂载在消费该服务的前端旁;[仅面向自动化的 ACP(Agent Client Protocol)应用](../simplification/2026-07-23-acp-automation-only-protocol.md)和无执行器、无 UI 的智能体 spine(主干)都省略该服务。TUI 注入该服务,命令生产者只依赖注册表及其操作的领域。
### 注册表契约
-`CommandDefinition` 包含不带 `/` 的小写名称、非空描述、可选的非结构化输入提示,以及可取消处理器。注册会校验并分离元数据、冻结有效定义,并返回准确的 Cordis effect disposer(副作用释放器)。同一层中的重复名称会失败。每个消费该注册表的适配器都能看到所有有效定义;若命令插件无法在某种部署中运行,它就不在该部署中注册,而不是把适配器身份编码进共享领域。
+`CommandDefinition` 包含不带 `/` 的小写名称、非空描述、可选的非结构化输入提示,以及可取消处理器。注册会校验并分离元数据、冻结有效定义,并返回准确的 Cordis effect disposer(副作用释放器)。同一层中的重复名称会失败。每个消费方都能看到所有有效定义;若命令插件无法在某种部署中运行,它就不在该部署中注册,而不是把消费方身份编码进共享领域。
`list(agent)` 在作用域遮蔽后返回不可变、按名称排序的描述符。`find(agent, name)` 解析有效定义。`execute(agent, line, signal)` 解析并运行已知定义,返回分离后的 `success` 或 `error` 结果;无效语法和未知名称返回 `undefined`,由适配器拥有直接错误文本。
@@ -36,46 +36,36 @@ TUI 拥有七个斜杠命令,而 ACP 定义了标准命令目录与调用形
### TUI 映射
-TUI 把 `help`、`clear`、`cancel`、`reasoning`、`tools`、`redraw` 和 `exit` 注册为智能体作用域命令定义,不再对字符串执行 switch。自动补全与帮助视图读取实时目录,因此插件命令会随其副作用出现和消失。任何以 `/` 开头的提交行都留在命令平面;未知输入产生终端警告,不会落入 `Agent.send()` 或 `Agent.steer()`。
+TUI 把内置斜杠命令注册为智能体作用域命令定义,不再对字符串执行 switch。自动补全与帮助视图读取实时目录,因此插件命令会随其副作用出现和消失。任何以 `/` 开头的提交行都留在命令平面;未知输入产生终端警告,不会落入 `Agent.send()` 或 `Agent.steer()`。
每个提交的命令拥有一个 `AbortController`。TUI 释放会中止未完成的分派、移除本地定义,并等待命令生产者 fiber(纤程)后再完成清理。
-### ACP 映射
-
-桥接遵循当前的 [ACP v1 斜杠命令契约](https://agentclientprotocol.com/protocol/v1/slash-commands)。`session/new` 与 `session/load` 发出准确智能体的完整 `available_commands_update` 快照;新会话的 RPC 响应会先引入服务端生成的 id,随后快照才会入队。每次注册表变更都会为每个实时会话发出替换快照。名称、描述和可选非结构化输入提示直接映射到 `AvailableCommand`。
-
-ACP 允许命令提示携带额外的受支持内容块。桥接应用普通的无损 `text` 与 `resource_link` 扁平化,然后在结果以 `/` 开头时进入命令平面。不支持的提示块由现有能力边界拒绝。已知命令直接执行;未知或格式错误的斜杠输入返回直接错误,绝不会到达模型。成功文本、预期错误和抛出失败的诊断作为实时 `agent_message_chunk` 输出流式发送,并以 `end_turn` 结束请求。
-
-每个 ACP 会话同时只能有一个模型提示或直接命令进行中,各会话彼此独立。当直接命令拥有请求时,`session/cancel` 会中止它;只有智能体提示才调用 `Agent.cancel()`,因此取消命令不会销毁无关的排队或注入智能体工作。连接清理会先中止命令,再释放所拥有的智能体。
-
## 测试
注册表测试覆盖语法边界、不可变规范化、运行时元数据校验、确定性排序、全局与作用域遮蔽、重复拒绝、准确释放、变更通知失败隔离、直接调用、预期和格式错误结果、同步与异步失败,以及每种中止时序边沿;该源文件达到逐文件 100% 语句、分支、函数和行覆盖率。
-TUI 测试覆盖全部迁移后的内置命令、实时插件发现、帮助与自动补全刷新、直接结果、未知命令拒绝、原始输入交付、定义移除、启动回滚和释放取消。ACP 测试使用真实 SDK 连接、智能体工厂、循环与 JSONL 持久化,验证创建/加载快照、动态更新、作用域多会话目录、受支持块扁平化、直接成功/错误/失败、未知命令隔离、取消,以及不存在模型请求或会话消息。SDK helper 测试固定直接 ACP 组合。无密钥 ACP 与终端快照固定新的协议和渲染记录形态。
+TUI 测试覆盖全部迁移后的内置命令、实时插件发现、帮助与自动补全刷新、直接结果、未知命令拒绝、原始输入交付、定义移除、启动回滚和释放取消。无密钥终端快照固定渲染后的帮助、错误与命令结果形态。
## 考虑过的替代方案
-- **保留适配器本地 switch**——不予采纳,因为可选插件无法贡献发现与行为,除非修改每个前端。
+- **保留适配器本地 switch**——不予采纳,因为可选插件无法贡献发现与行为,除非修改 TUI。
- **把人类命令表示为模型工具**——不予采纳,因为发现与直接调用属于人类 UI 行为;经由模型路由会增加延迟、token 成本和重新解释。
-- **把注册表放入核心智能体主干**——不予采纳,因为无头和 JSON-RPC 智能体不消费它,而两个 UI 应用组合包可以显式组合它。
+- **把注册表放入核心智能体主干**——不予采纳,因为无 UI 前端不消费它,而 TUI 可以显式组合它。
- **让 `dsh-agent-loop` 注入 commands**——不予采纳,因为循环不执行也不发现人类命令。智能体作用域生产者改为在子插件中声明 UI 依赖。
- **为每个定义附加适配器掩码**——不予采纳,因为支持能力是组合事实,而不是命令领域状态。每个已组合适配器都暴露已注册命令;不兼容插件不会在该部署中注册。
- **把未知斜杠输入发送给模型**——不予采纳,因为输入错误或不可用的直接操作必须可预测地失败,而不能改变执行平面。
- **持久化通用命令输入与输出**——不予采纳,因为适配器提示不是模型可见状态。改变持久行为的处理器会调用拥有该状态的领域 API,由后者记录自己的事件。
-- **把 ACP 命令限制为单个文本块**——不予采纳,因为 ACP v1 允许附带内容,而桥接已有无损的已接纳块转换。
## 后果
-- 命令生产者是普通的可移除插件,TUI 与 ACP 共享一个经过校验的目录和分派契约。
+- 命令生产者是普通的可移除插件,TUI 消费其经过校验的目录与分派契约。
- 智能体特定定义保留现有扁平作用域与遮蔽语义,不引入核心到 UI 的依赖。
- 未知斜杠输入与命令输出是确定性 UI 行为,直接模型 token 成本为零。
-- ACP 客户端在创建、加载、注册和 HMR 移除后收到当前的逐会话快照。
- 直接命令取消与模型轮次取消彼此隔离。
## 已知限制与延期工作
-- 输入元数据仅为 ACP 当前的非结构化文本提示。类型化表单、参数模式和补全提供器仍由命令拥有,或需要后续协议扩展。
-- 通用命令输出仅实时存在,TUI 重启或 ACP 重新连接后不会重建。
+- 输入元数据仅限非结构化文本提示。类型化表单、参数模式和补全提供器仍由命令拥有,或需要后续注册表或消费方扩展。
+- 通用命令输出仅实时存在,TUI 重启后不会重建。
- 注册表取消会立即停止等待,但外部工作只有在处理器配合信号时才会停止。
-- 无头 CLI 与 JSON-RPC SDK 前端不暴露命令平面;只有 TUI 和 ACP 消费它。
+- ACP 自动化服务器、无头 CLI 与 JSON-RPC SDK 前端不暴露命令平面;只有 TUI 消费它。
diff --git a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.i18n.yaml b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.i18n.yaml
index 27e29c10a9..8f948d3d45 100644
--- a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.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
-2026-07-19-same-session-goal-round-driver.md: f4eb2a279da14dca9b6ab2fe7297f9329575e4f8
-2026-07-19-same-session-goal-round-driver.zh.md: ae5cf7da2560a76131c11fe0c857ea5f48bc6809
+2026-07-19-same-session-goal-round-driver.md: d23af9a9b05d60d2dccad095455524844f1185b9
+2026-07-19-same-session-goal-round-driver.zh.md: f4f0cd6fd575d14427025bdbd8d10bc90e25f780
diff --git a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.md b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.md
index f4eb2a279d..0e6be9fe31 100644
--- a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.md
+++ b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.md
@@ -69,7 +69,7 @@ An inbox acceptance can win the microtask race immediately before plugin unload
The unit suite uses the real agent loop and session service with only the model scripted. It covers exact sequential admission and cap enforcement, load/resume inertness, every outcome classification, rate limiting, request errors, max tokens, downstream prompt veto, pre-admission and in-flight cancellation, unrelated-human cancellation, failed-pause fallback, human-input ordering, queued and downstream revision races, forged goal attribution, failed mutation and turn checkpoints including a later one-shot injection, scheduler and custom-agent failures, session-start reset, exact lifecycle retirement, and queued/running plugin teardown. The new driver source has per-file 100% statement, branch, function, and line coverage.
-A keyless ACP snapshot mounts the shipped editor app with the real goal domain, goal tools, goal driver, agent loop, persistence, and replay adapter through `cordis.yml`. One human turn creates and inspects a two-round goal, the first automatic turn stops normally, and ACP cancellation of a deliberately stalled second round records a durable pause. The normalized wire transcript and external JSONL assertions prove one session, round sources `1, 2`, the lifecycle mutation, and exact replay accounting without using `echo-agent` as an application surrogate.
+A keyless ACP snapshot mounts the shipped automation app with the real goal domain, goal tools, goal driver, agent loop, persistence, and replay adapter through `cordis.yml`. One human-originated turn creates and inspects a two-round goal, the first automatic turn stops normally, and ACP cancellation of a deliberately stalled second round records a durable pause. The normalized wire transcript and external JSONL assertions prove one session, round sources `1, 2`, the lifecycle mutation, and exact replay accounting without using `echo-agent` as an application surrogate.
The core cancellation test proves notification order and containment: observers run only for effective cancellation, can queue replacement work before the inbox clear, cannot veto later observers by throwing, and an idle call emits nothing.
diff --git a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.zh.md b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.zh.md
index ae5cf7da25..cfd9d1aa8c 100644
--- a/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-19-same-session-goal-round-driver.zh.md
@@ -69,7 +69,7 @@ Status: implemented
单元测试使用真实 agent loop 与会话服务,只对模型编写脚本。覆盖内容包括精确连续接纳和上限执行、加载与恢复的惰性、所有结果分类、限流、请求错误、最大 token、下游提示词否决、接纳前与执行中取消、无关人类工作取消、暂停失败回退、人类输入排序、排队时与下游修订竞争、伪造目标来源、变更与轮次检查点失败(包括后续一次性注入)、调度器与自定义 agent 失败、会话启动重置、精确生命周期退出,以及排队中和运行中的插件卸载。新驱动器源码达到逐文件 100% 语句、分支、函数和行覆盖率。
-无密钥 ACP 快照通过 `cordis.yml` 挂载已发布的编辑器应用,以及真实目标领域、目标工具、目标驱动器、agent loop、持久化和回放适配器。一个人类轮次创建并检查一个两回合目标;第一个自动轮次正常停止,ACP 随后取消刻意停滞的第二个回合并记录持久暂停。规范化线协议和外部 JSONL 断言证明只有一个会话、回合来源依次为 `1, 2`、生命周期变更与回放计数精确,并且没有把 `echo-agent` 当作应用替身。
+无密钥 ACP 快照通过 `cordis.yml` 挂载已发布的自动化应用,以及真实目标领域、目标工具、目标驱动器、agent loop、持久化和回放适配器。一个源自人类的轮次创建并检查一个两回合目标;第一个自动轮次正常停止,ACP 随后取消刻意停滞的第二个回合并记录持久暂停。规范化线协议和外部 JSONL 断言证明只有一个会话、回合来源依次为 `1, 2`、生命周期变更与回放计数精确,并且没有把 `echo-agent` 当作应用替身。
核心取消测试固定通知顺序与隔离:只有有效取消才会通知;观察者可以在清空收件箱前排入替代工作;抛错不能阻止后续观察者;空闲调用不会发出事件。
diff --git a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.i18n.yaml b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.i18n.yaml
index 0f1bead590..190a2b773f 100644
--- a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.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
-2026-07-20-code-mode-typed-tool-returns.md: 29f139a7e965de3a374d195ecc205210e6ae7e93
-2026-07-20-code-mode-typed-tool-returns.zh.md: 431c0b1717c6783771255ce8291c241f8f92c30b
+2026-07-20-code-mode-typed-tool-returns.md: 1b5cbb9f4664c371a03cfefd079d5dc531711b51
+2026-07-20-code-mode-typed-tool-returns.zh.md: 70fd9ee72803c5cd2fa228a1d38ddf7ae47a4814
diff --git a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.md b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.md
index 29f139a7e9..1b5cbb9f46 100644
--- a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.md
+++ b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.md
@@ -75,7 +75,7 @@ Dynamic Cordis mounting follows the same rule: `cordis_mount` returns `{ id, plu
Nested dispatch keeps the existing bounded `tool/code-dispatch.resultSummary` for diagnostics but does not persist canonical values. `tool/result` continues to persist only rendered content, error, and optional metadata. This is deliberately not a session-format change, so `SESSION_FORMAT_VERSION` remains unchanged and replay cannot recreate intermediate program values.
-The opaque `exec.parent` token marks nested calls. Presentation metadata and generic or tool-owned spill projections skip those calls because they have no direct result card and their canonical values never enter context. The outer `run_code` call alone produces one card and may spill its final post-policy presentation; `run_code` intentionally declares neither a result presenter nor presentation metadata, so ACP and TUI complete the card through their generic raw-content fallback using durable `tool/result.content`.
+The opaque `exec.parent` token marks nested calls. Presentation metadata and generic or tool-owned spill projections skip those calls because they have no direct result card and their canonical values never enter context. The outer `run_code` call alone produces one card and may spill its final post-policy presentation; `run_code` intentionally declares neither a result presenter nor presentation metadata, so UI adapters complete the card through their generic raw-content fallback using durable `tool/result.content`.
## Testing
@@ -95,7 +95,7 @@ Keyless real-worker integration tests pin the two handle workflows that prose re
## Consequences
-Code programs can compose tools through stable values instead of reverse-engineering Native prose. Native and Both Mode retain their existing text and editor presentation, while Code Mode receives output-schema types and exact runtime JSON. Tool authors must treat the canonical value as their programmatic API and put display-only formatting in the renderer.
+Code programs can compose tools through stable values instead of reverse-engineering Native prose. Native and Both Mode retain their existing text and UI presentation, while Code Mode receives output-schema types and exact runtime JSON. Tool authors must treat the canonical value as their programmatic API and put display-only formatting in the renderer.
The worker performs bounded-depth flat-wire transport and lossless validation but does not make intermediate values cheap or durable. Outer overflow is an explicit failed run, and error handling remains intentionally human-guided rather than a versioned code union.
diff --git a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.zh.md b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.zh.md
index 431c0b1717..70fd9ee728 100644
--- a/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-20-code-mode-typed-tool-returns.zh.md
@@ -75,7 +75,7 @@ Code Mode 通过运行时请求中的 `{ name: "ToolCallError", memberNameProper
嵌套分发会为诊断保留既有的有界 `tool/code-dispatch.resultSummary`,但不会持久化规范值。`tool/result` 继续只持久化渲染后的内容、错误和可选元数据。这并非会话格式变更,因此 `SESSION_FORMAT_VERSION` 保持不变,回放也无法重建程序的中间值。
-不透明的 `exec.parent` token 用于标识嵌套调用。由于这些调用没有直接对应的结果卡片,而且其规范值永远不会进入上下文,展示元数据以及通用或工具自有的输出落盘投影都会跳过它们。只有外层 `run_code` 调用会生成一张卡片,并且可能将 post-policy 处理后的最终展示写入落盘文件;`run_code` 有意既不声明结果展示器,也不声明展示元数据,因此 ACP 和 TUI 通过其通用的原始内容回退机制,使用持久化的 `tool/result.content` 补全该卡片。
+不透明的 `exec.parent` token 用于标识嵌套调用。由于这些调用没有直接对应的结果卡片,而且其规范值永远不会进入上下文,展示元数据以及通用或工具自有的输出落盘投影都会跳过它们。只有外层 `run_code` 调用会生成一张卡片,并且可能将 post-policy 处理后的最终展示写入落盘文件;`run_code` 有意既不声明结果展示器,也不声明展示元数据,因此 UI 适配器会通过通用的原始内容回退机制,使用持久化的 `tool/result.content` 补全该卡片。
## 测试
@@ -95,7 +95,7 @@ Code Mode 通过运行时请求中的 `{ name: "ToolCallError", memberNameProper
## 影响
-Code Mode 程序可以通过稳定值组合工具,无需逆向解析 Native 自然语言。Native 和 Both Mode 保留现有文本与编辑器展示,Code Mode 则获得输出 schema 类型和精确的运行时 JSON。工具作者必须把规范值视为程序化 API,并将仅用于展示的格式化放入渲染器。
+Code Mode 程序可以通过稳定值组合工具,无需逆向解析 Native 自然语言。Native 和 Both Mode 保留现有文本与 UI 展示,Code Mode 则获得输出 schema 类型和精确的运行时 JSON。工具作者必须把规范值视为程序化 API,并将仅用于展示的格式化放入渲染器。
worker 会以嵌套深度有界的扁平协议格式传输数据并执行无损校验,但不会降低中间值的开销,也不会使其具备持久性。外层输出溢出会显式导致运行失败,错误处理则有意由人类引导,而不是依赖带版本的错误代码联合。
diff --git a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.md b/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.md
deleted file mode 100644
index a2a22baafd..0000000000
--- a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.md
+++ /dev/null
@@ -1,39 +0,0 @@
-# Agent Note: Startup slogans replace the configured TUI welcome line
-
-Status: implemented
-
-English | [中文](2026-07-20-tui-startup-slogans.zh.md)
-
-> **Superseded** for the slogan/animation half by the [banner sweep Agent Note](2026-07-21-tui-banner-sweep.md): the slogan bank and typewriter reveal shipped, read as weird in use, and were replaced by a subtitle-free banner with a whole-banner sweep. The removal of the configured demo welcome and the animation-lifecycle groundwork (start after `ui.start()`, clear through `detachListeners`) stand.
-
-## Problem
-
-The TUI header subtitle came from a `welcome` config the demo leaf set to "TUI agent ready. Give it a coding task." — instructional filler that told a returning user nothing, restated what the product is on every boot, and had a hardcoded twin (`'ready.'`) as the schema default in two packages. The product wanted a startup moment with some character instead of a static banner caption.
-
-## Decision
-
-- `examples/tui-agent/cordis.yml` no longer configures `welcome`; the config key stays for deployments and fixtures that need a fixed, deterministic subtitle (the Code Mode overlay and every snapshot/scripted fixture keep theirs).
-- When `welcome` is unset, `dsh-tui` picks one member of an exported `STARTUP_SLOGANS` bank per boot (`pickStartupSlogan`, injectable random source) and reveals it with a typewriter animation: one character per 40 ms frame, a `▌` block cursor trailing until complete. The reveal starts only after `ui.start()` succeeds and its interval is cleared on dispose alongside the other listeners.
-- The slogan bank is presentation copy, deliberately not config: deployments that want controlled wording already have `welcome`. Slogans are ASCII-only by contract because the reveal slices per character.
-- `dsh-tui-demo` forwards `welcome` only when configured instead of defaulting it, so the app no longer decides the TUI's idle subtitle.
-- The keyless PTY boot scenario now waits for the reveal cursor (`▌` — the only source of that glyph in an empty transcript) instead of the removed welcome text.
-
-The same change restores `packages/ui/tui/src/index.ts` to 100 % per-file coverage, which the color-scheme merge had broken on the integration branch: the editor border-color reassignment inside `applyColorScheme` was dead (the `setStatus` call right after re-derives it) and is removed, and the color-scheme query's `.then`/`.catch` arrows became named, tested handlers (`applyReportedScheme`, `ignoreSchemeQueryFailure` — the latter pinned by a test whose terminal throws on the DSR query write).
-
-## Alternatives considered
-
-**A fixed cooler slogan.** Rejected: one string re-read on every boot decays into wallpaper exactly like the line it replaces; a small rotating bank keeps the moment alive at no complexity cost.
-
-**Making the bank and reveal speed configurable.** Rejected: that is two new knobs for presentation copy; `welcome` is already the escape hatch for deployments with an opinion, and the no-hardcoded-tunables rule targets deployment-varying behavior, not brand copy.
-
-**Animating in `HeaderComponent` itself.** Rejected: the component would need a TUI handle and its own lifecycle; the chat already owns a render loop, timers, and a disposal path, so the reveal lives beside the other `createTuiChat` effects and `detachListeners` clears it.
-
-## Consequences
-
-- Boot output is no longer byte-deterministic when `welcome` is unset (random slogan, timed frames). Every recorded or snapshot surface pins `welcome` explicitly, so no snapshot changed; the PTY smoke anchors on the reveal cursor and the session-id line instead.
-- The `welcome` schema default disappeared from both `dsh-tui` and `dsh-tui-demo`; a direct caller passing no welcome now gets a slogan, not `'ready.'`.
-- Adding a slogan is a one-line bank edit; tests assert membership, not specific text.
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` pins deterministic bank selection with an injected random source, the reveal (a bank member fully rendered, cursor frames observed), the configured-welcome path rendering verbatim with no cursor, and dispose stopping a mid-reveal animation. `examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` boots the real tree in a PTY and waits on the reveal cursor. Verified live in tmux (mid-reveal frame `no map below▌` then the full slogan).
diff --git a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.zh.md b/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.zh.md
deleted file mode 100644
index 58fa5790f3..0000000000
--- a/.agents/notes/implemented/feature/2026-07-20-tui-startup-slogans.zh.md
+++ /dev/null
@@ -1,39 +0,0 @@
-# Agent Note: 启动 slogan 取代配置化的 TUI 欢迎语
-
-Status: implemented
-
-[English](2026-07-20-tui-startup-slogans.md) | 中文
-
-> **已被取代**:slogan/动画的那一半由[横幅扫入 Agent Note](2026-07-21-tui-banner-sweep.md)取代:slogan 库和打字机动画上线后实际使用中显得怪异,已替换为无副标题的横幅加整体扫入。移除示例配置中欢迎语的决定与动画生命周期基础设施(`ui.start()` 后启动、经 `detachListeners` 清除)保持不变。
-
-## Problem
-
-TUI 头部副标题来自一个 `welcome` 配置,示例叶子配置把它设为 "TUI agent ready. Give it a coding task."——一句说明书式的填充语,对老用户毫无信息量,每次启动都在复述产品是什么,而且它还有一个硬编码的孪生兄弟(`'ready.'`)作为两个包里的 schema 默认值。产品需要的是一个有性格的启动时刻,而不是一条静态横幅说明。
-
-## Decision
-
-- `examples/tui-agent/cordis.yml` 不再配置 `welcome`;该配置键保留给需要固定、确定性副标题的部署与 fixture(Code Mode overlay 和所有快照/脚本化 fixture 都保留各自的欢迎语)。
-- `welcome` 未设置时,`dsh-tui` 每次启动从导出的 `STARTUP_SLOGANS` 库里挑选一条(`pickStartupSlogan`,随机源可注入),并以打字机动画逐字显示:每帧 40 ms 一个字符,完成前尾随一个 `▌` 块状光标。动画只在 `ui.start()` 成功后启动,其定时器与其他监听器一起在 dispose 时清除。
-- slogan 库是展示文案,刻意不做成配置:想控制措辞的部署已经有 `welcome` 这个出口。按契约 slogan 只含 ASCII,因为逐字显示按字符切片。
-- `dsh-tui-demo` 只在配置了 `welcome` 时才转发它,不再填默认值,应用不再替 TUI 决定空闲副标题。
-- 无 key 的 PTY 启动场景改为等待逐字显示的光标(`▌`——空 transcript 里该字形的唯一来源),不再等待已删除的欢迎文本。
-
-同一变更把 `packages/ui/tui/src/index.ts` 恢复到 100% 的单文件覆盖率(颜色方案合并曾在集成分支上破坏它):`applyColorScheme` 里对编辑器边框颜色的重新赋值是死代码(紧随其后的 `setStatus` 调用会重新推导它),已删除;颜色方案查询的 `.then`/`.catch` 箭头函数改为具名、有测试的处理器(`applyReportedScheme`、`ignoreSchemeQueryFailure`——后者由一个让终端在 DSR 查询写入时抛错的测试固定)。
-
-## Alternatives considered
-
-**换一条更酷的固定 slogan。** 否决:一条每次启动都重读的字符串会和它取代的那行一样退化成墙纸;一个小的轮换库以零复杂度代价让这个时刻保持新鲜。
-
-**把 slogan 库和显示速度做成配置。** 否决:那是为展示文案新增两个旋钮;对措辞有主张的部署已经有 `welcome` 这个出口,而「插件里不许硬编码可调参数」规则针对的是随部署变化的行为,不是品牌文案。
-
-**在 `HeaderComponent` 内部做动画。** 否决:组件将需要持有 TUI 句柄和自己的生命周期;聊天层已经拥有渲染循环、定时器和释放路径,所以逐字显示与 `createTuiChat` 的其他资源放在一起,由 `detachListeners` 清除。
-
-## Consequences
-
-- `welcome` 未设置时启动输出不再字节级确定(随机 slogan、定时帧)。所有录制或快照表面都显式固定 `welcome`,因此没有快照变化;PTY 冒烟测试改为锚定逐字显示光标和会话 id 行。
-- `welcome` 的 schema 默认值从 `dsh-tui` 和 `dsh-tui-demo` 中消失;不传 welcome 的直接调用方现在得到的是 slogan,而不是 `'ready.'`。
-- 新增一条 slogan 只需在库里加一行;测试断言成员归属,不断言具体文本。
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` 固定以下行为:注入随机源后的确定性选取、逐字显示(库中某条完整渲染、观察到光标帧)、配置了 welcome 时逐字动画不启动且原文渲染、以及 dispose 停止进行中的动画。`examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` 在 PTY 里启动真实配置树并等待显示光标。已在 tmux 中实机验证(中途帧 `no map below▌`,随后是完整 slogan)。
diff --git a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.i18n.yaml
index 9c0118a72d..b1f73ed15a 100644
--- a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.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
-2026-07-21-cross-session-references.md: bfa015b24cda6c8651829b6a7f0800326da5b502
-2026-07-21-cross-session-references.zh.md: e8e99124f7e2ccfe9fbe97323c17143372017562
+2026-07-21-cross-session-references.md: fc084b36e7920a72efff0f363278d24eaebc4c69
+2026-07-21-cross-session-references.zh.md: fe4a876b5265fa7ad298adf3b829bcec70e878e8
diff --git a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.md b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.md
index bfa015b24c..fc084b36e7 100644
--- a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.md
+++ b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.md
@@ -6,13 +6,13 @@ English | [中文](2026-07-21-cross-session-references.zh.md)
## Problem
-TUI and ACP users need to bring relevant work from another conversation into one new message without resuming, forking, or granting the source transcript authority over the current session. The harness already exposes exact session enumeration and raw event inspection, but every host independently parsing logs would duplicate compaction folding, provenance filtering, size limits, error behavior, and persistence. Encoding host markup directly into the agent message contract would also bind the core loop to one UI syntax.
+TUI users need to bring relevant work from another conversation into one new message without resuming, forking, or granting the source transcript authority over the current session. The harness already exposes exact session enumeration and raw event inspection, but every host independently parsing logs would duplicate compaction folding, provenance filtering, size limits, error behavior, and persistence. Encoding host markup directly into the agent message contract would also bind the core loop to one UI syntax.
## Decision
`@deepseek-ai/dsh-session-reference` is one context consumer service at `ctx.sessionReferences`. Hosts normalize their protocol into `SessionReferenceInput[]`, call `prepare()` before enqueue, and pass the returned contexts through the generic `SendOptions.contexts` boundary. Core agent packages know only that one queued message may carry frozen `HookContext[]`; they do not parse session URIs or read another log.
-`dsh-session:` is the canonical host-independent identifier. JSON string encoding precedes base64url so quotes, slashes, backslashes, Unicode, newlines, and every other JavaScript string value round-trip without delimiter ambiguity. TUI renders that URI inside `@[label](uri)` and ACP uses standard `resource_link`; text-only clients may use the same inline mention. Explicit Markdown mentions and resource links reject malformed URIs. Bare text becomes a reference only for a non-empty base64url-shaped payload, whose decode must still be canonical; empty or punctuation-only uses remain ordinary discussion text.
+`dsh-session:` is the canonical host-independent identifier. JSON string encoding precedes base64url so quotes, slashes, backslashes, Unicode, newlines, and every other JavaScript string value round-trip without delimiter ambiguity. TUI renders that URI inside `@[label](uri)`; text-only clients may use the same inline mention. Explicit Markdown mentions reject malformed URIs. Bare text becomes a reference only for a non-empty base64url-shaped payload, whose decode must still be canonical; empty or punctuation-only uses remain ordinary discussion text.
The service uses `ctx.sessionQuery.readSurface(sessionId)`, which loads one live-preferred corpus observation, folds it with the session package's canonical surface algorithm, and returns a detached header, capture seq, and current nodes. FTS is not a dependency: v1 discovery filters only id and cwd, and future title/body search can replace the candidate layer without changing reference identity or preparation.
@@ -28,13 +28,13 @@ One aggregated context is serialized as JSON beneath a fixed untrusted-backgroun
`send()` and `steer()` snapshot content, resolved source, and contexts together as one deeply frozen lossless-JSON inbox record. Synthetic `inject()` accepts source and model-hidden metadata but not attached contexts, which belong to inbox messages. A claimed ordinary message exposes its attached contexts as the default `agent/prompt-submit` additional contexts; a block writes neither user message nor contexts. The waterfall's returned allow is authoritative, so a listener wrapping `next()` preserves downstream content and contexts unless it intentionally replaces them. After admission, absent or `separate` placement writes an independent `context/message`, while `prompt-prefix` placement bakes context and the effective request into one prompt event. Drained steering bypasses `agent/prompt-submit` but applies the same placement split. Late steering retains the same record when converted to queued input, while cancellation, disposal, and terminal discard drop message and contexts together. `agent/queued` reports the frozen contexts so the observation event describes the complete retained item.
-This preserves host driving semantics: TUI decides `send()` versus `steer()` from the agent state after preparation, so only its queued path dispatches UserPromptSubmit hooks; ACP continues to call `send()` once per `session/prompt`. Reference preparation is not a new steering protocol and does not create a turn by itself.
+This preserves host driving semantics: TUI decides `send()` versus `steer()` from the agent state after preparation, so only its queued path dispatches UserPromptSubmit hooks. Reference preparation is not a new steering protocol and does not create a turn by itself.
## Host adapters
TUI combines session candidates with the existing `@` file provider. Each candidate displays the latest folded session title and falls back to the session id; lookup follows the editor's cancellation signal, and session id, cwd, and mention labels escape external terminal controls while the canonical URI retains the original id. TUI prepares only submissions containing structured mentions, disables duplicate submit while awaiting snapshots, restores failed input, renders the prompt envelope's display content as the user message, and renders its session-reference metadata as a compact source list instead of exposing the complete JSON in the terminal.
-ACP detects direct slash commands from ordinary prompt flattening before extracting `dsh-session:` resource links and canonical inline mentions, so URI-shaped command arguments remain opaque while ordinary resource-link rendering is preserved. Standard `session/list` exposes each loadable session's folded title and, when references are mounted, a canonical URI under `_meta["deepseek-harness/sessionReference"]`; a client can use `title ?? sessionId` as the resource-link name. A valid reference without the optional service returns a capability-unavailable RPC error, and preparation failure occurs before the in-flight turn slot and agent send. A preparation-specific abort owner makes `session/cancel` and bridge teardown stop pending reads. Picker UI remains an ACP client responsibility because ACP does not define a cross-session mention menu.
+The [automation-only ACP transport](../simplification/2026-07-23-acp-automation-only-protocol.md) deliberately does not mount session-query or session-reference services.
## Budget and retention
@@ -43,8 +43,8 @@ Each of at most three references is independently capped at 65,536 UTF-8 bytes b
## Alternatives considered
- **Wait for SQLite FTS5** — rejected because snapshot correctness requires exact id reads and canonical surface folding, not content search. FTS improves discovery only.
-- **Put mention syntax in `Agent.send()`** — rejected because it would make the core protocol parse TUI/ACP presentation and prevent typed non-text hosts from sharing the semantic layer.
-- **Implement references inside TUI and ACP separately** — rejected because projection, security warning, retention, and persistence would drift across hosts.
+- **Put mention syntax in `Agent.send()`** — rejected because it would make the core protocol parse one host's presentation syntax and prevent typed non-text hosts from sharing the semantic layer.
+- **Implement references separately in each host** — rejected because projection, security warning, retention, and persistence would drift across hosts.
- **Place a separate user-role context message beside the prompt** — rejected because two adjacent user messages weaken the prompt's deictic binding: in `@foo what does this session discuss?`, the model may resolve “this session” as the current conversation instead of the referenced snapshot.
- **Bake the prefix host-side before `send()`** — rejected because `agent/prompt-submit` must inspect and rewrite only the direct prompt. The effective prompt and attached contexts meet only after admission in AgentLoop, which can apply an `allow.content` rewrite consistently to both combined model content and `envelope.displayContent`; earlier host assembly would expose snapshot bytes to the hook or let those two views diverge.
- **Replay the raw source log or restore shadowed events** — rejected because compact defines the current model surface and may intentionally retire sensitive or expensive history.
@@ -53,8 +53,8 @@ Each of at most three references is independently capped at 65,536 UTF-8 bytes b
## Verification
-Unit and integration coverage pins URI round-trips and text-boundary punctuation, explicit malformed references, title-aware candidate ranking, terminal-control escaping, projection exclusions, non-recursive prompt-envelope projection, backend-independent compact checkpoints, tag-safe framing, deduplication, self-reference, count limits, all-or-nothing reads, prompt cancellation against a non-settling storage read, independent per-source byte retention, frozen message ownership, prompt blocking, send/steer placement, title isolation, missing capability, title-aware ACP session listing, ordinary ACP resource links, opaque ACP command arguments, and compact TUI/ACP replay. A keyless TUI snapshot runs the real agent loop: the source surface replaces old user/assistant history with a compact checkpoint, the target submits a mention, and the captured model request contains one user message ordered as snapshot, request delimiter, and current prompt, without either shadowed string.
+Unit and integration coverage pins URI round-trips and text-boundary punctuation, explicit malformed references, title-aware candidate ranking, terminal-control escaping, projection exclusions, non-recursive prompt-envelope projection, backend-independent compact checkpoints, tag-safe framing, deduplication, self-reference, count limits, all-or-nothing reads, prompt cancellation against a non-settling storage read, independent per-source byte retention, frozen message ownership, prompt blocking, send/steer placement, title isolation, missing capability, and compact TUI replay. A keyless TUI snapshot runs the real agent loop: the source surface replaces old user/assistant history with a compact checkpoint, the target submits a mention, and the captured model request contains one user message ordered as snapshot, request delimiter, and current prompt, without either shadowed string.
## Consequences
-The new plugin is the stable semantic boundary and adds no persistence schema, event type, FTS dependency, source subscription, or compact shadow access. Standard TUI/ACP demo bundles mount it explicitly and expose its count and per-source byte limits in their own config; custom hosts remain unchanged until they mount the service and adapt their input. Reference contexts increase target history size within configured bounds and can later be summarized by ordinary target compaction, after which the source session is irrelevant.
+The new plugin is the stable semantic boundary and adds no persistence schema, event type, FTS dependency, source subscription, or compact shadow access. The standard TUI demo bundle mounts it explicitly and exposes its count and per-source byte limits in its config; custom hosts remain unchanged until they mount the service and adapt their input. Reference contexts increase target history size within configured bounds and can later be summarized by ordinary target compaction, after which the source session is irrelevant.
diff --git a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.zh.md b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.zh.md
index e8e99124f7..fe4a876b52 100644
--- a/.agents/notes/implemented/feature/2026-07-21-cross-session-references.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-cross-session-references.zh.md
@@ -6,13 +6,13 @@ Status: implemented
## 问题
-TUI 与 ACP(Agent Client Protocol)用户需要把另一场对话中的相关工作带入一条新消息,但不恢复、不 fork,也不让源 transcript(文本记录)对当前会话拥有权威性。harness 已经提供准确的会话枚举与原始事件检查,但若每个宿主都独立解析日志,就会重复实现压缩(compaction)折叠、来源过滤、大小限制、错误行为和持久化。把宿主标记直接编码进 agent(智能体)消息契约,还会让核心循环绑定某一种 UI 语法。
+TUI 用户需要把另一场对话中的相关工作带入一条新消息,但不恢复、不 fork,也不让源 transcript(文本记录)对当前会话拥有权威性。harness 已经提供准确的会话枚举与原始事件检查,但若每个宿主都独立解析日志,就会重复实现压缩(compaction)折叠、来源过滤、大小限制、错误行为和持久化。把宿主标记直接编码进 agent(智能体)消息契约,还会让核心循环绑定某一种 UI 语法。
## 决策
`@deepseek-ai/dsh-session-reference` 是注册在 `ctx.sessionReferences` 上的单一上下文消费服务。宿主先把各自的协议规范化为 `SessionReferenceInput[]`,在入队前调用 `prepare()`,再通过通用的 `SendOptions.contexts` 边界传递返回的上下文。核心 agent 包只知道一条排队消息可以携带已冻结的 `HookContext[]`;它们既不解析会话 URI,也不读取其他日志。
-`dsh-session:` 是与宿主无关的规范标识符。系统先执行 JSON 字符串编码,再执行 base64url 编码,因此引号、正斜杠、反斜杠、Unicode、换行符以及其他任意 JavaScript 字符串值都能无损往返,不会因分隔符产生歧义。TUI 把该 URI 渲染到 `@[label](uri)` 中,ACP 使用标准 `resource_link`;纯文本客户端可以使用同一种行内提及标记。显式 Markdown 提及标记与资源链接会拒绝格式错误的 URI。裸文本只有在负载非空且形状符合 base64url 时才会成为引用,而且解码过程仍须通过规范性校验;空负载或只含标点符号的用法仍按普通讨论文本处理。
+`dsh-session:` 是与宿主无关的规范标识符。系统先执行 JSON 字符串编码,再执行 base64url 编码,因此引号、正斜杠、反斜杠、Unicode、换行符以及其他任意 JavaScript 字符串值都能无损往返,不会因分隔符产生歧义。TUI 把该 URI 渲染到 `@[label](uri)` 中;纯文本客户端可以使用同一种行内提及标记。显式 Markdown 提及标记会拒绝格式错误的 URI。裸文本只有在负载非空且形状符合 base64url 时才会成为引用,而且解码过程仍须通过规范性校验;空负载或只含标点符号的用法仍按普通讨论文本处理。
该服务使用 `ctx.sessionQuery.readSurface(sessionId)`:它优先从实时会话加载一次语料观察结果,使用会话包的规范表层算法执行折叠,并返回与源数据分离的会话头、捕获序号和当前节点。FTS 不是功能依赖:v1 的候选发现只按 id 和 cwd 过滤;未来的标题或正文搜索可以替换候选层,而无需改变引用标识或准备过程。
@@ -28,13 +28,13 @@ TUI 与 ACP(Agent Client Protocol)用户需要把另一场对话中的相关
`send()` 与 `steer()` 会把内容、解析后的来源和上下文一起快照为一条深度冻结、无损 JSON 的收件箱记录。合成的 `inject()` 接受来源和模型不可见的元数据,但不接受附加上下文,因为上下文属于收件箱消息。普通消息被认领后,其附带的上下文会作为 `agent/prompt-submit` 的默认附加上下文公开;提示词被阻止时,系统既不写入用户消息,也不写入上下文。waterfall(瀑布式事件)返回的 allow 结果具有最终权威性,因此监听器包装 `next()` 时会保留下游内容与上下文,除非它有意替换这些值。消息被接纳后,未指定放置方式或指定为 `separate` 时会写入独立的 `context/message`;指定为 `prompt-prefix` 时则会把上下文与最终生效的请求合并写入同一个提示词事件。排空 steering 消息时会绕过 `agent/prompt-submit`,但采用相同的放置方式分流。延迟到达的 steering 转换为排队输入时保留同一条记录;取消、dispose(资源释放)和到达终止态后的丢弃则会同时丢弃消息与上下文。`agent/queued` 会报告已冻结的上下文,使观察事件能够描述完整的保留项。
-这保留了宿主的驱动语义:TUI 在准备完成后根据 agent 状态决定调用 `send()` 还是 `steer()`,因此只有它的排队路径才会分派 UserPromptSubmit 钩子;ACP 则继续调用 `send()`,每个 `session/prompt` 调用一次。引用准备过程不是新的 steering 协议,本身也不会创建轮次。
+这保留了宿主的驱动语义:TUI 在准备完成后根据 agent 状态决定调用 `send()` 还是 `steer()`,因此只有它的排队路径才会分派 UserPromptSubmit 钩子。引用准备过程不是新的 steering 协议,本身也不会创建轮次。
## 宿主适配器
TUI 把会话候选与现有 `@` 文件提供方组合在一起。每个候选项显示最新折叠后的会话标题,没有标题时回退到 session id。候选查询遵循编辑器的取消信号;session id、cwd 和提及标签中的外部终端控制字符会被转义,但规范 URI 仍保留原始 id。TUI 只准备包含结构化提及标记的提交;等待快照时禁用重复提交;失败时恢复输入;它把提示词封套的显示内容渲染为用户消息,并把其中的会话引用元数据渲染为精简的来源列表,不在终端中暴露完整 JSON。
-ACP 先从普通提示词扁平化结果中检测直接斜杠命令,再提取 `dsh-session:` 资源链接和规范的行内提及标记,因此形如 URI 的命令参数保持不透明,同时保留普通资源链接的渲染方式。标准 `session/list` 会公开每个可加载会话折叠后的标题;挂载会话引用功能时,还会在 `_meta["deepseek-harness/sessionReference"]` 下公开规范 URI。客户端可以使用 `title ?? sessionId` 作为资源链接名称。若引用有效但可选服务未挂载,系统会返回「功能不可用」RPC 错误;准备失败会发生在占用进行中轮次槽位并调用 agent send 之前。引用准备过程单独拥有中止控制权,因此 `session/cancel` 和桥接释放都能停止待处理的读取。选择器 UI 仍由 ACP 客户端负责,因为 ACP 未定义跨会话提及菜单。
+[仅面向自动化的 ACP(Agent Client Protocol)传输层](../simplification/2026-07-23-acp-automation-only-protocol.md)有意不挂载会话查询或会话引用服务。
## 预算与保留策略
@@ -43,8 +43,8 @@ ACP 先从普通提示词扁平化结果中检测直接斜杠命令,再提取
## 考虑过的替代方案
- **等待 SQLite FTS5**:不予采纳,因为快照正确性依赖按准确 id 读取和规范表层折叠,而不是内容搜索。FTS 只改进候选发现。
-- **把提及标记语法放入 `Agent.send()`**:不予采纳,因为这会迫使核心协议解析 TUI/ACP 的表现层,并阻止带类型的非文本宿主复用同一语义层。
-- **在 TUI 和 ACP 中分别实现引用**:不予采纳,因为投影、安全警告、保留策略和持久化会在不同宿主之间逐渐偏离。
+- **把提及标记语法放入 `Agent.send()`**:不予采纳,因为这会迫使核心协议解析某个宿主的展示语法,并阻止带类型的非文本宿主复用同一语义层。
+- **在每个宿主中分别实现引用**:不予采纳,因为投影、安全警告、保留策略和持久化会在不同宿主之间逐渐偏离。
- **在提示词旁放置单独的用户角色上下文消息**:不予采纳,因为相邻的两条用户消息会削弱提示词的指示语绑定:在 `@foo what does this session discuss?` 中,模型可能把「this session」解析为当前对话,而不是被引用的快照。
- **在调用 `send()` 前由宿主合并前缀**:不予采纳,因为 `agent/prompt-submit` 必须只检查和改写直接提示词。最终生效的提示词与附加上下文只有在 AgentLoop 接纳后才汇合;此时 AgentLoop 可以把 `allow.content` 改写一致应用于合并后的模型内容和 `envelope.displayContent`。若由宿主更早组装,就会向该钩子暴露快照字节,或使这两个视图发生偏离。
- **回放原始源日志或恢复被遮蔽的事件**:不予采纳,因为压缩定义了当前模型表层,并且可能有意淘汰敏感或开销高昂的历史内容。
@@ -53,8 +53,8 @@ ACP 先从普通提示词扁平化结果中检测直接斜杠命令,再提取
## 验证
-单元与集成测试覆盖 URI 无损往返与文本边界标点、显式格式错误的引用、会考虑标题的候选排序、终端控制字符转义、投影排除规则、提示词封套的非递归投影、与后端无关的压缩检查点、标签安全封套、去重、自引用、数量限制、读取的全有或全无、存储读取不结束时取消提示词、逐源独立字节保留、冻结的消息所有权、提示词阻止、send/steer 放置方式、标题隔离、功能缺失、包含标题信息的 ACP 会话列表、普通 ACP 资源链接、不透明的 ACP 命令参数,以及精简的 TUI/ACP 回放。无密钥 TUI 快照会运行真实的 agent loop(智能体循环):源表层用一个压缩检查点替换旧的用户/assistant 历史,目标会话提交一个提及标记,捕获到的模型请求只包含一条用户消息,其中依次为快照、请求分隔符和当前提示词,并且不包含任一被遮蔽的字符串。
+单元与集成测试覆盖 URI 无损往返与文本边界标点、显式格式错误的引用、会考虑标题的候选排序、终端控制字符转义、投影排除规则、提示词封套的非递归投影、与后端无关的压缩检查点、标签安全封套、去重、自引用、数量限制、读取的全有或全无、存储读取不结束时取消提示词、逐源独立字节保留、冻结的消息所有权、提示词阻止、send/steer 放置方式、标题隔离、功能缺失和精简的 TUI 回放。无密钥 TUI 快照会运行真实的 agent loop(智能体循环):源表层用一个压缩检查点替换旧的用户/assistant 历史,目标会话提交一个提及标记,捕获到的模型请求只包含一条用户消息,其中依次为快照、请求分隔符和当前提示词,并且不包含任一被遮蔽的字符串。
## 后果
-新插件构成稳定的语义边界,不会新增持久化 schema、事件类型、FTS 依赖、源会话订阅或对压缩所遮蔽内容的访问。标准 TUI/ACP 演示组合包会显式挂载它,并在各自的配置中暴露引用数量和逐源字节上限;自定义宿主在挂载该服务并适配输入前保持不变。引用上下文会在配置的界限内增大目标历史,随后可由目标会话的普通压缩进行摘要;完成压缩后,源会话便不再相关。
+新插件构成稳定的语义边界,不会新增持久化 schema、事件类型、FTS 依赖、源会话订阅或对压缩所遮蔽内容的访问。标准 TUI 演示组合包会显式挂载它,并在自身配置中暴露引用数量和逐源字节上限;自定义宿主在挂载该服务并适配输入前保持不变。引用上下文会在配置的界限内增大目标历史,随后可由目标会话的普通压缩进行摘要;完成压缩后,源会话便不再相关。
diff --git a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.i18n.yaml
index f1b9829b73..2c0b4d3404 100644
--- a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.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
-2026-07-21-dsh-system-prompt-source-path.md: b54d01488fd7c0b49e06200c93af2b056c9fd00b
-2026-07-21-dsh-system-prompt-source-path.zh.md: 208e3dce072f63c280999e15276dce62ff4e5c43
+2026-07-21-dsh-system-prompt-source-path.md: 4cb89e8124840bba6633235d195e95957245137c
+2026-07-21-dsh-system-prompt-source-path.zh.md: 90c23bed4a3f95155e323c63a68fe2da09543ea6
diff --git a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.md b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.md
index b54d01488f..4cb89e8124 100644
--- a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.md
+++ b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.md
@@ -16,7 +16,7 @@ The testable logic lives in `dsh-app-boot`, not in `apps/cli`, because `apps/*`
## Scope
-Only the `dsh` CLI adds this. The demo bins (`dsh-tui-demo`, `dsh-acp-demo`) boot their committed trees verbatim and gain no source section: they are not the self-modification surface, and their checkout root is not a fact the model needs.
+Only the `dsh` CLI adds this. The demo bins (`dsh-cli-demo`, `dsh-acp-demo`) boot their committed trees verbatim and gain no source section: they are not the self-modification surface, and their checkout root is not a fact the model needs.
## HMR
diff --git a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.zh.md b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.zh.md
index 208e3dce07..90c23bed4a 100644
--- a/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-dsh-system-prompt-source-path.zh.md
@@ -16,7 +16,7 @@ Status: implemented
## Scope
-只有 `dsh` CLI 会加入这一段。demo bin(`dsh-tui-demo`、`dsh-acp-demo`)原样引导它们已提交的插件树,不会获得 source 段:它们不是自我修改的接口,其检出根目录也不是模型需要知道的事实。
+只有 `dsh` CLI 会加入这一段。demo bin(`dsh-cli-demo`、`dsh-acp-demo`)原样引导它们已提交的插件树,不会获得 source 段:它们不是自我修改的接口,其检出根目录也不是模型需要知道的事实。
## HMR
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml
index 929a802080..2dee4d6261 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.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
-2026-07-21-log-backed-session-titles.md: 183aa6909fecffdaf18c77c2a66fbc38c67c2d2c
-2026-07-21-log-backed-session-titles.zh.md: c6a0c2ce2ad4b2cddec2ada36f655fe55adb143b
+2026-07-21-log-backed-session-titles.md: 4cf238a278a4eec7b894a0fcfb0f2ac464325bb0
+2026-07-21-log-backed-session-titles.zh.md: 933cc14eb245581c128cb18055df726174f953ec
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
index 183aa6909f..4cf238a278 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.md
@@ -40,7 +40,7 @@ Automatic provider failures are nonfatal warnings and retain the latest title. E
A fork inherits seed title events unchanged, like the rest of its source log. The first-message provider does not automatically retitle a fork. The all-messages provider may append a child-owned revision after a later child prompt, using inherited and new eligible messages.
-`ctx.sessionQuery.readTitle()` folds one live-preferred or persisted log without loading titles during `listSessions()`. ACP maps the event to `session_info_update` during both live streaming and load replay, using the event timestamp for `updatedAt`. The TUI uses the latest title as its header subtitle and sets the terminal window title to ` — ` after terminal-safe rendering. The Web host folds the same log state into a validated mux control frame after each attached-session subscription baseline and immediately after forwarding a live raw title event. The browser retains only newer title event seqs even when the control frame precedes list or session-instance creation; sidebar labels, search, breadcrumbs, and the browser title then react to the projected revision. `session.list` remains metadata-only, so a cold persisted session uses the cwd basename or id until opening or resuming it attaches the log. The browser title uses ` — ` only for a selected titled session and otherwise preserves the product title. A synthetic title turn remains a completed durability boundary for the metadata write; consumers reporting agent completion use the core `findLastMessageTurnEnd()` fold so a later title, injection, or other plugin-owned turn cannot replace the preceding message-triggered outcome.
+`ctx.sessionQuery.readTitle()` folds one live-preferred or persisted log without loading titles during `listSessions()`. The TUI uses the latest title as its header subtitle and sets the terminal window title to ` — ` after terminal-safe rendering. The Web host folds the same log state into a validated mux control frame after each attached-session subscription baseline and immediately after forwarding a live raw title event. The browser retains only newer title event seqs even when the control frame precedes list or session-instance creation; sidebar labels, search, breadcrumbs, and the browser title then react to the projected revision. `session.list` remains metadata-only, so a cold persisted session uses the cwd basename or id until opening or resuming it attaches the log. The browser title uses ` — ` only for a selected titled session and otherwise preserves the product title. A synthetic title turn remains a completed durability boundary for the metadata write; consumers reporting agent completion use the core `findLastMessageTurnEnd()` fold so a later title, injection, or other plugin-owned turn cannot replace the preceding message-triggered outcome.
## Alternatives considered
@@ -54,7 +54,7 @@ A fork inherits seed title events unchanged, like the rest of its source log. Th
## Consequences
-- Titles survive JSONL and SQLite persistence, replay through ACP, and follow fork inheritance without a separate mutable record.
+- Titles survive JSONL and SQLite persistence, replay, and fork inheritance without a separate mutable record.
- Web title delivery stays incremental and log-backed without a title index or persisted-list scan; cold list rows improve after attach.
- A fallback appears immediately. Each fresh Web session adds one first-message auxiliary call; other compositions choose whether better titles justify model cost and whether later prompts should retitle a session.
- Auxiliary request records and late accepted titles consume event seqs and may create balanced zero-step turns, so persistence exposes both attempted dispatches and accepted updates even though model history and KV-cache identity do not change.
diff --git a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
index c6a0c2ce2a..933cc14eb2 100644
--- a/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-log-backed-session-titles.zh.md
@@ -40,7 +40,7 @@ Status: implemented
与源日志的其他部分相同,fork 会原样继承作为种子的标题事件。首消息提供方不会自动为 fork 重新生成标题。全部消息提供方可以在子会话出现后续提示词后追加一项归子会话所有的修订,并使用继承的合格消息和新增的合格消息。
-`ctx.sessionQuery.readTitle()` 会折叠一份实时优先或已持久化的日志,而不会在 `listSessions()` 期间加载标题。ACP(Agent Client Protocol)会在实时流式输出和加载回放期间把该事件映射到 `session_info_update`,并使用事件时间戳作为 `updatedAt`。TUI 使用最新标题作为其标题栏副标题,并在完成终端安全渲染后,将终端窗口标题设置为 ` — `。Web host 会在每个已附加会话的订阅基线之后,以及转发实时原始标题事件后立即,将同一份日志状态折叠为经过校验的 mux 控制帧。即使控制帧先于列表或会话实例创建抵达,浏览器也只保留标题事件 seq 较新的版本;侧边栏标签、搜索、面包屑和浏览器标题会随投影后的修订更新。`session.list` 仍只包含元数据,因此尚未打开的持久化会话会继续以 cwd 基名或 id 作为回退,直至打开或恢复会话时附加其日志。浏览器仅在选中已有标题的会话时将标题设置为 ` — `,否则保留产品标题。合成标题轮次本身仍会完成,并作为元数据写入的持久性边界;报告 agent 完成情况的消费方使用核心的 `findLastMessageTurnEnd()` 折叠逻辑,因此后续的标题轮次、注入轮次或其他归插件所有的轮次无法取代此前由消息触发的结果。
+`ctx.sessionQuery.readTitle()` 会折叠一份实时优先或已持久化的日志,而不会在 `listSessions()` 期间加载标题。TUI 使用最新标题作为其标题栏副标题,并在完成终端安全渲染后,将终端窗口标题设置为 ` — `。Web host 会在每个已附加会话的订阅基线之后,以及转发实时原始标题事件后立即,将同一份日志状态折叠为经过校验的 mux 控制帧。即使控制帧先于列表或会话实例创建抵达,浏览器也只保留标题事件 seq 较新的版本;侧边栏标签、搜索、面包屑和浏览器标题会随投影后的修订更新。`session.list` 仍只包含元数据,因此尚未打开的持久化会话会继续以 cwd 基名或 id 作为回退,直至打开或恢复会话时附加其日志。浏览器仅在选中已有标题的会话时将标题设置为 ` — `,否则保留产品标题。合成标题轮次本身仍会完成,并作为元数据写入的持久性边界;报告 agent 完成情况的消费方使用核心的 `findLastMessageTurnEnd()` 折叠逻辑,因此后续的标题轮次、注入轮次或其他归插件所有的轮次无法取代此前由消息触发的结果。
## 考虑过的替代方案
@@ -54,7 +54,7 @@ Status: implemented
## 后果
-- 标题可以在 JSONL 和 SQLite 持久化中存续,通过 ACP 回放,并遵循 fork 继承语义,而无需单独的可变记录。
+- 标题可以在 JSONL 和 SQLite 持久化中存续、重放并遵循 fork 继承语义,而无需单独的可变记录。
- Web 标题仍以增量方式从日志交付,无需标题索引或扫描持久化列表;冷启动列表项会在会话附加后改用标题。
- 回退标题会立即出现。每个新建的 Web 会话都会增加一次针对首消息的辅助调用;其他组合可以自行决定更优标题是否值得模型成本,以及后续提示词是否需要重新生成会话标题。
- 辅助请求记录和延迟接受的标题会占用事件 seq,并可能创建平衡的零步骤轮次,因此持久化会同时呈现尝试发起的调用与已接受的更新,尽管模型历史和 KV 缓存标识保持不变。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.i18n.yaml
deleted file mode 100644
index 737a9da6ca..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-21-tui-auto-pane-title.md: 069fd33a8874d9ad3d4472dd13f5130b2df65f08
-2026-07-21-tui-auto-pane-title.zh.md: 580f36b2563e21231a22cab3f0c1689c6f3e8d9d
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.md b/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.md
deleted file mode 100644
index 069fd33a88..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.md
+++ /dev/null
@@ -1,41 +0,0 @@
-# Agent Note: Auto-titled terminal from the first message
-
-Status: implemented
-
-English | [中文](2026-07-21-tui-auto-pane-title.zh.md)
-
-> **Superseded** by the [session-title consolidation Agent Note](../simplification/2026-07-22-tui-titles-from-session-title-service.md): the TUI-local `autoTitle` generation is removed; titles come from the log-backed session-title service, and the terminal rename consumes `session/title` events.
-
-> **Superseded** for the default and the resume behavior by the [auto-title default-on Agent Note](2026-07-21-tui-auto-title-default-on.md): `autoTitle` now defaults on, and a resumed session re-derives its title from the stored first message instead of keeping the static one. The OSC 0 path, the one-shot latch, the model-summary shape, the fire-and-forget call, and every failure fallback below stand.
-
-## Problem
-
-The TUI's terminal title is a single static string (`title`, default `DeepSeek Harness`) shared by every session. A user who runs one agent per tmux pane or terminal tab sees the same label on all of them, so panes are indistinguishable at a glance and the tab bar carries no signal about what each session is doing.
-
-## Decision
-
-- `TuiConfig` gains an `autoTitle` boolean (default `false`). When it is on, the TUI issues one background model call after the first user message of a fresh session and replaces the terminal title with a short, model-generated label; the static `title` is the pre-title and the fallback.
-- The label is a model summary, not a truncation of the prompt. The request carries a fixed task instruction (summarize the request as a short title of two to five lowercase words, no punctuation) plus the user's first message and no tools; the TUI takes the first non-empty line of the reply and caps it at 40 characters (39 plus an ellipsis).
-- The title is set through `runtime.terminal.setTitle`, the same OSC 0 path the static `title` already uses. No new terminal-control surface is introduced, and pi-tui keeps ownership of terminal writes.
-- The call is fire-and-forget and one-shot per session. A `titleSettled` latch guards it: with `autoTitle` off it is pre-settled and never runs; on a resumed session whose first `user/message` is already logged it is pre-settled so the static title stands; a whitespace-only first message is skipped without consuming the slot. Any failure, an empty reply, a missing `llm` service, or a missing agent provider/model leaves the static title untouched. A dedicated `AbortController` cancels an in-flight request on shutdown.
-- The title call reaches `ctx.llm.stream` directly rather than through `agent.send`, so it never appends to the session or transcript and cannot perturb the agent loop.
-- The feature defaults off and is enabled only in the interactive product config (`examples/tui-agent/cordis.yml`) and the scripted PTY fixture. Enabling it in the shared `dsh-tui-demo` schema default would fire an extra model call in keyless replay and boot scenarios that send no user message.
-
-## Alternatives considered
-
-**Truncate the first user message instead of a model title.** Rejected: the user chose a short model-made label; a truncated raw prompt is noisy, often begins with boilerplate, and rarely reads as a title.
-
-**Rename the window (OSC 2) or the tmux window.** Rejected: OSC 0 sets only `pane_title`, so it labels the pane without renaming or leaking into the user's window title; the user confirmed OSC is the right lever.
-
-**Default the feature on.** Rejected: enabling it in the shared demo schema perturbs keyless replay and boot snapshots and spends a model call on every fresh session; opt-in per deployment keeps the default surface inert.
-
-**Fold this into the log-backed session-title work (PR #451).** Rejected: that change is session metadata persisted to the log; this is a terminal label with no persistence. Keeping them independent leaves each self-contained and avoids a shared dependency.
-
-**Block the first turn until the title resolves.** Rejected: awaiting the title before sending the user's message adds latency to the actual request; fire-and-forget makes the rename invisible to the turn.
-
-## Consequences
-
-- When enabled, a fresh session spends one extra, tool-less model call with a single short user message and a few output tokens; off by default, it costs nothing.
-- Because the title call stamps `sessionId`, it shares the session's `llm-replay` cursor: enabling `autoTitle` in a replay-backed snapshot scenario would consume a recorded script entry. This is why the default is off and the scripted PTY fixture answers the call with a tool-branching adapter rather than replay.
-- `packages/ui/tui/tests/tui.spec.ts` pins the behavior with a mock `llm` adapter: a generated title replaces the static one, over-long output is truncated with an ellipsis, a whitespace-only first message keeps the one-shot slot, empty or failing replies leave the title, a resumed session never fires, and the feature-off / no-service / missing-provider / missing-model paths keep the static title. A shutdown test asserts the in-flight request is aborted.
-- `examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` proves the real Loader-booted path: the scripted adapter answers the tool-less title call with a fixed string, and the conversation scenario asserts the OSC 0 sequence reaches the PTY. Boot scenarios send no user message, so they never fire the call.
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.zh.md
deleted file mode 100644
index 580f36b256..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-pane-title.zh.md
+++ /dev/null
@@ -1,41 +0,0 @@
-# Agent Note: 从首条消息自动命名终端
-
-Status: implemented
-
-[English](2026-07-21-tui-auto-pane-title.md) | 中文
-
-> **已被取代**:见[标题归一 Agent Note](../simplification/2026-07-22-tui-titles-from-session-title-service.md)。TUI 本地的 `autoTitle` 生成已移除;标题来自日志承载的 session-title 服务,终端重命名消费 `session/title` 事件。
-
-> **已被取代**(就默认值与恢复行为而言),见[自动标题默认开启 Agent Note](2026-07-21-tui-auto-title-default-on.md):`autoTitle` 现默认开启,恢复会话会从已存储的首条消息重新推导标题,而非保留静态标题。下文的 OSC 0 路径、一次性门闩、模型概括形态、发出后不等待其返回的调用,以及每一条失败兜底,均仍然成立。
-
-## Problem
-
-TUI 的终端标题是一个所有会话共用的静态字符串(`title`,默认 `DeepSeek Harness`)。在 tmux 每个窗格或每个终端标签页各跑一个 agent(智能体)的用户看来,它们的标签全都一样,因此窗格一眼看去无从区分,标签栏也不携带任何关于各会话正在做什么的信号。
-
-## Decision
-
-- `TuiConfig` 新增布尔字段 `autoTitle`(默认 `false`)。开启后,TUI 会在全新会话的首条用户消息之后发起一次后台模型调用,并用一个简短的、模型生成的标签替换终端标题;静态 `title` 是替换前的初值,也是兜底。
-- 该标签是模型概括,而非对提示词的截断。请求携带一段固定的任务指令(将该请求概括为两到五个小写单词、不含标点的简短标题)加上用户的首条消息,且不带工具;TUI 取回复的首个非空行并截断到 40 个字符(39 个字符加一个省略号)。
-- 标题通过 `runtime.terminal.setTitle` 设置——静态 `title` 已经在用的同一条 OSC 0 路径。不引入任何新的终端控制面,终端写入仍归 pi-tui 所有。
-- 该调用发出后不等待其返回,且每会话仅一次。一个 `titleSettled` 门闩守护它:`autoTitle` 关闭时它预先置为已结算、从不运行;在首条 `user/message` 已入日志的恢复会话中它预先结算,因此静态标题得以保留;仅含空白的首条消息被跳过且不消耗名额。任何失败、空回复、缺少 `llm` 服务、或缺少 agent 的 `provider` 或 `model`,都会让静态标题保持不动。一个专用的 `AbortController` 在关闭时取消尚在进行的请求。
-- 标题调用直接抵达 `ctx.llm.stream`,而非经由 `agent.send`,因此它从不追加进会话或 transcript(文本记录),也无法扰动 agent loop(智能体循环)。
-- 该功能默认关闭,仅在交互式产品配置(`examples/tui-agent/cordis.yml`)与脚本化 PTY fixture(测试前置数据)中开启。若在共享的 `dsh-tui-demo` schema 默认值里开启,会在不发送任何用户消息的无密钥回放与启动场景中多发一次模型调用。
-
-## Alternatives considered
-
-**截断首条用户消息,而非用模型生成标题。** 否决:用户选择的是简短的、模型制作的标签;截断后的原始提示词嘈杂、常以样板文字开头,且很少读起来像标题。
-
-**重命名窗口(OSC 2)或 tmux 窗口。** 否决:OSC 0 只设置 `pane_title`,因此它标记窗格而不重命名、也不泄漏进用户的窗口标题;用户确认 OSC 是正确的手段。
-
-**让该功能默认开启。** 否决:在共享的 demo schema 里开启会扰动无密钥回放与启动快照,并在每个全新会话上花掉一次模型调用;按部署选择性开启可让默认面保持惰性。
-
-**并入日志支撑的会话标题工作(PR #451)。** 否决:那项改动是持久化到日志的会话元数据;本项是不做持久化的终端标签。让二者相互独立可使各自自成一体,并避免共享依赖。
-
-**阻塞首轮直到标题就绪。** 否决:在发送用户消息前先等待标题,会给实际请求增加延迟;发出后不等待其返回可让重命名对该轮次不可见。
-
-## Consequences
-
-- 开启时,全新会话会多花一次无工具的模型调用,只带单条简短的用户消息和少量输出 token;默认关闭时它不产生任何开销。
-- 由于标题调用会打上 `sessionId`,它与会话的 `llm-replay` 游标共享:在以回放支撑的快照场景中开启 `autoTitle` 会消耗一条录制脚本条目。这正是它默认关闭、且脚本化 PTY fixture 用按工具分支的适配器而非回放来回答该调用的原因。
-- `packages/ui/tui/tests/tui.spec.ts` 用一个 mock `llm` 适配器固定该行为:生成的标题替换静态标题、过长输出以省略号截断、仅含空白的首条消息保留一次性名额、空回复或失败回复保留标题、恢复的会话从不触发,以及功能关闭 / 无服务 / 缺提供方 / 缺模型各路径都保留静态标题。一项关闭测试断言尚在进行的请求被中止。
-- `examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` 证明真实的经 Loader 启动的路径:脚本化适配器以固定字符串回答无工具的标题调用,对话场景断言 OSC 0 序列抵达 PTY。启动场景不发送用户消息,因此它们从不触发该调用。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.i18n.yaml
deleted file mode 100644
index 830ca3e2e0..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-21-tui-auto-title-default-on.md: 35809e1ef6bade3e09c34b17608eff5f8fb5bd22
-2026-07-21-tui-auto-title-default-on.zh.md: aa20cfde1359605f2ac5a8f0427f4518c611ecd1
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.md b/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.md
deleted file mode 100644
index 35809e1ef6..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.md
+++ /dev/null
@@ -1,32 +0,0 @@
-# Agent Note: Auto-title on by default, re-derived on resume
-
-Status: implemented
-
-English | [中文](2026-07-21-tui-auto-title-default-on.zh.md)
-
-> **Superseded** by the [session-title consolidation Agent Note](../simplification/2026-07-22-tui-titles-from-session-title-service.md): the TUI-local `autoTitle` generation is removed; titles come from the log-backed session-title service, and the terminal rename consumes `session/title` events.
-
-## Problem
-
-The [auto-title Agent Note](2026-07-21-tui-auto-pane-title.md) shipped `autoTitle` off by default and, on a resumed session, kept the static title because the first `user/message` was already logged. In use both choices defeated the feature's purpose. A per-session descriptive pane title is what makes one tmux pane or terminal tab distinguishable from the next; leaving it off by default means the product ships an inert feature that almost no user turns on, and skipping re-derivation on resume means a resumed session — exactly the long-lived session most worth labelling — falls back to the shared static string. The user asked for a descriptive per-session name to be the normal experience.
-
-## Decision
-
-- `autoTitle` defaults **on** (`z.boolean().default(true)`, mirrored by `resolveTuiConfig`'s `?? true`). A deployment with an `llm` service and an agent provider/model gets a model-made pane title on every session without opting in; one without them keeps the static title, so default-on is inert where the call cannot run.
-- A **resumed** session re-derives the title on mount from its already-logged first `user/message`: `createTuiChat` scans `agent.session.events` for the first such event and feeds its text to the same one-shot `generateTitle`. The title is never persisted (the session header carries no title field), so it is always derived, never restored.
-- The one-shot latch is now simply `titleSettled = !resolved.autoTitle`. The prior pre-settle-on-resume clause is gone: on resume `generateTitle` runs once from the stored first message and then latches, so a message that arrives *after* the resume does not re-title. A fresh session has no stored `user/message` at mount, so the resume scan is a no-op and the live `session/event` listener titles the first message instead.
-- Everything else from the [auto-title Agent Note](2026-07-21-tui-auto-pane-title.md) stands unchanged: the OSC 0 `runtime.terminal.setTitle` path, the model-summary shape (two-to-five lowercase words, first non-empty line, 40-char cap), the fire-and-forget `ctx.llm.stream` call that never touches the session or transcript, the shutdown `AbortController`, and every failure fallback (empty reply, missing `llm`, missing provider/model, whitespace-only prompt).
-
-## Alternatives considered
-
-**Keep the feature off by default.** Rejected: this is a direct reversal of the [auto-title Agent Note](2026-07-21-tui-auto-pane-title.md)'s "default off" decision at the user's request. Off-by-default ships an inert feature; the descriptive name is only useful if it is the normal experience. The keyless-replay concern that motivated off-by-default is addressed by pinning `autoTitle: false` in the replay-backed snapshot scenarios rather than by suppressing it for every deployment.
-
-**Persist the derived title in the session header.** Rejected: the header has no title field and adding one would make a terminal label into session metadata — the boundary the [auto-title Agent Note](2026-07-21-tui-auto-pane-title.md) already drew against the log-backed session-title work. Re-deriving from the stored first message costs one tool-less call on resume and keeps the label a pure function of the conversation.
-
-**Re-derive on resume from the latest message instead of the first.** Rejected: the title summarises what the session is *about*, which its opening request captures; a mid-conversation message would make the pane label drift as the work moves on.
-
-## Consequences
-
-- A fresh session with a working `llm` now spends one extra tool-less model call by default (previously only when opted in); a resumed session spends one on mount. Deployments without an `llm` or provider/model are unaffected.
-- The replay-backed `examples/tui-agent/tests/tui.snapshot.ts` must opt **out**: it pins `autoTitle: false`, because a default-on title request is not among the recorded turns and `installLlmReplay` fails loud on an unrecorded request. The unit `packages/ui/tui/tests/tui.snapshot.ts` needs no opt-out — it mounts no `llm` service, so `generateTitle` short-circuits and the default flip is inert there. The interactive `examples/tui-agent/cordis.yml` and the scripted PTY fixture already set `autoTitle: true`, so the keyless smoke's OSC 0 assertion is unchanged.
-- `packages/ui/tui/tests/tui.spec.ts` pins the new defaults: the config-default test expects `autoTitle: true`; the disabled-path test now sets `autoTitle: false` explicitly; and the former "resumed session never fires" test is rewritten to assert re-derivation from the stored first message and that a later live message does not re-title. `docs/config-catalog.md` regenerates to "On by default".
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.zh.md
deleted file mode 100644
index aa20cfde13..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-auto-title-default-on.zh.md
+++ /dev/null
@@ -1,32 +0,0 @@
-# Agent Note: 自动标题默认开启,恢复时重新推导
-
-Status: implemented
-
-[English](2026-07-21-tui-auto-title-default-on.md) | 中文
-
-> **已被取代**:见[标题归一 Agent Note](../simplification/2026-07-22-tui-titles-from-session-title-service.md)。TUI 本地的 `autoTitle` 生成已移除;标题来自日志承载的 session-title 服务,终端重命名消费 `session/title` 事件。
-
-## Problem
-
-[自动标题 Agent Note](2026-07-21-tui-auto-pane-title.md) 交付时 `autoTitle` 默认关闭,并且在恢复会话中因首条 `user/message` 已入日志而保留静态标题。实际使用中这两个选择都违背了该功能的初衷。让一个 tmux 窗格或终端标签页区别于下一个的,正是每会话各异的描述性窗格标题;默认关闭意味着产品交付了一个几乎无人开启的惰性功能,而恢复时不重新推导,则意味着恢复会话——恰恰是最值得标记的长命会话——退回到共用的静态字符串。用户要求把每会话的描述性名称做成常态体验。
-
-## Decision
-
-- `autoTitle` 默认**开启**(`z.boolean().default(true)`,`resolveTuiConfig` 以 `?? true` 与之对齐)。带有 `llm` 服务与 agent 提供方/模型的部署无需选择性开启即可在每个会话获得模型制作的窗格标题;不具备它们的部署保留静态标题,因此在调用无法运行处,默认开启是惰性的。
-- **恢复**会话在挂载时从其已入日志的首条 `user/message` 重新推导标题:`createTuiChat` 在 `agent.session.events` 中扫描首个此类事件,并把其文本喂给同一个一次性的 `generateTitle`。标题从不持久化(会话头不携带标题字段),因此它始终是推导得来,而非恢复而来。
-- 一次性门闩现在只是 `titleSettled = !resolved.autoTitle`。此前"恢复即预先结算"的分句已删除:恢复时 `generateTitle` 从已存储的首条消息运行一次随后上闩,因此恢复*之后*到达的消息不会再改标题。全新会话在挂载时没有已存储的 `user/message`,因此恢复扫描是空操作,改由实时的 `session/event` 监听器为首条消息命名。
-- [自动标题 Agent Note](2026-07-21-tui-auto-pane-title.md) 的其余一切保持不变:OSC 0 的 `runtime.terminal.setTitle` 路径、模型概括形态(两到五个小写单词、首个非空行、40 字符上限)、从不触碰会话或 transcript(文本记录)的发出后不等待其返回的 `ctx.llm.stream` 调用、关闭时的 `AbortController`,以及每一条失败兜底(空回复、缺 `llm`、缺提供方/模型、仅含空白的提示词)。
-
-## Alternatives considered
-
-**让该功能保持默认关闭。** 否决:这是应用户要求,对[自动标题 Agent Note](2026-07-21-tui-auto-pane-title.md)"默认关闭"决策的直接反转。默认关闭交付的是惰性功能;只有当描述性名称成为常态体验时它才有用。当初促成默认关闭的无密钥回放顾虑,改由在以回放支撑的快照场景中固定 `autoTitle: false` 来处理,而非为每个部署都压制该功能。
-
-**把推导出的标题持久化进会话头。** 否决:会话头没有标题字段,加一个会把终端标签变成会话元数据——正是[自动标题 Agent Note](2026-07-21-tui-auto-pane-title.md)已经对日志支撑的会话标题工作划出的边界。从已存储的首条消息重新推导,代价是恢复时一次无工具调用,并让标签保持为对话的纯函数。
-
-**恢复时从最新消息而非首条消息重新推导。** 否决:标题概括的是会话*关于什么*,而这由其开场请求捕获;一条对话中途的消息会让窗格标签随工作推进而漂移。
-
-## Consequences
-
-- 带可用 `llm` 的全新会话现在默认多花一次无工具的模型调用(此前只在选择性开启时才有);恢复会话在挂载时花掉一次。不具备 `llm` 或提供方/模型的部署不受影响。
-- 以回放支撑的 `examples/tui-agent/tests/tui.snapshot.ts` 必须选择**关闭**:它固定 `autoTitle: false`,因为默认开启的标题请求不在录制轮次之列,而 `installLlmReplay` 对未录制的请求会显式报错。单元 `packages/ui/tui/tests/tui.snapshot.ts` 无需选择关闭——它不挂载 `llm` 服务,因此 `generateTitle` 提前短路,默认值的翻转在那里是惰性的。交互式的 `examples/tui-agent/cordis.yml` 与脚本化 PTY fixture(测试前置数据)已设 `autoTitle: true`,因此无密钥冒烟测试的 OSC 0 断言保持不变。
-- `packages/ui/tui/tests/tui.spec.ts` 固定新的默认值:config 默认测试期望 `autoTitle: true`;关闭路径测试现在显式设 `autoTitle: false`;此前的"恢复会话从不触发"测试改写为断言从已存储首条消息重新推导,并断言之后的实时消息不会再改标题。`docs/config-catalog.md` 重新生成为"On by default"。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.i18n.yaml
deleted file mode 100644
index a06145f092..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-21-tui-banner-sweep.md: c146424d53e75a72b63e346f87a5bbd206d67350
-2026-07-21-tui-banner-sweep.zh.md: 01cc153e88f067b7b8d2eb6317648f3892fe8a5a
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.md b/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.md
deleted file mode 100644
index c146424d53..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.md
+++ /dev/null
@@ -1,35 +0,0 @@
-# Agent Note: The banner sweeps in; the subtitle line is gone
-
-Status: implemented
-
-English | [中文](2026-07-21-tui-banner-sweep.zh.md)
-
-> **Superseded** by the [no-banner Agent Note](2026-07-21-tui-no-banner.md): the banner itself was removed, taking the sweep with it.
-
-## Problem
-
-The [startup-slogans Agent Note](2026-07-20-tui-startup-slogans.md) replaced the instructional welcome line with a random slogan bank revealed by a per-character typewriter. In use the quotes read as weird — random flavor text in a tool's header — and the animation was slow (40 ms/char over a full sentence) while animating only one line of a four-line banner. This note supersedes that decision's slogan half; the removal of the configured demo welcome and the animation-lifecycle groundwork stand.
-
-## Decision
-
-- The slogan bank, `pickStartupSlogan`, and the typewriter reveal are deleted. When `welcome` is unset the banner simply has **no subtitle line** — title and model/session detail only. The `welcome` config remains for deployments and fixtures that want a fixed subtitle, rendered frame-deterministically with no animation.
-- The startup animation is now the **whole banner**: `HeaderComponent` gains a `revealWidth` clip, and the header box wipes in left-to-right over ~24 frames at 15 ms (~360 ms total, ~60 fps), started after `ui.start()` succeeds and cleared through the same `detachListeners` path the typewriter used. `stopBannerReveal` also resets the clip so a disposed-mid-sweep header re-renders whole.
-- The PTY smoke's boot marker changes from the typewriter cursor (`▌`) to the banner's top-right corner (`╮`), which only renders once the sweep completes.
-
-## Alternatives considered
-
-**Keep the animation as-is and only change the copy.** Rejected: any fixed or rotating phrase re-read on every boot decays into wallpaper; the user's judgment was that the quotes themselves, not just their content, were wrong for the surface.
-
-**Animate per banner line (top-down) instead of a left-right sweep.** Rejected: with only four lines the animation would have four visible steps — closer to a flicker than a reveal; the horizontal sweep uses the full terminal width for a smooth motion at the same total duration.
-
-**Character-level clipping via `revealWidth` on styled text.** Adopted with `truncateToWidth` from pi-tui, the same ANSI-aware clipper the header already uses for width overflow, so the sweep cannot tear escape sequences.
-
-## Consequences
-
-- Boot output with `welcome` unset is again animation-dependent but no longer random: every boot sweeps the same banner. Configured welcomes (all snapshot/scripted fixtures, the Code Mode overlay) stay frame-deterministic and unchanged.
-- The `STARTUP_SLOGANS`/`pickStartupSlogan` exports are gone; no consumer outside the deleted tests referenced them.
-- The default banner is one line shorter (no subtitle), so PTY assertions anchored on banner geometry use the corner glyph rather than any subtitle text.
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` pins: the sweep completes to a full banner (both corners + title) and produced at least one clipped mid-sweep frame; a configured welcome renders verbatim with no clipped frames; the unset-welcome banner has no subtitle; and dispose clears the sweep's own interval handle. The PTY smoke boots on the `╮` completion marker across the tui-demo bin, the dsh CLI, and the personal-overlay scenarios. Verified live in tmux.
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.zh.md
deleted file mode 100644
index 01cc153e88..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-banner-sweep.zh.md
+++ /dev/null
@@ -1,35 +0,0 @@
-# Agent Note: 横幅整体扫入;副标题行移除
-
-Status: implemented
-
-[English](2026-07-21-tui-banner-sweep.md) | 中文
-
-> **已被取代**:由[移除启动横幅 Agent Note](2026-07-21-tui-no-banner.md)取代:横幅本身已移除,扫入动画随之移除。
-
-## Problem
-
-[启动 slogan Agent Note](2026-07-20-tui-startup-slogans.md) 用随机 slogan 库加逐字打字机动画取代了说明书式的欢迎行。实际使用中这些引语显得怪异——工具头部出现随机的风味文案——而且动画很慢(每字符 40 ms,扫完一整句),却只动画四行横幅中的一行。本 note 取代该决定中 slogan 的那一半;移除示例配置中欢迎语的决定与动画生命周期的基础设施保持不变。
-
-## Decision
-
-- 删除 slogan 库、`pickStartupSlogan` 和打字机动画。`welcome` 未设置时横幅直接**没有副标题行**——只有标题和模型/会话详情。`welcome` 配置保留给想要固定副标题的部署与 fixture,无动画、逐帧确定地渲染。
-- 启动动画现在作用于**整个横幅**:`HeaderComponent` 增加 `revealWidth` 裁剪,头部盒子以约 24 帧、每帧 15 ms(总计约 360 ms、约 60 fps)从左到右扫入,在 `ui.start()` 成功后启动,经打字机动画用过的同一条 `detachListeners` 路径清除。`stopBannerReveal` 同时重置裁剪,因此扫入中途被 dispose 的头部会重新完整渲染。
-- PTY 冒烟测试的启动标记从打字机光标(`▌`)改为横幅右上角(`╮`),它只在扫入完成后才渲染。
-
-## Alternatives considered
-
-**保留动画原样、只改文案。** 否决:任何每次启动都被重读的固定或轮换语句都会退化成墙纸;用户的判断是引语本身——而不只是内容——对这个表面来说就是错的。
-
-**按横幅行逐行(自上而下)动画而非左右扫入。** 否决:只有四行时动画只有四个可见步骤——更像闪烁而不是展开;水平扫入用满终端宽度,在相同总时长内动作更平滑。
-
-**用 `revealWidth` 对带样式文本做字符级裁剪。** 采用 pi-tui 的 `truncateToWidth`——头部处理宽度溢出时已在使用的同一个 ANSI 感知裁剪器——因此扫入不可能撕裂转义序列。
-
-## Consequences
-
-- `welcome` 未设置时启动输出再次依赖动画但不再随机:每次启动扫入同一幅横幅。配置了欢迎语的场景(全部快照/脚本化 fixture、Code Mode overlay)保持逐帧确定且不变。
-- `STARTUP_SLOGANS`/`pickStartupSlogan` 导出移除;除被删除的测试外没有消费者引用它们。
-- 默认横幅少一行(无副标题),因此锚定横幅几何的 PTY 断言使用角落字形而非任何副标题文本。
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` 固定:扫入完成为完整横幅(两个角 + 标题)且产生了至少一个裁剪的中途帧;配置的欢迎语原文渲染且无裁剪帧;未设置欢迎语的横幅没有副标题;dispose 清除扫入自己的定时器句柄。PTY 冒烟测试在 tui-demo bin、dsh CLI 和个人 overlay 场景中以 `╮` 完成标记启动。已在 tmux 中实机验证。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.i18n.yaml
index 8732101ab2..5d3ddbd972 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.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
-2026-07-21-tui-borderless-banner.md: 37263854b6cc77283215c3c1378f9908ff966611
-2026-07-21-tui-borderless-banner.zh.md: ca796e49cb9d3a9abc0acd64a39448bc3f9ad50e
+2026-07-21-tui-borderless-banner.md: 2fcb414c11f91df0914b17aa973e45746bbdfc67
+2026-07-21-tui-borderless-banner.zh.md: 8f80b21e6425bb38fff52529f1df8d262c34338f
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.md b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.md
index 37263854b6..2fcb414c11 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.md
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.md
@@ -6,34 +6,41 @@ English | [中文](2026-07-21-tui-borderless-banner.zh.md)
## Problem
-The [no-banner Agent Note](2026-07-21-tui-no-banner.md) removed the boxed startup banner: it deleted `HeaderComponent` and its sweep, moved the model into the footer, dropped the session id, and rendered `welcome` as the transcript's first line. The user's verdict reversed that: bring the banner back — "just remove the border". The four-row box frame was the objectionable chrome, not the identifying facts it carried (model, session id) nor the sweep-in motion.
+An intermediate no-banner design removed the boxed startup banner: it deleted `HeaderComponent` and its sweep, moved the model into the footer, dropped the session id, and rendered `welcome` as the transcript's first line. The user's verdict reversed that: bring the banner back — "just remove the border". The four-row box frame was the objectionable chrome, not the identifying facts it carried (model, session id) nor the sweep-in motion.
## Decision
-- `HeaderComponent` and its left-to-right sweep return, but render **borderless**: no `╭─╮`/`╰─╯` corners and no `│` side bars. Each line is a single leading space plus `truncateToWidth`-clipped content, so the sweep's width clip can never tear an escape sequence and no fixed frame is drawn.
-- The header carries the title (`DEEPSEEK HARNESS`), a ` • ` detail line, and — when `welcome` is set — a muted subtitle. With `welcome` unset the header is title + detail only.
-- The model **also** stays in the footer's left segment. The no-banner note's footer model prefix is kept, not reverted, so the driving model stays glanceable after the transient banner scrolls out of view.
+- `HeaderComponent` and its left-to-right sweep return, but render **borderless**: no `╭─╮`/`╰─╯` corners and no `│` side bars. Each line is a single leading space plus `truncateToWidth`-clipped content, so the sweep's width clip can never tear an escape sequence and no fixed frame is drawn. The reveal advances through about 24 frames at 15 ms each.
+- The header carries the title (`DEEPSEEK HARNESS`), a ` • ` detail line, and — when `welcome` is set — a muted subtitle. With `welcome` unset the header is title + detail only: there is no fixed or random slogan.
+- The model **also** stays in the footer's left segment, so the driving model remains glanceable after the transient banner scrolls out of view.
- `welcome` reverts to a banner subtitle; the transcript-first-line notice is removed from `rebuildTranscript`.
- The sweep animates only when `welcome` is unset. A configured `welcome` renders the whole banner immediately, keeping fixtures and snapshots frame-deterministic. The sweep starts after `ui.start()` succeeds and is cleared through the same `detachListeners` path via `stopBannerReveal`, which also resets the clip so a header disposed mid-sweep re-renders whole.
-This supersedes the [no-banner Agent Note](2026-07-21-tui-no-banner.md) (which superseded the [banner-sweep Agent Note](2026-07-21-tui-banner-sweep.md)): the banner and its sweep return borderless, while the model's footer home the no-banner note added stays.
+This note owns the current result of the discarded startup variants: random slogans with a per-character typewriter, a boxed whole-banner sweep, and no banner. The example composition does not set `welcome`; deployments and deterministic fixtures may still provide one. The model's persistent footer home from the no-banner variant remains.
## Alternatives considered
**Keep the box but thin it or use lighter glyphs.** Rejected: the instruction was "just remove the border"; any surrounding glyph is the frame chrome the user objected to.
-**Drop the model from the footer now that the banner shows it again.** Rejected: the banner is transient and scrolls away with the transcript, while the footer keeps the model visible for the whole session — the reason the no-banner note put it there, deliberately preserved.
+**Keep a random or fixed slogan when `welcome` is unset.** Rejected because repeated flavor copy becomes wallpaper and the per-character reveal was slow while animating only one line. An unset welcome therefore produces no subtitle, and the whole banner supplies the startup motion.
-**Leave the session id out, as the no-banner note decided.** Rejected: with the box gone the detail line costs one row, and the user asked for the banner "as before", which carried `model • session-id`.
+**Remove the banner entirely.** Rejected because the persistent footer is a good home for the model but not for the full identifying detail, while putting `welcome` in the transcript makes presentation configuration behave like conversation content.
+
+**Reveal the banner top-down.** Rejected because four row-sized steps read as a flicker. The horizontal width clip uses the terminal span for smooth motion and reuses the ANSI-aware truncation path.
+
+**Drop the model from the footer now that the banner shows it again.** Rejected: the banner is transient and scrolls away with the transcript, while the footer keeps the model visible for the whole session; that persistent location is deliberately preserved.
+
+**Leave the session id out of the banner.** Rejected: with the box gone the detail line costs one row, and the user asked for the banner "as before", which carried `model • session-id`.
## Consequences
- Boot output with `welcome` unset is animation-dependent again (the sweep); configured welcomes stay frame-deterministic, so every snapshot and scripted fixture keeps a fixed subtitle.
+- The demo no longer supplies instructional welcome filler; an unset `welcome` means a subtitle-free banner, while the config remains the deterministic escape hatch for deployments and fixtures.
- The model now appears twice at boot — banner detail and footer — intended redundancy: the banner is transient, the footer persistent.
-- `/clear` empties the transcript but not the header, so the banner and its configured subtitle survive `/clear`, unlike the no-banner welcome line that `/clear` wiped.
+- `/clear` empties the transcript but not the header, so the banner and its configured subtitle survive `/clear`, unlike a transcript-based welcome line.
- All pi-tui terminal snapshots and the examples/tui-agent replay snapshots re-recorded (`test:snapshot:refresh`): banner rows return with no box glyphs; footer rows keep the model prefix.
- Anything that anchored on banner absence re-anchors on its presence: the PTY smoke boots on the detail line's `main-session-` id (revealed late in the sweep) and asserts `DEEPSEEK`/`HARNESS` present with no box corners.
## Testing
-`packages/ui/tui/tests/tui.spec.ts` pins: the borderless banner sweeps to natural completion — no box corners, title and `main-session` detail present — with at least one clipped mid-sweep frame; a configured `welcome` renders the whole banner with no clipped frame; the unset-welcome banner has no subtitle; and dispose clears the sweep interval mid-sweep. The tui-agent and dsh-CLI PTY smokes boot on the `main-session-` detail marker and assert no box corners. Snapshots verify the full frames.
+`packages/ui/tui/tests/tui.spec.ts` pins: the borderless banner sweeps to natural completion — no box corners, title and `main-session` detail present — with at least one clipped mid-sweep frame; a configured `welcome` renders the whole banner with no clipped frame; the unset-welcome banner has no subtitle; and dispose clears the sweep interval mid-sweep. Independent color-scheme cases cover reported light/dark transitions, a same-scheme no-op, and a terminal that throws on the DSR query write; `applyColorScheme` relies on `setStatus` to rederive the editor border instead of repeating the dead assignment that had broken per-file coverage. The tui-agent and dsh-CLI PTY smokes boot on the `main-session-` detail marker and assert no box corners. Snapshots verify the full frames.
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.zh.md
index ca796e49cb..8f80b21e64 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-borderless-banner.zh.md
@@ -6,34 +6,41 @@ Status: implemented
## Problem
-[移除横幅 Agent Note](2026-07-21-tui-no-banner.md) 删掉了带框的启动横幅:它删除了 `HeaderComponent` 及其扫入动画,把模型移入页脚,丢弃了会话 id,并把 `welcome` 渲染为 transcript 的第一行。用户的裁决把这一切反转:把横幅拿回来——"just remove the border"。令人反感的装饰是那四行盒子边框,而不是它承载的识别信息(模型、会话 id),也不是扫入动效。
+一个中间的无横幅设计删掉了带框的启动横幅:它删除了 `HeaderComponent` 及其扫入动画,把模型移入页脚,丢弃了会话 id,并把 `welcome` 渲染为 transcript 的第一行。用户的裁决把这一切反转:把横幅拿回来——"just remove the border"。令人反感的装饰是那四行盒子边框,而不是它承载的识别信息(模型、会话 id),也不是扫入动效。
## Decision
-- `HeaderComponent` 及其从左到右的扫入动画回归,但以**无边框**方式渲染:没有 `╭─╮`/`╰─╯` 边角,也没有 `│` 侧边。每一行都是一个前导空格加上经 `truncateToWidth` 裁剪的内容,因此扫入的宽度裁剪永远不会撕裂转义序列,也不绘制任何固定边框。
-- 头部承载标题(`DEEPSEEK HARNESS`)、一条 ` • ` 详情行,以及——当设置了 `welcome` 时——一条弱化的副标题。`welcome` 未设置时头部只有标题加详情。
-- 模型**同时**保留在页脚的左段。移除横幅那版 note 加入的页脚模型前缀被保留而非回退,因此在短暂的横幅滚出视野后,会话使用的模型仍可一瞥可见。
+- `HeaderComponent` 及其从左到右的扫入动画回归,但以**无边框**方式渲染:没有 `╭─╮`/`╰─╯` 边角,也没有 `│` 侧边。每一行都是一个前导空格加上经 `truncateToWidth` 裁剪的内容,因此扫入的宽度裁剪永远不会撕裂转义序列,也不绘制任何固定边框。扫入大约经过 24 帧完成,每帧间隔 15 ms。
+- 头部承载标题(`DEEPSEEK HARNESS`)、一条 ` • ` 详情行,以及——当设置了 `welcome` 时——一条弱化的副标题。`welcome` 未设置时头部只有标题加详情:不含固定或随机标语。
+- 模型**同时**保留在页脚的左段,因此在短暂的横幅滚出视野后,会话使用的模型仍可一瞥可见。
- `welcome` 恢复为横幅副标题;transcript 第一行的通知从 `rebuildTranscript` 中移除。
- 仅当 `welcome` 未设置时才播放扫入动画。配置了 `welcome` 会立即渲染整个横幅,使 fixture 和快照保持帧确定性。扫入在 `ui.start()` 成功后启动,并经与之前相同的 `detachListeners` 路径通过 `stopBannerReveal` 清理;后者还会重置裁剪,使扫入中途被销毁的头部重新完整渲染。
-本 note 取代[移除横幅 Agent Note](2026-07-21-tui-no-banner.md)(后者取代了[横幅扫入 Agent Note](2026-07-21-tui-banner-sweep.md)):横幅及其扫入动画以无边框方式回归,而移除横幅那版 note 为模型设立的页脚归宿得以保留。
+本 Agent Note 统一记录几种已弃用启动方案的当前结论:带逐字打字机效果的随机标语、带边框的整幅横幅扫入动画,以及完全移除横幅。示例组装不设置 `welcome`;部署和确定性 fixture 仍可提供该值。无横幅方案为模型设置的常驻页脚位置继续保留。
## Alternatives considered
**保留盒子但做细或改用更轻的字符。** 否决:指令是 "just remove the border";任何环绕的字符都是用户所反对的边框装饰。
-**既然横幅重新显示模型,就把模型从页脚移除。** 否决:横幅是短暂的,会随 transcript 滚走,而页脚在整个会话中保持模型可见——这正是移除横幅那版 note 把它放在那里的原因,此处刻意保留。
+**在未设置 `welcome` 时保留随机或固定标语。** 否决:反复出现的氛围文案很快失去信息价值,而逐字揭示仅为一行制作动画,速度又慢。因此,未设置 `welcome` 时不显示副标题,由整个横幅提供启动动效。
-**像移除横幅那版 note 那样,把会话 id 留在外面。** 否决:盒子去掉后详情行只占一行,且用户要求横幅"和以前一样",而以前它承载 `model • session-id`。
+**完全移除横幅。** 否决:常驻页脚很适合显示模型,却无法承载完整识别详情;把 `welcome` 放入 transcript 还会使展示配置表现成对话内容。
+
+**自上而下揭示横幅。** 否决:按四行分成四步看起来像闪烁。横向宽度裁剪利用终端横向空间实现平滑动效,并复用 ANSI 感知的截断路径。
+
+**既然横幅重新显示模型,就把模型从页脚移除。** 否决:横幅是短暂的,会随 transcript 滚走,而页脚在整个会话中保持模型可见;这个常驻位置被刻意保留。
+
+**将会话 id 留在横幅之外。** 否决:盒子去掉后详情行只占一行,且用户要求横幅"和以前一样",而以前它承载 `model • session-id`。
## Consequences
- `welcome` 未设置时的启动输出再次依赖动画(扫入);配置了欢迎语则保持帧确定性,因此每个快照和脚本 fixture 都保留一个固定副标题。
+- demo 不再提供教学性质的欢迎填充文案;`welcome` 未设置就表示横幅没有副标题,而该配置仍是部署和 fixture 获得确定性输出的配置手段。
- 模型现在在启动时出现两次——横幅详情与页脚——这是有意的冗余:横幅短暂,页脚常驻。
-- `/clear` 清空 transcript 但不清头部,因此横幅及其配置的副标题在 `/clear` 后存活,不同于被 `/clear` 清掉的移除横幅那版的欢迎行。
+- `/clear` 清空 transcript 但不清头部,因此横幅及其配置的副标题在 `/clear` 后存活,不同于基于 transcript 的欢迎行。
- 全部 pi-tui 终端快照与 examples/tui-agent 回放快照重新录制(`test:snapshot:refresh`):横幅行以无盒子字符方式回归;页脚行保留模型前缀。
- 一切锚定横幅缺失的内容改为锚定其存在:PTY 冒烟测试以详情行的 `main-session-` id 为启动标记(它在扫入后段才被揭示),并断言 `DEEPSEEK`/`HARNESS` 出现且无盒子角。
## Testing
-`packages/ui/tui/tests/tui.spec.ts` 固定:无边框横幅扫入至自然完成——无盒子角、标题与 `main-session` 详情出现——且至少有一帧扫入中途被裁剪;配置的 `welcome` 完整渲染横幅且无裁剪帧;未设置 `welcome` 的横幅无副标题;销毁会在扫入中途清掉扫入定时器。tui-agent 与 dsh CLI 的 PTY 冒烟测试以 `main-session-` 详情标记为启动标记并断言无盒子角。快照验证完整帧。
+`packages/ui/tui/tests/tui.spec.ts` 固定:无边框横幅扫入至自然完成——无盒子角、标题与 `main-session` 详情出现——且至少有一帧扫入中途被裁剪;配置的 `welcome` 完整渲染横幅且无裁剪帧;未设置 `welcome` 的横幅无副标题;销毁会在扫入中途清掉扫入定时器。独立的配色方案用例覆盖终端报告的浅色/深色转换、相同方案下的空操作,以及写入 DSR 查询时抛出异常的终端;`applyColorScheme` 依靠 `setStatus` 重新推导编辑器边框,而不再重复那个导致逐文件覆盖率未达标的无效赋值。tui-agent 与 dsh CLI 的 PTY 冒烟测试以 `main-session-` 详情标记为启动标记并断言无盒子角。快照验证完整帧。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.i18n.yaml
deleted file mode 100644
index 56333563f5..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-21-tui-no-banner.md: f5f4b1b847740e741ec3e33a6116e7497e955bd1
-2026-07-21-tui-no-banner.zh.md: 956fe03e2c0b09ea7378ffd53ffbe8d712d1e152
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.md b/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.md
deleted file mode 100644
index f5f4b1b847..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.md
+++ /dev/null
@@ -1,39 +0,0 @@
-# Agent Note: No startup banner
-
-Status: implemented
-
-English | [中文](2026-07-21-tui-no-banner.zh.md)
-
-> **Superseded** by the [borderless-banner Agent Note](2026-07-21-tui-borderless-banner.md): the banner and its sweep return without the box. The model's footer home this note added stays.
-
-## Problem
-
-The TUI opened with a boxed product banner ("DEEPSEEK HARNESS" + model/session detail), most recently with a sweep-in animation ([banner sweep Agent Note](2026-07-21-tui-banner-sweep.md)). The user's verdict: remove it. A product title re-read on every boot is chrome, the box spends four rows before any content, and the identifying facts it carried (model, session) have better homes.
-
-## Decision
-
-- `HeaderComponent`, the sweep animation, and its lifecycle wiring are deleted. The TUI mounts straight into the transcript; startup renders nothing above the separator.
-- The model name moves into the footer status line's left segment (` ↑tokens ↓tokens`), so the session's driving model stays visible at all times, not just at boot. The session id is no longer displayed — it lives in the session log and `./.sessions` filenames, and `RESUME_SESSION_ID` consumers retrieve it there.
-- `welcome`, when configured, renders as the transcript's first line (a muted notice) inside `rebuildTranscript`, so palette swaps preserve it. Unset renders nothing. Fixtures keep their configured welcomes; the PTY smoke's boot marker becomes the footer's model name, the only mounted-TUI text guaranteed to render regardless of cwd length.
-
-This supersedes the [banner sweep Agent Note](2026-07-21-tui-banner-sweep.md) entirely: both the sweep and the banner it animated are gone.
-
-## Alternatives considered
-
-**Keep a one-line header (no box).** Rejected: the only load-bearing fact was the model name, and the footer already aggregates session status; a dedicated header row for one fact is the same chrome, smaller.
-
-**Show the session id in the footer too.** Rejected: a 36-char UUID dominates the 100-column footer and clips the status segment; it identifies the session for resume, which is a log/filesystem concern, not a glanceable one.
-
-**Print the welcome outside the transcript (above the separator).** Rejected: any fixed region above the transcript is a banner again; as a transcript line it scrolls away naturally and survives rebuilds through the same path as every other transcript element.
-
-## Consequences
-
-- Startup output is fully deterministic again — no animation frames at all; the interval-lifecycle machinery from the two animation iterations is gone.
-- All 26 pi-tui terminal snapshots re-recorded (`test:snapshot:refresh`): banner rows gone, footer rows gain the model prefix.
-- Anything that anchored on banner text (`DEEPSEEK`, box corners) re-anchors on the footer model name; `main-session-` no longer appears in boot output.
-- `/clear` now wipes the welcome line too: it is an ordinary transcript line, and `/clear` empties the transcript (the old banner survived `/clear` only by sitting outside it).
-- The footer's left segment is wider; on narrow terminals the right status segment clips earlier.
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` pins: no box corners/product title and an empty transcript when `welcome` is unset, with the model in the footer; a configured welcome as the first transcript line without a banner; and the welcome surviving a palette-swap transcript rebuild. The PTY smoke boots on the footer model name and asserts `DEEPSEEK HARNESS` is absent. Snapshots verify the full frames.
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.zh.md
deleted file mode 100644
index 956fe03e2c..0000000000
--- a/.agents/notes/implemented/feature/2026-07-21-tui-no-banner.zh.md
+++ /dev/null
@@ -1,39 +0,0 @@
-# Agent Note: 移除启动横幅
-
-Status: implemented
-
-[English](2026-07-21-tui-no-banner.md) | 中文
-
-> **已被取代**,见[无边框横幅 Agent Note](2026-07-21-tui-borderless-banner.md):横幅及其扫入动画回归,只是去掉了盒子。本 note 为模型设立的页脚归宿得以保留。
-
-## Problem
-
-TUI 启动时展示一个带框的产品横幅("DEEPSEEK HARNESS" + 模型/会话详情),最近一版还带扫入动画([横幅扫入 Agent Note](2026-07-21-tui-banner-sweep.md))。用户的裁决:删掉它。每次启动都被重读的产品标题是装饰,盒子在任何内容之前先占掉四行,而它承载的识别信息(模型、会话)有更好的去处。
-
-## Decision
-
-- 删除 `HeaderComponent`、扫入动画及其生命周期接线。TUI 直接挂载进 transcript;启动时分隔线之上不渲染任何东西。
-- 模型名移入页脚状态行的左段(` ↑tokens ↓tokens`),会话使用的模型因此始终可见,而不只是启动时。会话 id 不再显示——它存在于会话日志和 `./.sessions` 文件名中,`RESUME_SESSION_ID` 的使用者从那里获取。
-- 配置了 `welcome` 时,它作为 transcript 的第一行(一条弱化的通知)在 `rebuildTranscript` 内渲染,因此调色板切换会保留它。未设置则什么也不渲染。fixture 保留各自配置的欢迎语;PTY 冒烟测试的启动标记改为页脚的模型名——无论 cwd 多长都保证渲染的唯一挂载后文本。
-
-本 note 完全取代[横幅扫入 Agent Note](2026-07-21-tui-banner-sweep.md):扫入动画和它所动画的横幅都已移除。
-
-## Alternatives considered
-
-**保留单行头部(去掉盒子)。** 否决:唯一有承载价值的信息是模型名,而页脚已经聚合会话状态;为一条信息保留专用头部行仍是同一种装饰,只是小一点。
-
-**把会话 id 也放进页脚。** 否决:36 字符的 UUID 会占满 100 列页脚并裁掉状态段;它的用途是恢复会话的标识,属于日志/文件系统关注点,不是需要一瞥可见的信息。
-
-**把欢迎语渲染在 transcript 之外(分隔线上方)。** 否决:transcript 上方任何固定区域都会再次变成横幅;作为 transcript 行它自然滚走,并通过与其他 transcript 元素相同的路径在重建后保留。
-
-## Consequences
-
-- 启动输出再次完全确定——没有任何动画帧;两轮动画迭代留下的定时器生命周期机制全部移除。
-- 全部 26 个 pi-tui 终端快照重新录制(`test:snapshot:refresh`):横幅行消失,页脚行增加模型前缀。
-- 锚定横幅文本(`DEEPSEEK`、盒子角)的内容改为锚定页脚模型名;启动输出中不再出现 `main-session-`。
-- `/clear` 现在也会清掉欢迎行:它是普通的 transcript 行,而 `/clear` 清空 transcript(旧横幅能在 `/clear` 后存活只因为它在 transcript 之外)。
-- 页脚左段变宽;窄终端上右侧状态段更早被裁剪。
-
-## Testing
-
-`packages/ui/tui/tests/tui.spec.ts` 固定:`welcome` 未设置时无盒子角/产品标题、transcript 为空、模型在页脚;配置的欢迎语作为 transcript 第一行且无横幅;欢迎语在调色板切换的 transcript 重建后保留。PTY 冒烟测试以页脚模型名为启动标记并断言 `DEEPSEEK HARNESS` 不出现。快照验证完整帧。
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.i18n.yaml b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.i18n.yaml
index 319f28ea61..8cac245f00 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.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
-2026-07-21-tui-verbose-status-line.md: f277afd3a874b30a29dc0ef193740f636d22290b
-2026-07-21-tui-verbose-status-line.zh.md: 9fa7cf29c67245382bbee6b72f2710c5550d7f54
+2026-07-21-tui-verbose-status-line.md: 71584ee91a911cc8652512ec26b00dae8c818f36
+2026-07-21-tui-verbose-status-line.zh.md: bda3c5e8394f7707916c6fc76045b1a6f38fa95b
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.md b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.md
index f277afd3a8..71584ee91a 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.md
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.md
@@ -13,7 +13,7 @@ While a turn ran, the [full-screen TUI](2026-07-17-dedicated-full-screen-tui-fro
- While a turn runs, the status line above the editor shows a derived phase label with elapsed time, keeping the trailing `— Enter sends steering, Esc cancels` hint. The four phases and their labels are `waiting` → "Waiting for the first token", `thinking` → "Thinking", `responding` → "Responding", and `executing` → "Executing tools".
- The phase is presentation state the TUI derives from live session events, not a session event or agent status of its own. `step/start` enters `waiting`; an `assistant/chunk` reasoning delta or reasoning block-start enters `thinking`; a text delta or text block-start enters `responding`; a `tool/call` enters `executing`. The event map is merge-extensible, so every other event kind falls through a default and leaves the phase unchanged.
- The label reports two clocks — ` · total ` — except `waiting`, which shows only the step total. The phase clock resets on a genuine phase change or a new step; the step clock resets on `step/start`. Durations format as `8s` below a minute and `1m05s` at or above one. Tool time between `step/end` and the next `step/start` accrues to the finishing step's total.
-- A single `RunningStatus` controller — the loader, the phase, the two baselines, and a refresh timer — exists only while a turn runs. A one-second `setInterval` refreshes the elapsed time; a phase event refreshes it immediately. `clearStatus` clears the interval, stops the loader, and drops the controller, so any transition to idle or disposed leaves no live timer, matching the [banner sweep](2026-07-21-tui-banner-sweep.md)'s timer hygiene. A mid-turn palette rebuild (`setStatus` re-derives the editor border on a terminal color-scheme change) carries the phase and both baselines across, so a running status never snaps back to `waiting`.
+- A single `RunningStatus` controller — the loader, the phase, the two baselines, and a refresh timer — exists only while a turn runs. A one-second `setInterval` refreshes the elapsed time; a phase event refreshes it immediately. `clearStatus` clears the interval, stops the loader, and drops the controller, so any transition to idle or disposed leaves no live timer, matching the [borderless banner](2026-07-21-tui-borderless-banner.md)'s timer hygiene. A mid-turn palette rebuild (`setStatus` re-derives the editor border on a terminal color-scheme change) carries the phase and both baselines across, so a running status never snaps back to `waiting`.
## Alternatives considered
diff --git a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.zh.md b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.zh.md
index 9fa7cf29c6..bda3c5e839 100644
--- a/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.zh.md
+++ b/.agents/notes/implemented/feature/2026-07-21-tui-verbose-status-line.zh.md
@@ -13,7 +13,7 @@ Status: implemented
- 轮次运行期间,编辑器上方的状态行显示一个派生的阶段标签及已用时长,并保留末尾的 `— Enter sends steering, Esc cancels` 提示。四个阶段及其标签为 `waiting` → "Waiting for the first token"、`thinking` → "Thinking"、`responding` → "Responding"、`executing` → "Executing tools"。
- 阶段是 TUI 从实时会话事件派生出的呈现状态,而非它自有的会话事件或 agent 状态。`step/start` 进入 `waiting`;`assistant/chunk` 的 reasoning 分片或 reasoning 块开始(`block-start`)进入 `thinking`;text 分片或 text 块开始进入 `responding`;`tool/call` 进入 `executing`。该事件映射可合并扩展,因此其余任何事件类型都落入默认分支,保持阶段不变。
- 标签汇报两个时钟——` · total `——但 `waiting` 只显示步骤总时长。阶段时钟在真正发生阶段切换或进入新步骤时重置;步骤时钟在 `step/start` 时重置。时长在不足一分钟时格式化为 `8s`,达到或超过一分钟时格式化为 `1m05s`。`step/end` 与下一个 `step/start` 之间的工具时间计入结束步骤的总时长。
-- 单一的 `RunningStatus` 控制器——loader、阶段、两个基准时刻以及一个刷新定时器——仅在轮次运行期间存在。一个每秒触发的 `setInterval` 刷新已用时长;阶段事件则立即刷新。`clearStatus` 清除该 interval、停止 loader 并丢弃控制器,因此任何向 idle 或 disposed 的转变都不会遗留活动定时器,与 [banner 扫入动画](2026-07-21-tui-banner-sweep.md)的定时器清理保持一致。轮次进行中的调色板重建(终端颜色方案变化时 `setStatus` 会重新派生编辑器边框)会将阶段与两个基准时刻一并沿用过来,因此运行中的状态绝不会退回 `waiting`。
+- 单一的 `RunningStatus` 控制器——loader、阶段、两个基准时刻以及一个刷新定时器——仅在轮次运行期间存在。一个每秒触发的 `setInterval` 刷新已用时长;阶段事件则立即刷新。`clearStatus` 清除该 interval、停止 loader 并丢弃控制器,因此任何向 idle 或 disposed 的转变都不会遗留活动定时器,与[无边框横幅](2026-07-21-tui-borderless-banner.md)的定时器清理保持一致。轮次进行中的调色板重建(终端颜色方案变化时 `setStatus` 会重新派生编辑器边框)会将阶段与两个基准时刻一并沿用过来,因此运行中的状态绝不会退回 `waiting`。
## 曾考虑的替代方案
diff --git a/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.i18n.yaml b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.i18n.yaml
new file mode 100644
index 0000000000..9087db9dd7
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+# pnpm run verify-translation-pairing --write
+2026-07-24-model-facing-session-query-tools.md: 2a9a20a8b39dea309e759f4eb6ddcdabe25dd8be
+2026-07-24-model-facing-session-query-tools.zh.md: 7fbe746681b329b3e50ae74608a8b2e5167c5ae9
diff --git a/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.md b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.md
new file mode 100644
index 0000000000..2a9a20a8b3
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.md
@@ -0,0 +1,55 @@
+# Agent Note: Model-facing session query tools
+
+Status: implemented
+
+English | [中文](2026-07-24-model-facing-session-query-tools.zh.md)
+
+## Problem
+
+The unified `ctx.sessionQuery` service exposes exact reads, filters, relationship traces, and full-text search over live-preferred session logs, but models cannot use that service directly. Giving a model the provider request types would also expose unstable pagination cursors, trusted corpus scope, storage-shaped time values, and result records that are more convenient for programmatic consumers than for reasoning. Large traces and event payloads introduce a separate output-size concern, but solving that concern inside this consumer would duplicate the harness-wide spill mechanism and make session-query tools disagree with other tools.
+
+## Decision
+
+`@deepseek-ai/dsh-tool-session-query` is the model-facing consumer of `ctx.sessionQuery`. It registers five narrow read-only tools: `session_search`, `session_event_search`, `session_trace`, `session_event_trace`, and `session_event_read`. The package imports the interface rather than the SQLite implementation, owns model argument validation and readable text rendering, and contributes one concise prompt section that teaches the prior-history search and search-to-trace/read workflow.
+
+The package entrypoint is only the public composition root for configuration, prompt registration, and tool registration. Its internal modules follow the execution boundary: `input.ts` owns model schemas, normalization, and filter construction; `service-boundary.ts` contains provider calls and model-safe error translation; `workspace-access.ts` owns caller identity, workspace authorization, title access, and lineage projection; `operations.ts` orchestrates the five service workflows; and `presentation.ts` renders tool results and call cards. This keeps policy in its owning layer without changing the package contract.
+
+`session_search` groups full-text matches by session and exposes typed session and event metadata filters. `session_event_search` searches one session, defaulting to the caller's current session. `session_trace` returns the complete authorized ancestor chain and recursive descendant trees. `session_event_trace` returns every known positional replacement and direct provenance relationship for one event. `session_event_read` returns the exact target event as unabridged JSON and optionally summarizes a bounded raw-event window; omitted `before` and `after` values mean target-only.
+
+Model-facing filters use flat snake-case fields. Timestamps are timezone-qualified ISO 8601 strings at the tool boundary, convert to inclusive epoch-millisecond ranges for the service, and render as UTC ISO 8601. List values are ORed inside one filter while separate filters are ANDed. Requested parent ids are deduplicated and authority-filtered before FTS, so only parents in the caller workspace enter the provider clause; missing and cross-workspace guesses behave identically, while the root-session marker remains independently ORed into that clause. Event type strings remain open because `SessionEventMap` is merge-extensible; availability and event surface use closed values.
+
+## Workspace authority
+
+Every executor derives its caller from immutable `ToolExecution.exec.agent` identity and never accepts a model-supplied workspace. A target is authorized only when its observed `cwd` exactly equals the caller session's `cwd`. Cross-session search always adds that workspace filter. Direct operations preflight the target and then validate the header returned from the same service observation as every event-search page, event trace, event read, lineage target, or folded title before rendering its payload. This prevents a live or persisted target replacement between the check and use from crossing the workspace boundary. Lineage rendering stops at an unauthorized ancestor or descendant subtree without revealing the hidden session id. A caller whose session has no `cwd` can inspect only its own session; missing agent identity fails closed.
+
+Every trusted `ctx.sessionQuery` call crosses one model-boundary sanitizer. It checks the execution signal first, preserving caller cancellation exactly. For other failures it records the available corpus or provider diagnostic chain in the internal log on a best-effort basis, substituting a fixed placeholder when the value cannot be safely inspected. Diagnostic formatting and error classification are independently guarded, so an unprintable nested cause cannot escape or prevent a safely classified outer error, while unsafe classification or logging returns the fixed generic `SESSION_QUERY_TOOL_FAILED` code and message. Per-title failures use the same sanitizer before becoming unavailable markers. Tool-owned input-validation and authorization errors remain precise because they are created outside this service boundary.
+
+The search tools expose prior work rather than the operation that is performing the search. `session_search` omits the caller's session. When `session_event_search` targets the caller's session, it intersects the requested sequence range with the event immediately before the current `step/start`, excluding the current assistant message and tool call as well as the query arguments indexed from that call.
+
+## Cursor-free results and spill
+
+Neither search tool exposes a cursor, offset, page size, or model-controlled result limit. One execution follows provider cursors while the observed generation remains valid and collects up to the configured `maxSearchResults`, which defaults to 100. A capped result tells the model to narrow its query or filters; a generation change reports that the whole search must be retried. Search execution carries a configurable `searchTimeoutMs`, defaulting to 30 seconds, through the tool deadline and the service abort signal. Because internal pages share generation-bound cursors, both search tools are exclusive in the agent-loop scheduler; the exact trace and read tools opt into parallel sibling execution because their observations tolerate intervening commits.
+
+Trace and read tools likewise expose no lineage or character pagination. Canonical results are plain text and remain complete within the service's existing event-window and search-count resource bounds. Each exact executor passes its unchanged tool-execution signal through target authorization and the service trace or read. Within service resolution, known-live event traces, event reads, and title reads remain persistence-free while honoring pre-abort. Session lineage tracing passes the signal to whole-corpus persistence listing; persisted event tracing and reading pass it to target listing and inspection. Each started backend call is awaited for cleanup before the exact abort reason is preserved, even when that backend ignored cancellation. The generic `tools/post-execute` spill policy owns inline byte retention: when a configured deployment receives oversized text, it replaces that text with a bounded preview plus an opaque locator and retrieval hint while preserving the complete result in its spill store. The session-query consumer neither imports `ctx.spillStore` nor implements a second truncation format.
+
+Session-level results include the latest folded title when available. Each tool execution batches its unique title ids through one live-preferred corpus observation with at most the service's configured `persistedInspectConcurrency` workers, which defaults to four, and passes the exact tool-execution signal through persisted listing and inspection. Live sources fold directly; each persisted worker folds its completed source to a detached header/title observation and releases the full log before dequeuing another id, so the batch retains only small projected values. For the search tools, the execution signal carries the configured search deadline. Cancellation starts no queued title inspections and rejects the complete tool execution after already-started inspections settle; a missing, malformed, or operationally failed title remains isolated to that id, preserves the base result, renders an unavailable marker, and logs the underlying error, while an authorization mismatch fails closed. Search results include the strongest matching event and provider excerpt, traces include complete authorized relationships, and event reads keep neighbor presentation readable while reserving exact JSON for the requested target.
+
+## Host composition
+
+The consumer is an opt-in plugin. The shipped ACP and TUI apps mount `ctx.sessionQuery` for non-model consumers, while the shared Web/headless composition mounts neither the query service nor the consumer. No shipped composition mounts `@deepseek-ai/dsh-tool-session-query`, so default model requests gain no query prompt or schemas. A composition that opts in also chooses whether to mount the generic timeout and spill policies; the dedicated ACP snapshot fixture mounts both and uses private local spill storage. Generic tool presentation requires no session-query-specific client plugin.
+
+## Alternatives considered
+
+- **Expose provider cursors to the model** — rejected because recording a tool result or starting the next model step changes the relevant session or global generation, so a cursor is usually stale before the model can reuse it.
+- **Add tool-local truncation, offsets, or spill files** — rejected because the post-execute spill policy already owns complete-result retention and retrieval across tools.
+- **Allow every persisted session or model-supplied workspace filters** — rejected because `ctx.sessionQuery` is a trusted service and the model-facing consumer must enforce the caller's authority boundary.
+- **Combine search, tracing, and exact reads into one operation selector** — rejected because narrow names give the model clearer schemas, defaults, presentation intents, and follow-up choices.
+- **Return only one lineage hop** — rejected because spill removes the inline-size motivation while one-hop output would omit relationships with no continuation path.
+
+## Verification
+
+Package tests pin argument validation, filter translation, timestamp normalization, exact-workspace authorization, parent-filter preauthorization and oracle resistance, changed-observation rejection, service-diagnostic redaction for ordinary and adversarial unknown values, best-effort cyclic-cause logging, logger-failure containment, missing-identity behavior, hidden-boundary pruning, current-step exclusion, internal provider paging, exclusive search and parallel exact-read classification, count caps, exact-signal forwarding, abort-reason preservation, persistence cleanup quiescence, one-scan bounded batch title enrichment, projection-before-dequeue ordering, queued-work suppression, started-worker quiescence, per-header validation, title fallbacks, representative search/trace/read rendering, generic presentation, and disposable registration. Integration coverage uses the real SQLite FTS provider over live and persisted sessions. Shipped configuration, app composition tests, and assembled ACP request-header snapshots prove that the model-facing consumer remains absent while `ctx.sessionQuery` stays available where mounted. A package-owned Loader smoke and dedicated keyless ACP snapshot explicitly mount the consumer with timeout and spill support, pinning its prompt guidance, schemas, and path-independent exact event-read retention behavior.
+
+## Consequences
+
+Models gain provider-independent access to prior session work without receiving storage authority or continuation state. Search has a finite per-call work bound and may require a narrower query to reach matches beyond the first 100; search calls cannot overlap siblings, while exact observations retain parallel scheduling. Complete traces and event payloads may become spill references instead of inline text. Exact string `cwd` equality favors a conservative security boundary over resolving symlink-equivalent paths. Custom compositions may mount the tool without spill, but then they explicitly accept complete inline trace and read results.
diff --git a/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.zh.md b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.zh.md
new file mode 100644
index 0000000000..7fbe746681
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-07-24-model-facing-session-query-tools.zh.md
@@ -0,0 +1,55 @@
+# Agent Note: 面向模型的会话查询工具
+
+Status: implemented
+
+[English](2026-07-24-model-facing-session-query-tools.md) | 中文
+
+## 问题
+
+统一的 `ctx.sessionQuery` 服务对优先使用实时数据的会话日志提供精确读取、过滤、关系追踪与全文搜索,但模型无法直接使用该服务。若把提供方请求类型交给模型,还会暴露不稳定的分页游标、受信任的语料范围、存储形态的时间值,以及更适合程序化消费者而非模型推理的结果记录。大型追踪与事件负载另有输出大小问题,但若在该消费者内部解决,就会重复 harness 的通用 spill 机制,并使会话查询工具与其他工具的行为不一致。
+
+## 决策
+
+`@deepseek-ai/dsh-tool-session-query` 是 `ctx.sessionQuery` 面向模型的消费者。它注册五个职责单一的只读工具:`session_search`、`session_event_search`、`session_trace`、`session_event_trace` 和 `session_event_read`。该包依赖接口而非 SQLite 实现,负责模型参数校验与易读文本渲染,并贡献一个精简的提示词段,说明历史搜索以及从搜索转向追踪/读取的工作流。
+
+该包入口仅作为配置、提示词注册与工具注册的公开组合根。内部模块沿执行边界划分:`input.ts` 负责模型 schema、规范化与过滤条件构造;`service-boundary.ts` 包含提供方调用与面向模型的安全错误转换;`workspace-access.ts` 负责调用者身份、工作区授权、标题访问与谱系投影;`operations.ts` 编排五个服务工作流;`presentation.ts` 渲染工具结果与调用卡片。这样可让策略留在其所属层,同时不改变包契约。
+
+`session_search` 按会话聚合全文匹配,并公开带类型的会话与事件元数据过滤条件。`session_event_search` 搜索一个会话,默认目标为调用者的当前会话。`session_trace` 返回完整的已授权祖先链与递归后代树。`session_event_trace` 返回一个事件所有已知的位置替换关系与直接来源关系。`session_event_read` 以未删节 JSON 返回准确的目标事件,并可选择汇总一个有界的原始事件窗口;省略 `before` 与 `after` 时只返回目标。
+
+面向模型的过滤条件使用扁平的 snake-case 字段。工具边界上的时间戳采用带时区的 ISO 8601 字符串,转换为服务使用的闭区间毫秒时间戳,并以 UTC ISO 8601 渲染。同一个过滤条件中的列表值按 OR 组合,不同过滤条件按 AND 组合。请求的父会话 id 会在 FTS 之前去重并按权限过滤,因此只有调用者工作区中的父会话会进入提供方条件;缺失与跨工作区的猜测具有相同行为,而根会话标记仍会独立按 OR 加入该条件。由于 `SessionEventMap` 可通过声明合并扩展,事件类型字符串保持开放;可用状态与事件表层使用封闭取值。
+
+## 工作区权限
+
+每个执行器都从不可变的 `ToolExecution.exec.agent` 身份推导调用者,绝不接受模型提供的工作区。只有当目标观测中的 `cwd` 与调用者会话的 `cwd` 完全相同时,目标才获授权。跨会话搜索始终附加该工作区过滤条件。直接操作先预检目标,然后在渲染负载前,校验与每一页事件搜索结果、事件追踪、事件读取、谱系目标或折叠标题来自同一服务观测的会话头。这样,即使实时或持久化目标在检查与使用之间被替换,也无法跨越工作区边界。谱系渲染在遇到未授权的祖先或后代子树时停止,且不泄露被隐藏的会话 id。调用者会话没有 `cwd` 时只能检查自身会话;缺少 agent 身份时按失败关闭处理。
+
+每个受信任的 `ctx.sessionQuery` 调用都会经过同一个模型边界净化器。它首先检查执行信号,准确保留调用者取消。对于其他失败,它会尽力把可获得的语料或提供方诊断链写入内部日志;当值无法安全检查时,则改用固定占位符。诊断格式化与错误分类各自受到保护,因此无法打印的嵌套 cause 既不会逃逸,也不会阻止对外层错误进行安全分类;分类不安全或日志记录失败时,则返回固定的通用错误码 `SESSION_QUERY_TOOL_FAILED` 及其消息。逐标题失败也会先经过同一个净化器,再转为不可用标记。工具自身的输入校验与授权错误在该服务边界之外创建,因此仍保留精确消息。
+
+搜索工具公开的是既往工作,而不是正在执行搜索的操作本身。`session_search` 排除调用者会话。`session_event_search` 以调用者会话为目标时,会把请求的序号范围与当前 `step/start` 之前的最后一个事件取交集,从而排除当前 assistant 消息、工具调用,以及从该次调用中建立索引的查询参数。
+
+## 无游标结果与 spill
+
+两个搜索工具都不向模型公开游标、偏移量、页大小或模型可控的结果限制。一次执行会在观察到的代保持有效时持续跟随提供方游标,并收集不超过配置项 `maxSearchResults` 的结果,其默认值为 100。达到上限的结果会要求模型缩小查询或过滤范围;代发生变化时会报告必须重试完整搜索。搜索执行通过工具截止时间与服务中止信号传递可配置的 `searchTimeoutMs`,默认值为 30 秒。由于内部页面共享与代绑定的游标,两个搜索工具在 agent loop 调度器中都以独占方式执行;精确追踪与读取工具则允许和兄弟工具并行执行,因为其观测可以容忍期间发生的提交。
+
+追踪与读取工具同样不公开谱系分页或字符分页。规范结果采用纯文本,并在服务已有的事件窗口与搜索数量资源边界内保持完整。每个精确执行器都会将未经替换的工具执行信号传递给目标授权与服务追踪或读取。在服务解析过程中,已知实时事件追踪、事件读取与标题读取在遵循预中止的同时仍不访问持久化。会话谱系追踪会将该信号传递给全语料持久化列表;持久化事件追踪与读取则将其传递给目标列表和检查。每个已启动的后端调用都会等待清理完成后再保留准确的中止原因,即使该后端忽略了取消也不例外。通用的 `tools/post-execute` spill 策略负责行内字节保留:当已配置的部署收到过大的文本时,该策略会用有界预览、不可透明推导的定位符与读取提示替换文本,同时在 spill 存储中保留完整结果。会话查询消费者既不导入 `ctx.spillStore`,也不实现第二套截断格式。
+
+会话级结果在可用时包含最新折叠标题。每次工具执行都会通过一次优先使用实时数据的语料观测批量读取唯一标题 id,最多使用服务通过 `persistedInspectConcurrency` 配置的持久化检查 worker,其默认值为 4,并将准确的工具执行信号传递给持久化列表与检查操作。实时来源会直接折叠;每个持久化 worker 都会把已完成的来源折叠为分离的会话头/标题观测,并在取出下一个 id 前释放完整日志,因此批次只保留小型投影值。对于搜索工具,该执行信号携带已配置的搜索截止时间。取消不会启动排队中的标题检查,并会在已经启动的检查全部完成后拒绝完整的工具执行;标题缺失、格式错误或发生操作性失败时,错误只影响对应 id,同时保留基础结果、渲染不可用标记并记录底层错误,而授权不匹配则按失败关闭处理。搜索结果包含最强匹配事件与提供方摘录,追踪包含完整的已授权关系,事件读取保持邻近事件表现易读,同时只为被请求的目标保留精确 JSON。
+
+## 宿主组合
+
+该消费方是一个需显式启用的插件。发布的 ACP 与 TUI 应用为非模型消费方挂载 `ctx.sessionQuery`,而 Web/headless 共享组合既不挂载查询服务,也不挂载该消费方。发布的组合均未挂载 `@deepseek-ai/dsh-tool-session-query`,因此默认模型请求中不包含查询提示词或 schema。选择启用该插件的组合还要决定是否挂载通用的超时与 spill 策略;专用的 ACP 快照 fixture(测试前置数据)同时挂载这两项策略,并使用私有的本地 spill 存储。通用工具表现无需会话查询专用客户端插件。
+
+## 考虑过的替代方案
+
+- **向模型公开提供方游标**:不予采纳,因为记录工具结果或开始下一个模型步骤会改变相关会话或全局代,导致游标通常在模型能够复用前就已过期。
+- **增加工具本地截断、偏移量或 spill 文件**:不予采纳,因为执行后 spill 策略已经统一负责各工具的完整结果保留与读取。
+- **允许访问所有持久化会话或由模型提供工作区过滤条件**:不予采纳,因为 `ctx.sessionQuery` 是受信任服务,面向模型的消费者必须执行调用者权限边界。
+- **把搜索、追踪与精确读取合并为一个带操作选择器的工具**:不予采纳,因为职责单一的名称能为模型提供更清晰的 schema、默认值、表现意图与后续选择。
+- **只返回一层谱系**:不予采纳,因为 spill 已消除行内大小方面的理由,而单层输出会遗漏关系且没有继续读取路径。
+
+## 验证
+
+包级测试固定参数校验、过滤条件转换、时间戳规范化、精确工作区授权、父级过滤预授权与抵御预言机探测、变更观测拒绝、普通值与对抗性未知值的服务诊断脱敏、尽力记录循环 cause、日志失败隔离、身份缺失行为、隐藏边界裁剪、当前步骤排除、内部提供方翻页、搜索独占与精确读取并行分类、数量上限、精确信号传递、中止原因保留、持久化清理静止、单次扫描且并发有界的批量标题扩充、先投影再取出下一个任务的顺序、抑制排队工作、等待已启动 worker 静止、逐会话头校验、标题回退、代表性搜索/追踪/读取渲染、通用表现与可释放注册。集成覆盖使用真实 SQLite FTS 提供方查询实时与持久化会话。发布配置、应用组合测试与组装后的 ACP 请求头快照证明:面向模型的消费方仍未挂载,而 `ctx.sessionQuery` 在已经挂载该服务的组合中保持可用。包自身的 Loader 冒烟测试与专用无密钥 ACP 快照显式挂载该消费方,并配套启用超时与 spill 支持,固定其提示词指引、schema 以及与路径无关的精确事件读取保留行为。
+
+## 后果
+
+模型无需获得存储权限或继续状态,即可通过与提供方无关的方式访问既往会话工作。搜索具有有限的单次调用工作边界,若要命中前 100 条以后的结果,可能需要缩小查询;搜索调用不能与兄弟工具重叠执行,而精确观测仍可并行调度。完整追踪与事件负载可能表现为 spill 引用而不是行内文本。严格的 `cwd` 字符串相等选择了保守安全边界,而不解析通过符号链接等价的路径。自定义组合可以在不挂载 spill 的情况下使用该工具,但这表示它们明确接受完整追踪与读取结果直接出现在行内。
diff --git a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.i18n.yaml b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.i18n.yaml
similarity index 63%
rename from .agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.i18n.yaml
rename to .agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.i18n.yaml
index 91d675acea..3295a845f3 100644
--- a/.agents/notes/implemented/feature/2026-06-14-acp-agent-client-protocol.i18n.yaml
+++ b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.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
-2026-06-14-acp-agent-client-protocol.md: 61e2fd609b9f2762b50ef0fa76173259506c1470
-2026-06-14-acp-agent-client-protocol.zh.md: 608f445432913e3a332f5e8320a4154768217af4
+2026-07-25-workspace-ui-product-flow.md: a02087235a36f2c257de407facf2dc02ed072f3b
+2026-07-25-workspace-ui-product-flow.zh.md: 8ccbf5b98401bef9c3fd40e948d35ec5f0818202
diff --git a/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.md b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.md
new file mode 100644
index 0000000000..a02087235a
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.md
@@ -0,0 +1,117 @@
+# Agent Note: Workspace UI Complete Product Flow
+
+Status: implemented
+
+English | [中文](2026-07-25-workspace-ui-product-flow.zh.md)
+
+## Problem
+
+[Domain KV Storage and the Workspace Entity](../../proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.md) defines the persistent Workspace entity, path conventions, and ordered Session ledger, but not the Host wiring, historical-data initialization, or GUI flow. The GUI presents both Workspaces and Sessions; users must be able to type immediately after entering New Session, even when no Host Session or Host Workspace exists yet.
+
+Pending Workspaces, pending Sessions, retained input, and Host entity publication need clear owners and must preserve the same page identity when RPC completions and Host frames arrive in either order. Eagerly creating a Host Session for the zero state would bring a page with no input into the Host lifecycle. Historical Sessions also expose only the lightweight `SessionHeader.cwd` for grouping; initialization cannot read event bodies.
+
+## Decision
+
+### Host and persistent data
+
+The Host provides the following GUI wiring on the Workspace entity:
+
+| RPC | Behavior |
+| --- | --- |
+| `workspace.list` | Returns persistent Workspaces in order and filters out Session ids that fail header validation |
+| `workspace.create({ name })` | Creates a directory and Workspace at `workspaceRoot/name`; fails on a display-name conflict |
+| `workspace.create({ path })` | Adopts an existing directory and does not create an arbitrary path |
+| `session.create({ workspaceId, sessionId? })` | Resolves cwd from the Workspace, idempotently creates a Session with an optional preallocated id, and attaches it |
+| `session.create({ cwd })` | Remains available to non-Workspace callers and creates an Ungrouped Session |
+
+`workspaceRoot` is an independent Host setting that falls back to the Host cwd when unset; it is unrelated to `storageRoot`, which stores Workspace domain data. The Host stream pushes Workspace and Session deltas, and the Client refreshes the `workspace.list` and `session.list` baselines separately after reconnecting.
+
+A Workspace's `sessionIds` is an ordered candidate index. A membership projection requires both that an id appear in the index and that the corresponding canonicalized `SessionHeader.cwd` equal the Workspace path; SessionHeader does not gain a `workspaceId`. A Session whose cwd matches but whose id is absent from the index remains Ungrouped, while an indexed id is filtered out if its header is missing, its cwd is invalid, or its cwd does not match. Two Workspace indexes claiming the same Session is corrupt state and fails loudly.
+
+The Workspace domain uses a durable marker to distinguish “never initialized” from “initialized but empty.” When the marker is absent, the Registry calls only `SessionPersistence.list()` to read header metadata; it calls neither `load` nor `inspect`, reads no history, and parses no event bodies. Valid cwd values are grouped by canonical path, and both Sessions within each group and the Workspace groups themselves are initialized in descending header `createdAt` order. Bootstrap is reentrant and writes the marker last; after the marker is written, new Sessions created without `workspaceId` are no longer adopted automatically.
+
+### Client object model
+
+`Session` and `Workspace` are frontend objects from the page Intent stage onward.
+
+- A frontend Session preallocates a SessionId when created and owns its Intent target and `pendingPrompt`; it remains the same Session object after Host `session.create` succeeds.
+- Before materialization, a frontend Workspace has no WorkspaceId and owns its create input, phase, and error; after Host `workspace.create` succeeds, the same Workspace object adopts the returned view.
+- `SessionManager` and `WorkspaceManager` own object indexes and merge Host baselines and deltas; the objects are the sole source of state for both Intents and Host views.
+- `SessionsService` provides Session objects, real selection, scope, and list projections; `WorkspacesService` depends on `SessionsService` and owns the default Workspace, cross-object New Session flow, and Workspace materialization.
+
+A page has at most one frontend Session Intent and one accompanying Workspace Intent that exists only in the zero-Workspace state. Intents exist only on the current page and disappear on refresh; real Session selection can be restored persistently. Selecting a real Session or starting another Session Intent revokes the old Intent's eligibility for automatic sending, but does not roll back a Session already published by the Host or any accepted message.
+
+The Session owns the first input and drives one internal pipeline: when necessary, it attaches to a Workspace with its preallocated id, then sends `pendingPrompt`. Both attach and send failures return to the same Session. Workspace creation phase and error belong only to the Workspace object; the Session does not simulate the Workspace lifecycle.
+
+### User flow
+
+On initial entry, the application waits until both the Workspace and Session baselines are ready. It restores a real Session selection that remains valid; otherwise, it enters New Session and selects the most recent Workspace exactly once. The most recent Workspace is determined by the maximum `updatedAt` of its member Sessions, falling back to `createdAt` for an empty Workspace. This derived value chooses only the default target: it does not alter the Host Workspace order or trigger another selection after later hydration.
+
+When no Workspace exists, the page creates a frontend Workspace object named `workspace` and a frontend Session that targets it. Neither writes to the Host, and the composer always accepts input; the first send materializes the Workspace, attaches the Session, and sends the message in that order.
+
+Top-level New Session, the plus button on a Workspace row, and the Workspace picker all invoke the same New Session action. An explicit Workspace id becomes the target directly; when none is specified, the action uses the most recent Workspace, or the Workspace Intent if no real Workspace exists. The Workspace picker's Use an existing folder and Create a new workspace actions immediately create a real Workspace when the user confirms, then retarget the frontend Session to it; an explicitly created empty Workspace remains even if the user sends no message.
+
+Create a new workspace temporarily uses the same input as both the directory name and display name. The UI prevents duplicate confirmation based on current Workspace titles, while the Host continues to reject same-name requests that bypass the UI or race concurrently. Rename, Delete, moving across Workspaces, drag-and-drop ordering, manual adoption from Ungrouped, and separate display-name and directory-name inputs are outside this iteration's scope.
+
+### First send and recovery
+
+A frontend Session's `pendingPrompt` retains its original text until the Host accepts the message. The first send advances through Workspace materialization, Session attachment, and prompt sending in order:
+
+1. If Workspace creation fails, the Workspace Intent retains its input and error, and the Session continues to target that object.
+2. If Session creation fails before publication, the Session Intent returns to an editable state and retries with the same preallocated SessionId.
+3. `workspace-attach-failed` proves that the Session has been published; the same Session object enters the real list and retains the prompt, and subsequent retries attach it.
+4. If the prompt fails, the Session retains it and retries only send without recreating the Workspace or Session.
+5. If the page switches to another Intent while a Session is being created, the old Session does not send automatically even if it is subsequently published; it retains its original prompt and visible error.
+
+Lost RPC responses, Host frames arriving before completions, and completions arriving before Host frames all converge through the preallocated SessionId and object identity. The Manager performs ordered upserts of Host views and prioritizes preserving the original object identity during local materialization, rather than creating a temporary second row with the same id.
+
+### Sidebar and ordering
+
+Workspace groups strictly follow the persistent order returned by the Host. Bootstrap determines the historical order once, explicitly created Workspaces are placed first, and Session activity does not move Workspace groups.
+
+Within each group, order strictly follows `Workspace.sessionIds`. A newly attached Session is placed first; when a Session later becomes active, the Host moves only that id to the front and persists the change. The Client does not reorder the entire group by time after the Session list arrives, so it never displays one Workspace order and then jumps to another during hydration.
+
+A frontend Session Intent appears as a “New session” row and temporarily counts toward the group's Session total only when it targets a real Workspace. When it targets a Workspace Intent, neither the Workspace nor the Session appears in the sidebar. After the Intent is published, the real row with the same preallocated id takes its place; after refresh, both the Intent row and temporary count disappear. Search mode neither retains nor filters Intent rows.
+
+Real Sessions that cannot be assigned to any Workspace appear under Ungrouped. Host `session-added` and `workspace-changed` events may arrive in either order; list merging does not depend on frame order.
+
+### React and slot boundaries
+
+React components only consume `useSessions`, `useWorkspaces`, and session-scoped hooks; they do not own entity lifecycles. The Zustand store retains only layout, the current view, composer text for ordinary real Sessions, and other purely presentational state. Session and Workspace Intents, materialization phases, errors, and retained prompts reside in the React-free runtime object layer.
+
+The Sidebar and conversation empty hero receive standardized actions through slots: `startSession`, `updateSessionPrompt`, `sendSession`, `open`, and `toggleSidebar`. The Workspace picker reuses the same component and the `createWorkspace` seam; its owner supplies only popover state, an anchor, and a selection callback. The presentation layer does not send `host/workspace-changed` directly; Host events originate only from Host mutations and the stream adapter.
+
+## Alternatives considered
+
+**Store separate page records for pending Workspaces and Sessions.** This approach must replace identities after materialization and hand off input, errors, focus, and sidebar rows; Intent state owned by the objects preserves identity continuity.
+
+**Let the presentation layer or root Zustand store orchestrate object lifecycles.** This approach duplicates Manager and Service responsibilities and brings domain state back into React. Runtime services provide standardized actions, while slots inject only the narrow interfaces required by presentation.
+
+**Immediately create a Host Session or Host persistence intent in the zero state.** A page with no input would enter the Host lifecycle and change refresh semantics; before the first send, the frontend Session retains only a page-local Intent.
+
+**Delay an explicit Create Workspace until the first send.** After confirmation, the sidebar would still show no real empty Workspace, conflating “create a Workspace” with “prepare a Session”; only the zero-Workspace Intent generated automatically by the system delays materialization.
+
+**Continuously derive Workspaces dynamically from cwd.** This cannot represent empty Workspaces, stable display names, or explicit ordering, and would automatically adopt non-Workspace callers; cwd is used only for one historical bootstrap and bidirectional membership validation.
+
+**Have the Client batch-reorder by time after the Session list arrives.** The initial screen would first show the Host order and then jump as a whole, and reconnecting could change positions again; the Host's persistent ledger owns ordering, while the Client merges only individual updates.
+
+**Add workspaceId to SessionHeader.** This would create two persistent ownership fields alongside the Workspace index and require double writes; the header retains the Session's own cwd fact, while the Workspace index owns explicit membership.
+
+## Verification
+
+- The zero state with no Workspace writes nothing to the Host and accepts input; explicit Create Workspace immediately creates and displays an empty Workspace.
+- Frontend Sessions and Workspaces preserve object identity across materialization; input, errors, focus, and sidebar projections always originate from the object layer.
+- The first send advances through Workspace, Session, and prompt in order; successful stages are not rolled back, input is not lost before the prompt is accepted, and creation retries use the same SessionId.
+- Workspace list performs one reentrant bootstrap using only headers; an initialized empty registry does not initialize again after restart, and membership reads validate both the index and canonical cwd.
+- The initial default target is determined exactly once after both baselines are ready; Workspace groups are not reordered as a whole by hydration or Session activity, and an active Session moves only itself to the front.
+- A frontend Session under a real Workspace temporarily counts toward the sidebar total, while a Workspace Intent remains hidden; neither publication nor refresh leaves duplicate rows or counts.
+- Both the UI and Host reject duplicate Workspace names; cwd-only Sessions, Sessions with invalid historical cwd values, and unattached Sessions remain Ungrouped.
+- Keyless runnable snapshots cover the zero state, explicit creation, and the first send; package-level tests cover bootstrap, membership validation, ordering, idempotency, failure recovery, and arbitrary frame order.
+
+## Consequences
+
+- SessionHeader does not record last-active time, so historical bootstrap can initialize order only by `createdAt`; real Session activity events move individual entries afterward.
+- Historical Sessions with a missing cwd, an invalid directory, or a failed realpath remain Ungrouped; this iteration has no manual-adoption entry point.
+- Refreshing the page discards unmaterialized Workspace and Session Intents and input not yet accepted by the Host; this is the page-local contract.
+- Explicit Create Workspace writes to disk immediately, so leaving without sending still leaves an empty Workspace.
+- Before its first event, a Host Session retains the existing lazy-persistence semantics; frontend Intents do not change empty-Session behavior after a Host restart.
diff --git a/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.zh.md b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.zh.md
new file mode 100644
index 0000000000..8ccbf5b984
--- /dev/null
+++ b/.agents/notes/implemented/feature/2026-07-25-workspace-ui-product-flow.zh.md
@@ -0,0 +1,117 @@
+# Agent Note: Workspace UI 完整产品动线
+
+[English](2026-07-25-workspace-ui-product-flow.md) | 中文
+
+Status: implemented
+
+## Problem
+
+[Domain KV storage 与 Workspace entity](../../proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.md)定义了 Workspace 的持久实体、路径规范和有序 Session 账本,但没有定义 Host 接线、历史数据初始化或 GUI 动线。GUI 同时呈现 Workspace 和 Session;用户进入 New Session 后必须立即输入,即使此时还没有 Host Session,甚至没有 Host Workspace。
+
+待创建 Workspace、待创建 Session、输入保留与 Host 实体发布必须具有明确所有者,并在 RPC completion 与 Host frame 以任意顺序到达时保持同一页面身份。若零态提前创建 Host Session,则无输入的页面状态会进入 Host 生命周期。历史 Session 又只有轻量 `SessionHeader.cwd` 可用于归组,初始化不能读取事件正文。
+
+## Decision
+
+### Host 与持久数据
+
+Host 在 Workspace entity 上提供以下 GUI 接线:
+
+| RPC | 行为 |
+| --- | --- |
+| `workspace.list` | 返回持久有序的 Workspace,并过滤未通过 header 校验的 Session id |
+| `workspace.create({ name })` | 在 `workspaceRoot/name` 创建目录和 Workspace;显示名冲突时失败 |
+| `workspace.create({ path })` | 收编已经存在的目录,不为任意路径创建目录 |
+| `session.create({ workspaceId, sessionId? })` | 从 Workspace 解析 cwd,以可选预分配 id 幂等创建 Session 并 attach |
+| `session.create({ cwd })` | 保留给非 Workspace 调用方,创建 Ungrouped Session |
+
+`workspaceRoot` 是独立 Host 配置,未配置时回退到 Host cwd;它与保存 Workspace domain 数据的 `storageRoot` 无关。Host stream 推送 Workspace 与 Session 增量,Client 重连后分别刷新 `workspace.list` 与 `session.list` 基线。
+
+Workspace 的 `sessionIds` 是有序候选索引。成员投影同时要求 id 位于索引且对应 `SessionHeader.cwd` canonical 后等于 Workspace path;SessionHeader 不增加 `workspaceId`。cwd 匹配但未入索引的 Session 保持 Ungrouped,索引命中但 header 缺失、cwd 无效或 cwd 不匹配的 id 被过滤。同一 Session 被两个 Workspace 索引占用属于损坏状态并 fail loud。
+
+Workspace domain 以 durable marker 区分“从未初始化”和“已初始化但为空”。marker 未设置时,Registry 只调用 `SessionPersistence.list()` 读取 header 元数据,不调用 `load`、`inspect`、history 或解析事件正文;有效 cwd 按 canonical path 分组,组内 Session 与 Workspace 组均按 header `createdAt` 降序初始化。Bootstrap 可重入,最后才写 marker;marker 写入后,绕过 `workspaceId` 的新 Session 不再被自动收编。
+
+### Client 对象模型
+
+`Session` 与 `Workspace` 从页面 Intent 阶段开始就是前端对象。
+
+- 前端 Session 创建时预分配 SessionId,并在对象内持有 Intent target 与 `pendingPrompt`;Host `session.create` 成功后仍是同一个 Session 对象。
+- 前端 Workspace 在 materialize 前没有 WorkspaceId,并在对象内持有 create input、phase 与 error;Host `workspace.create` 成功后同一个 Workspace 对象 adopt 返回的 view。
+- `SessionManager` 与 `WorkspaceManager` 负责对象索引、Host 基线和增量合并;对象是 Intent 与 Host view 的唯一状态源。
+- `SessionsService` 提供 Session 对象、真实 selection、scope 与列表投影;`WorkspacesService` 依赖 `SessionsService`,负责默认 Workspace、跨对象 New Session 动线和 Workspace materialize。
+
+页面至多有一个前端 Session Intent 和一个仅在零 Workspace 状态下配套的 Workspace Intent。Intent 只存在于当前页面,刷新后消失;真实 Session selection 可以持久恢复。选择真实 Session 或启动另一个 Session Intent 会放弃旧 Intent 的自动发送资格,但已经由 Host 发布的 Session 和已经接受的消息不会回滚。
+
+Session 自己持有首条输入并驱动一条内部流水线:必要时以预分配 id attach 到 Workspace,然后发送 `pendingPrompt`。attach 与 send 的失败都落回同一 Session。Workspace 创建 phase/error 只属于 Workspace 对象,Session 不模拟 Workspace 生命周期。
+
+### 用户动线
+
+应用首次进入时等待 Workspace 与 Session 两份基线 ready。仍有效的真实 Session selection 被恢复;否则进入 New Session,并固定选择一次最近 Workspace。最近 Workspace 取其成员 Session 的最大 `updatedAt`,空 Workspace 回退到 `createdAt`;该派生只决定默认目标,不改变 Host Workspace 顺序,也不会在后续 hydration 时二次改选。
+
+完全没有 Workspace 时,页面创建默认名为 `workspace` 的前端 Workspace 对象和指向它的前端 Session。两者不写 Host,composer 始终可输入;首次发送才依次 materialize Workspace、attach Session、发送消息。
+
+顶部 New Session、Workspace 行内加号和 Workspace picker 最终都调用同一 New Session 动作:显式 Workspace id 直接成为目标,未指定时使用最近 Workspace,没有真实 Workspace 时使用 Workspace Intent。Workspace picker 的 Use an existing folder 与 Create a new workspace 会在用户确认时立即创建真实 Workspace,再把前端 Session 定位到该 Workspace;即使用户不发送消息,显式创建的空 Workspace 也保留。
+
+Create a new workspace 暂时用同一个输入作为目录名和显示名。UI 根据当前 Workspace title 禁止重复确认,Host 继续拒绝绕过 UI 或并发产生的同名请求。Rename、Delete、跨 Workspace 移动、拖拽排序、Ungrouped 手动收编和显示名/目录名双输入不在本期范围。
+
+### 首次发送与恢复
+
+前端 Session 的 `pendingPrompt` 在 Host 接受消息前始终保留原文。首次发送按 Workspace materialize、Session attach、prompt send 顺序推进:
+
+1. Workspace 创建失败时,Workspace Intent 保留输入与错误,Session 仍指向该对象。
+2. Session 创建在发布前失败时,Session Intent 回到可编辑状态,以同一预分配 SessionId 重试。
+3. `workspace-attach-failed` 证明 Session 已发布;同一 Session 对象进入真实列表并保留 prompt,后续重试 attach。
+4. prompt 失败时,Session 保留 prompt 并只重试 send,不重复创建 Workspace 或 Session。
+5. Session 创建期间若页面切换到另一个 Intent,旧 Session 即使随后发布也不自动发送;它保留原 prompt 和可见错误。
+
+RPC lost response、Host frame 先于 completion 和 completion 先于 Host frame 都通过预分配 SessionId 与对象身份收敛。Manager 对 Host view 做有序 upsert,本地 materialize 时优先保留原对象身份,不生成同 id 的临时第二行。
+
+### Sidebar 与排序
+
+Workspace 组严格使用 Host 返回的持久顺序。Bootstrap 一次性确定历史顺序,显式创建的新 Workspace 放在首位;Session 活跃不会移动 Workspace 组。
+
+组内严格使用 `Workspace.sessionIds`。新 attach 的 Session 放在首位,后续某个 Session 活跃时 Host 只前移该 id 并持久化。Client 不在 Session list 到达后按时间整体重排,因此不会先显示一套 Workspace 顺序再因 hydration 瞬间跳动。
+
+前端 Session Intent 只有在目标是真实 Workspace 时才作为 “New session” 行显示,并临时计入该组 Session 数量;目标是 Workspace Intent 时,Workspace 与 Session 都不进入 sidebar。Intent 发布后由同一预分配 id 对应的真实行接替,刷新后 Intent 行和临时计数一起消失。搜索模式不保存或筛选 Intent 行。
+
+无法归入任何 Workspace 的真实 Session 进入 Ungrouped。Host `session-added` 与 `workspace-changed` 可以任意顺序到达,列表合并不依赖 frame 顺序。
+
+### React 与 slot 边界
+
+React 组件只消费 `useSessions`、`useWorkspaces` 与 session-scoped hooks,不拥有实体生命周期。Zustand store 只保留布局、当前 view、普通真实 Session 的 composer 文本和其他纯呈现状态;Session/Workspace Intent、materialize phase、错误和 retained prompt 位于 React-free runtime 对象层。
+
+Sidebar 与 conversation empty hero 通过 slot 获得标准化动作:`startSession`、`updateSessionPrompt`、`sendSession`、`open` 与 `toggleSidebar`。Workspace picker 复用同一组件与 `createWorkspace` seam;owner 只提供 popover 开关、锚点和选中回调。呈现层不直接发送 `host/workspace-changed`,Host event 只由 Host mutation 与 stream adapter 产生。
+
+## Alternatives considered
+
+**为待创建 Workspace 与 Session 保存独立页面记录。** 该方案在 materialize 后需要替换身份并转交输入、错误、焦点和 sidebar 行;对象自身的 Intent 状态可以保持身份连续。
+
+**由呈现层或 root Zustand store 编排对象生命周期。** 该方案会重复 Manager/Service 的职责,并把领域状态带回 React。标准化动作由 runtime service 提供,slot 只注入呈现所需的窄接口。
+
+**零态立即创建 Host Session 或 Host persistence intent。** 未输入页面会进入 Host 生命周期,并改变刷新语义;前端 Session 在首次发送前只保留 page-local Intent。
+
+**显式 Create Workspace 延迟到首次发送。** 用户确认后 sidebar 仍看不到真实空 Workspace,“创建 Workspace”与“准备 Session”语义混合;只有系统自动产生的零 Workspace Intent 延迟 materialize。
+
+**持续按 cwd 动态派生 Workspace。** 该方案无法表达空 Workspace、稳定显示名和显式顺序,也会自动收编非 Workspace 调用方;cwd 只用于一次历史 bootstrap 与成员双向校验。
+
+**Client 在 Session list 到达后按时间批量重排。** 首屏会先展示 Host 顺序再整体跳动,重连也可能改变位置;排序由 Host 持久账本拥有,Client 只合并单项更新。
+
+**在 SessionHeader 增加 workspaceId。** 它会与 Workspace 索引形成两个持久归属字段并要求双写;header 保留 Session 自身 cwd 事实,Workspace 索引负责显式归属。
+
+## Verification
+
+- 完全无 Workspace 的零态不写 Host 且允许输入;显式 Create Workspace 立即创建并显示空 Workspace。
+- 前端 Session 与 Workspace 在 materialize 前后保持对象身份,输入、错误、焦点和 sidebar 投影始终来自对象层。
+- 首发按 Workspace、Session、prompt 顺序推进,各成功阶段不回滚,输入在 prompt 接受前不丢失,创建重试使用同一 SessionId。
+- Workspace list 只读取 header 完成一次可重入 bootstrap;initialized 的空 registry 重启不重复初始化,成员读取同时校验索引与 canonical cwd。
+- 初始默认目标只在两份基线 ready 后确定一次;Workspace 组不因 hydration 或 Session 活跃整体重排,单个活跃 Session 只前移自身。
+- 真实 Workspace 下的前端 Session 临时计入 sidebar 数量,Workspace Intent 保持隐藏,发布与刷新都不会留下重复行或重复计数。
+- UI 与 Host 两层拒绝同名 Workspace;cwd-only Session、无效历史 cwd 和未 attach Session 保持 Ungrouped。
+- keyless runnable snapshot 覆盖零态、显式创建和首次发送;包级测试覆盖 bootstrap、成员校验、排序、幂等、失败恢复及任意 frame 顺序。
+
+## Consequences
+
+- SessionHeader 不记录最后活跃时间,历史 bootstrap 只能按 `createdAt` 初始化;此后由真实 Session 活跃事件逐项前移。
+- 历史 cwd 缺失、目录无效或 realpath 失败的 Session 留在 Ungrouped;本期没有手动收编入口。
+- 页面刷新会丢弃未 materialize 的 Workspace/Session Intent 和尚未被 Host 接受的输入,这是 page-local 契约。
+- 显式 Create Workspace 立即落盘,用户不发送就离开也会留下空 Workspace。
+- Host Session 在首个事件前仍遵循现有懒持久化语义;前端 Intent 不改变 Host 重启后的空 Session 行为。
diff --git a/.agents/notes/implemented/process/2026-06-17-ts-build-config.i18n.yaml b/.agents/notes/implemented/process/2026-06-17-ts-build-config.i18n.yaml
index 80f7838d45..d8530ac919 100644
--- a/.agents/notes/implemented/process/2026-06-17-ts-build-config.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-06-17-ts-build-config.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
-2026-06-17-ts-build-config.md: 1275a635242ea887c941db9dc0554fd33acd4102
-2026-06-17-ts-build-config.zh.md: 7691118dd152874e2849f1ace686069c9ea3f61f
+2026-06-17-ts-build-config.md: 17036438b83a77f72b49f55abf29632af3f4ffef
+2026-06-17-ts-build-config.zh.md: 70e49c61deba418894a48be3016898d1d85c78a0
diff --git a/.agents/notes/implemented/process/2026-06-17-ts-build-config.md b/.agents/notes/implemented/process/2026-06-17-ts-build-config.md
index 1275a63524..17036438b8 100644
--- a/.agents/notes/implemented/process/2026-06-17-ts-build-config.md
+++ b/.agents/notes/implemented/process/2026-06-17-ts-build-config.md
@@ -43,6 +43,8 @@ In-package relative imports use explicit `.ts` specifiers.
- Referenced package/vendor projects keep the same emit behavior as build, so typecheck refreshes their `lib/types` outputs instead of using a separate no-emit graph. Project-specific strictness changes live in the owning `packages/*/*/tsconfig.json` or `vendor/*/tsconfig.json`.
- The no-emit aggregates disable `rewriteRelativeImportExtensions`; they emit nothing and include tests that import helpers across project-reference boundaries. Package/vendor emit projects keep the rewrite enabled.
+Composite projects keep their incremental build information inside their project-local `lib/` output. `pnpm run clean` derives live output directories from the root TypeScript project-reference graph, removes legacy root build information, and removes deleted `packages/*/*` directories that contain only known generated residue. It preserves `node_modules` for every package that still has a `package.json`, and refuses to remove a manifest-less directory containing unknown files. Build does not invoke clean automatically, so ordinary builds retain incremental state.
+
The command orchestration shape is:
```sh
@@ -55,6 +57,9 @@ tsx scripts/verify-node-next-types.ts
pnpm run typecheck:
tsc -b
+
+pnpm run clean:
+tsx scripts/clean.ts
```
`pnpm run demo:*` still runs `src` directly through tsx and root paths, without a compile step.
@@ -63,6 +68,8 @@ tsc -b
- **Keep `tsdown`/oxc as the TypeScript transformer** — oxc's transform is not `tsc` behavior (decorator transform differs, bundled JS differs from per-file emit), and its bundled `.d.ts` conflicts with Cordis' internal relative module augmentation shape.
- **One root strict program over packages, vendor, examples, tests, and scripts** — vendor source triggers type errors outside this project's ownership under the root strict flags; project references with per-project strictness are the boundary that works.
+- **Clean before every build** — this would discard the incremental state owned by `tsc` and the bundler even when the workspace layout is unchanged.
+- **Remove every package-level `node_modules`** — valid package dependency links do not cause the workspace-discovery failure, and deleting them would turn build cleanup into dependency reinstallation.
## Consequences
@@ -76,5 +83,6 @@ Build responsibilities are clearer:
- `lib/index.*` is the publish runtime output and is generated by the bundler, currently `tsdown`.
- `pnpm run verify-node-next-types` scans built declarations for relative specifiers without file extensions, then typechecks a temporary external ESM consumer with `moduleResolution: "NodeNext"` against the built `types`/`exports` surface, so declaration specifier regressions fail before publish.
- The `typecheck` command uses `tsconfig.json`. Examples, tests, and scripts are checked by the root no-emit project, while packages and vendor modules keep the same emit behavior as `build`. Package and vendor source stays behind project-reference boundaries.
+- After changing branches or updating a checkout that deleted packages, contributors can run `pnpm run clean` to remove stale package directories before rebuilding. Unknown files in a manifest-less package directory require manual classification instead of being deleted.
The Cordis vendor copy now has one more type-structure divergence from upstream. During upstream sync, that divergence must be reapplied or explicitly retired.
diff --git a/.agents/notes/implemented/process/2026-06-17-ts-build-config.zh.md b/.agents/notes/implemented/process/2026-06-17-ts-build-config.zh.md
index 7691118dd1..70e49c61de 100644
--- a/.agents/notes/implemented/process/2026-06-17-ts-build-config.zh.md
+++ b/.agents/notes/implemented/process/2026-06-17-ts-build-config.zh.md
@@ -43,6 +43,8 @@ Status: implemented
- 被引用的包/vendor 项目保持与构建相同的输出行为,因此类型检查会刷新它们的 `lib/types` 输出,而无需使用独立的 no-emit 图。项目特定的严格度变更放在各自的 `packages/*/*/tsconfig.json` 或 `vendor/*/tsconfig.json` 中。
- 两个 no-emit 聚合禁用 `rewriteRelativeImportExtensions`;它们不输出任何文件,且包含跨 project-reference 边界导入 helper 的测试。包/vendor 的 emit 项目保持重写开启。
+复合项目将增量构建信息保存在各项目本地的 `lib/` 输出中。`pnpm run clean` 会根据根 TypeScript project-reference 图确定当前有效的输出目录,删除遗留的根目录构建信息,并删除已删除包留下且仅包含已知生成残留的 `packages/*/*` 目录。对于仍有 `package.json` 的每个包,该命令都会保留 `node_modules`;如果不含 `package.json` 的目录中存在未知文件,则拒绝删除。构建不会自动调用 clean,因此常规构建会保留增量状态。
+
命令编排结构如下:
```sh
@@ -55,6 +57,9 @@ tsx scripts/verify-node-next-types.ts
pnpm run typecheck:
tsc -b
+
+pnpm run clean:
+tsx scripts/clean.ts
```
`pnpm run demo:*` 仍通过 tsx 和根路径直接运行 `src`,无需编译步骤。
@@ -63,6 +68,8 @@ tsc -b
- **继续使用 `tsdown`/oxc 作为 TypeScript 转换器**:oxc 的转换行为与 `tsc` 不同(装饰器转换有差异、打包 JS 与逐文件输出不同),且其打包 `.d.ts` 与 Cordis 内部的相对模块增强结构冲突。
- **用一个根目录严格程序覆盖包、vendor、示例、测试和脚本**:vendor 源码在根目录严格标志下会触发不属于本项目所有权范围的类型错误;带有逐项目严格度的 project references 才是可行的边界。
+- **每次构建前都执行清理**:即使工作区布局没有变化,这也会丢弃 `tsc` 和打包器拥有的增量状态。
+- **删除所有包级 `node_modules`**:有效的包依赖链接不会导致工作区发现失败,而删除这些链接会使构建清理变成重新安装依赖。
## 后果
@@ -76,5 +83,6 @@ tsc -b
- `lib/index.*` 是发布用的运行时输出,由打包器(当前为 `tsdown`)生成。
- `pnpm run verify-node-next-types` 扫描构建出的声明文件,检查是否存在缺少文件扩展名的相对说明符,然后以 `moduleResolution: "NodeNext"` 对构建出的 `types`/`exports` 接口进行临时外部 ESM 消费方的类型检查,确保声明说明符的回归在发布前被捕获。
- `typecheck` 命令使用 `tsconfig.json`。示例、测试和脚本由根 no-emit 项目检查,包和 vendor 模块保持与 `build` 相同的输出行为。包和 vendor 源码始终处于 project-reference 边界之后。
+- 切换分支或更新工作副本后,如果其中删除了包,贡献者可在重新构建前运行 `pnpm run clean`,删除残留的包目录。不含 `package.json` 的包目录如果存在未知文件,必须手动判定其类别,不能直接删除。
Cordis 的 vendor 副本现在与上游多了一处类型结构差异。在上游同步时,该差异必须被重新应用或明确废弃。
diff --git a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.i18n.yaml b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.i18n.yaml
index dbaa77f5ef..d3cd78a06b 100644
--- a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.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
-2026-07-03-documentation-graph-atlas.md: 9a30b13f9db6ceb2715517230e349cbe083850ec
-2026-07-03-documentation-graph-atlas.zh.md: 0734a2c659459f40ee81b9960f259681cb35891c
+2026-07-03-documentation-graph-atlas.md: 8b532fdd6f600eba05411588f8277b7cc43b3613
+2026-07-03-documentation-graph-atlas.zh.md: d426735407c0395fd96aed65a49aea0f0fdf6a90
diff --git a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.md b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.md
index 9a30b13f9d..8b532fdd6f 100644
--- a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.md
+++ b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.md
@@ -28,7 +28,7 @@ Every graph page declares one maintenance mode:
### First shipped index
-The index links eleven relationship surfaces. Package topology and tool-package affordances live in the existing generated catalogs that already own those facts; the remaining focused diagrams are generated by `scripts/gen-doc-graphs.ts`.
+The index links ten relationship surfaces. Package topology and tool-package affordances live in the existing generated catalogs that already own those facts; the remaining focused diagrams are generated by `scripts/gen-doc-graphs.ts`.
| Graph | Maintenance mode | Source of truth |
|---|---|---|
@@ -42,7 +42,6 @@ The index links eleven relationship surfaces. Package topology and tool-package
| [event producer/consumer matrix](../../../../docs/event-producer-consumer.md) | hybrid generated | Cordis event declarations, AST-scanned `ctx.on/emit/parallel/serial/waterfall` sites, and explicit dynamic dispatch overrides |
| [agent turn and step lifecycle](../../../../docs/agent-lifecycle.md) | curated | architecture.md loop lifecycle, Cordis catalog links, and session event semantics |
| [tool execution pipeline](../../../../docs/tool-execution-pipeline.md) | curated | tool pipeline semantics and the `tools/execute` waterfall |
-| [ACP snapshot replay](../../../../packages/ui/acp/snapshot-replay.md) | curated | snapshot harness behavior |
### Why generators own the docs
@@ -64,7 +63,7 @@ Committed diagrams use Mermaid because GitHub renders it in Markdown and it adds
## Consequences
-- Maintainers get visual entry points for topology, seams, event flow, lifecycle, app composition, and snapshot behavior.
+- Maintainers get visual entry points for topology, seams, event flow, lifecycle, and app composition.
- SDK users get a path from use case to package composition instead of only bottom-up package references.
- `doc-sync` now includes `verify-doc-graphs` and `verify-mermaid`, so graph drift and Mermaid syntax errors are caught with the other doc freshness gates.
- Future fs and hooks work has a concrete place to land new complexity: fs should expand the capability docs and tool catalog, while hooks should expand the event matrix and tool execution pipeline.
diff --git a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.zh.md b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.zh.md
index 0734a2c659..d426735407 100644
--- a/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.zh.md
+++ b/.agents/notes/implemented/process/2026-07-03-documentation-graph-atlas.zh.md
@@ -28,7 +28,7 @@ Status: implemented
### 首批发布的索引
-该索引链接十一种关系表面。包拓扑和工具包所提供的功能位于已经拥有这些事实的现有生成式目录中;其余聚焦图表由 `scripts/gen-doc-graphs.ts` 生成。
+该索引链接十种关系表面。包拓扑和工具包所提供的功能位于已经拥有这些事实的现有生成式目录中;其余聚焦图表由 `scripts/gen-doc-graphs.ts` 生成。
| 关系图 | 维护模式 | 真源 |
|---|---|---|
@@ -42,7 +42,6 @@ Status: implemented
| [事件生产者/消费方矩阵](../../../../docs/event-producer-consumer.md) | 混合生成式 | Cordis 事件声明、经 AST 扫描的 `ctx.on/emit/parallel/serial/waterfall` 位置,以及显式动态分派覆盖 |
| [agent 轮次与步骤生命周期](../../../../docs/agent-lifecycle.md) | 人工维护 | architecture.md 循环生命周期、Cordis 目录链接,以及会话事件语义 |
| [工具执行管线](../../../../docs/tool-execution-pipeline.md) | 人工维护 | 工具管线语义与 `tools/execute` waterfall(瀑布式事件)|
-| [ACP(Agent Client Protocol)快照回放](../../../../packages/ui/acp/snapshot-replay.md) | 人工策划 | 快照 harness 行为 |
### 为什么由生成器拥有文档
@@ -64,7 +63,7 @@ Status: implemented
## 后果
-- 维护者获得了拓扑、seam、事件流、生命周期、应用组合与快照行为的可视化入口。
+- 维护者获得了拓扑、seam、事件流、生命周期与应用组合的可视化入口。
- SDK 用户获得了从用例到包组合的路径,而非仅有自底向上的包参考。
- `doc-sync` 现在包含 `verify-doc-graphs` 和 `verify-mermaid`,因此关系图漂移和 Mermaid 语法错误与其他文档新鲜度门禁一起被捕获。
- 未来的文件系统和钩子工作有了承载新复杂度的具体位置:文件系统应扩展能力文档和工具目录,钩子应扩展事件矩阵和工具执行流水线。
diff --git a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.i18n.yaml b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.i18n.yaml
index d49010e985..45455444d5 100644
--- a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.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
-2026-07-04-doc-tiers-and-budgets.md: 9993eda6b0c7f1f8e5908dbd85fcfb4e5c6b3d1d
-2026-07-04-doc-tiers-and-budgets.zh.md: a40e7c2fe56ef86137bb7bb3ee8c818bdeed7eb9
+2026-07-04-doc-tiers-and-budgets.md: 7acdba8bfd96d183c418b3935d7b8e7f237d3607
+2026-07-04-doc-tiers-and-budgets.zh.md: 0f07a92740d31dd13cb523d73ac8a3699d666d30
diff --git a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.md b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.md
index 9993eda6b0..7acdba8bfd 100644
--- a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.md
+++ b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.md
@@ -18,7 +18,7 @@ Standing docs accumulated repeated rules, retold incidents, duplicated package m
## Alternatives considered
- **Skill and review discipline without a gate** — rejected: the accretion above happened while the current-state rule and reviewer attention already existed; a prose rule with no mechanical backstop demonstrably does not hold here, and this repo's own [quality-gates stance](2026-06-11-quality-gates.md) says invariants worth keeping are worth encoding.
-- **A broad gate over every doc tier** — rejected: a blanket ceiling punishes exactly the right kind of long doc (a feature matrix or type catalog where every row is a fact, e.g. `packages/ui/acp/acp-feature-support.md`) and generates per-file override churn that trains contributors to rubber-stamp raises.
+- **A broad gate over every doc tier** — rejected: a blanket ceiling punishes exactly the right kind of long doc (a feature matrix or type catalog where every row is a fact) and generates per-file override churn that trains contributors to rubber-stamp raises.
- **Housing the standard inside the skill** — rejected: contracts live in docs and workflows in skills; a standard packed into SKILL.md is invisible to an agent that edits docs without invoking the skill, and `docs/AGENTS.md` already loads as subtree instructions for anyone working under `docs/`.
## Consequences
diff --git a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.zh.md b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.zh.md
index a40e7c2fe5..0f07a92740 100644
--- a/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.zh.md
+++ b/.agents/notes/implemented/process/2026-07-04-doc-tiers-and-budgets.zh.md
@@ -18,7 +18,7 @@ Status: implemented
## 曾考虑的替代方案
- **仅靠 skill 和评审纪律,不设门禁**:否决。上述膨胀正是在现行规则和评审注意力已经存在的情况下发生的;一条没有机械后盾的行文规则在此处已被证明无法维持,而本仓库自身的[质量门禁立场](2026-06-11-quality-gates.md)认为值得保持的不变式就值得编码。
-- **对所有文档层级全面设限**:否决。一刀切的上限恰好惩罚了那些正当的长文档(如特性矩阵或类型目录,每一行都是事实,例如 `packages/ui/acp/acp-feature-support.md`),并产生逐文件的例外变更,训练贡献者机械地批准提限。
+- **对所有文档层级全面设限**:否决。一刀切的上限恰好惩罚了那些正当的长文档(如特性矩阵或类型目录,每一行都是事实),并产生逐文件的例外变更,训练贡献者机械地批准提限。
- **将标准放在 skill 内部**:否决。契约归文档,工作流归 skill;如果标准被塞进 SKILL.md,那些不调用该 skill 而直接编辑文档的 agent(智能体)就看不到它,而 `docs/AGENTS.md` 已经作为子树指令被任何在 `docs/` 下工作的人加载。
## 后果
diff --git a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.i18n.yaml b/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.i18n.yaml
deleted file mode 100644
index 9aa8a03c52..0000000000
--- a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-06-parallel-github-ci-gates.md: 5c276f6a75936021369bc5ad9494c9aa6e4e3fc3
-2026-07-06-parallel-github-ci-gates.zh.md: 7d98f842ef1d60a3a5b727f975cb1d93ea6f253c
diff --git a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.md b/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.md
deleted file mode 100644
index 5c276f6a75..0000000000
--- a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.md
+++ /dev/null
@@ -1,50 +0,0 @@
-# Agent Note: Parallel GitHub CI gates
-
-Status: implemented
-
-English | [中文](2026-07-06-parallel-github-ci-gates.zh.md)
-
-## Problem
-
-The keyless GitHub CI gates are mostly orthogonal: typecheck, lint, documentation freshness, coverage, snapshot replay, build, package-publication hygiene, demo smoke, and built-bin smoke fail for different reasons and do not need each other's runtime state. Running them as one ordered command chain makes the workflow wall clock equal the sum of those gates, while splitting every short leaf into its own GitHub job repeats checkout, Node setup, pnpm restore, and install work until orchestration overhead becomes the bottleneck.
-
-The original broad-lane split stopped meeting that balance as the workspace grew. On the merge of PR #404, Linux static, coverage, snapshot, and artifact jobs took 148, 195, 94, and 230 seconds; Windows static and artifacts took 251 and 482 seconds. Package-manager packing once per package dominated both artifact validators, coverage needlessly rebuilt output before a source-only suite, and CPU-heavy gates contended inside the static and coverage lanes.
-
-The artifact boundary remains load-bearing. `publint`, `verify-node-next-types`, compiled invariant loading, and built-bin smoke tests need emitted `lib/` output. Sharding cannot race those consumers ahead of build or replace their published-artifact signal with source execution.
-
-## Decision
-
-The production topology below is historical and is superseded by [Evidence-based larger hosted runners](2026-07-22-evidence-based-larger-hosted-runners.md). The larger-runner decision removes its shard selectors and workflow jobs; this note preserves why that earlier topology was implemented.
-
-[CI](../../../../.github/workflows/ci.yml) treats one minute for non-Windows jobs and three minutes for Windows jobs as observed performance targets, not cancellation deadlines. Hosted-runner variance should leave complete timing evidence and useful failure logs instead of cancelling an otherwise-correct gate. The [serial cross-platform CI reference](2026-07-21-serial-cross-platform-ci-reference.md) independently runs the complete unsharded primary Node aggregate on Linux, macOS, and Windows so the optimized lane inventory is not its own completeness oracle.
-
-In that topology, [scripts/run-gates.ts](../../../../scripts/run-gates.ts) was the common bounded scheduler and GitHub supplied explicit shard names for the expensive gate families. `scripts/static-shards.ts` partitioned static gates into foundation, documentation-type, API-contract, catalog, prose, documentation-projection, and documentation-build ownership and rejected a missing or duplicate gate assignment. Linux lint used disjoint A-C, D-M, N-S, and T-Z package-source and package-test lanes, while Windows used complete package-source and package-test lanes; both included a repository complement starting from `.` so new top-level targets could not disappear between shards and owned the single cross-file duplication run. `scripts/coverage-shards.ts` assigned every workspace package to exactly one source-coverage lane. Directory filters retained a trailing separator because Vitest positional filters match substrings and would otherwise admit prefix-named siblings. Each coverage lane included only its owned source files, repeated the exhaustive companion topology test, and ran without a preceding build because the complete coverage suite passes from a tree with every generated `lib/` removed.
-
-Snapshot replay used two explicit multi-file lanes and eight scenario partitions of the large ACP file. `scripts/snapshot-shards.ts` owned that inventory, and its test discovered every file admitted by the snapshot config. Each snapshot job installed dependencies while its Linux runner prepared Bubblewrap, built the shipped runtime, and ran only its assigned replay surface. The suite retained bounded concurrency of five subprocesses because replay spent most of its time waiting on child protocol I/O. Fixture guards still inspected the complete ACP scenario table in every partition.
-
-Cold standalone documentation typechecking rebuilds the complete project-reference graph, so a dedicated documentation-type lane builds once and checks Markdown blocks against those declarations. The Linux documentation lane uses VitePress's MPA build to retain page rendering and dead-link validation within the observed non-Windows target; separate blocking Windows build and production-site lanes preserve the emitted-package and shipped-site checks without putting both critical paths in one job.
-
-Artifacts use two lanes: one metadata lane for `publint`, NodeNext declarations, and compiled invariant loading, plus one built-bin smoke lane. Each lane produces its own build before its consumers. Repeating the short build costs runner minutes but avoids an upload/download dependency and keeps each job's critical path bounded.
-
-[scripts/publint-all.ts](../../../../scripts/publint-all.ts) calls publint's supported API in-process against an in-memory publication view made from each manifest's declared files and npm's mandatory metadata files. This preserves the distinction between workspace files and published files without spawning a package-manager pack command 103 times. [scripts/verify-built-package-invariants.mjs](../../../../scripts/verify-built-package-invariants.mjs) stages those structurally validated manifest-declared `lib/` files below the real package, then imports the compiled self-reference through plain Node and Cordis Loader normalization. A companion that reaches an undeclared runtime chunk still fails.
-
-Compatibility lanes run the source worker and Zstandard runtime smokes on every advertised Node line. TypeScript checks the source graph once in a dedicated primary Node 24 lane; repeating the same compiler analysis in runtime compatibility jobs added time without runtime-specific signal.
-
-The workflow caches the pnpm store, keys each immutable ESLint cache to its owning lint shard, preserves native PowerShell for Windows measurements, and retains one aggregate `all checks passed` status for branch protection. Windows reuses the three exhaustive lint partitions and groups foundation/catalog/prose plus documentation-type/API-contract gates behind shared runner setups; only scheduling differs from the Linux partitions. Windows build and production-site validation remain blocking, while the wider Windows static, lint, and artifact matrix remains observational.
-
-## Alternatives considered
-
-- **Keep the broad lanes** - minimizes workflow YAML, but it preserves the measured multi-minute feedback loop.
-- **Run every leaf gate as a separate GitHub job** - maximizes fan-out, but short generators and prose checks would spend more time preparing a runner than checking the repository.
-- **Upload one build to artifact consumers** - avoids repeated compilation, but upload/download and dependency scheduling lengthen wall time; the clean build is short enough to repeat inside bounded lanes.
-- **Keep package-manager packing in both publication gates** - delegates inventory selection to pnpm, but repeats more than 200 package-manager processes. The manifest structural gate plus publication-view fixtures make the optimized inventory contract explicit and fail on an on-disk but unpublished dependency.
-- **Keep build before coverage** - provides emitted output the source suite no longer consumes; a clean-tree coverage proof showed it was pure latency.
-- **Typecheck on every Node version** - repeats compiler work while the compatibility smokes already exercise actual Node-specific loading and compression behavior.
-
-## Consequences
-
-The shard inventories and matrix jobs described above are not part of the current repository contract. The superseding larger-runner decision keeps the complete primary inventory in one process and uses the serial suite as its independent completeness oracle.
-
-The optimized publication validators rely on the manifest `files` contract enforced by `verify-package-invariants`. If publication rules grow beyond that contract, the structural gate and both staged views must change together.
-
-Compatibility jobs no longer claim that TypeScript itself was exercised under every Node runtime. They prove runtime-sensitive source loading on Node 22, 24, and 26, while the primary runtime owns the single source-graph typecheck.
diff --git a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.zh.md b/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.zh.md
deleted file mode 100644
index 7d98f842ef..0000000000
--- a/.agents/notes/implemented/process/2026-07-06-parallel-github-ci-gates.zh.md
+++ /dev/null
@@ -1,50 +0,0 @@
-# Agent Note: 并行 GitHub CI 门禁
-
-Status: implemented
-
-[English](2026-07-06-parallel-github-ci-gates.md) | 中文
-
-## 问题
-
-无密钥 GitHub CI 门禁大多相互正交:类型检查、lint、文档新鲜度、覆盖率、快照重放、构建、包(package)的发布卫生检查、demo 冒烟和已构建二进制冒烟会因不同原因失败,也不需要彼此的运行时状态。将它们作为一条有序命令链运行,会使工作流墙钟时间等于所有门禁耗时之和;而把每个短小叶子拆成独立 GitHub job,又会反复执行 checkout、Node 设置、pnpm 恢复和安装,直到编排开销成为瓶颈。
-
-随着 workspace 增长,原有的宽车道拆分不再满足这一平衡。PR(Pull Request)#404 合并时,Linux 的静态、覆盖率、快照和产物 job 分别耗时 148、195、94 和 230 秒;Windows 的静态和产物 job 分别耗时 251 和 482 秒。每个包都调用一次包管理器打包,主导了两个产物验证器的耗时;覆盖率在仅运行源码的套件前无谓地重建输出;CPU 密集型门禁则在静态与覆盖率车道内争用资源。
-
-产物边界仍然承载关键约束。`publint`、`verify-node-next-types`、已编译不变量加载和已构建二进制冒烟测试都需要生成的 `lib/` 输出。分片不能让这些消费方抢在构建前运行,也不能用源码执行取代它们对已发布产物的信号。
-
-## 决策
-
-下述生产拓扑已经成为历史,并由[基于证据采用更大的托管 runner](2026-07-22-evidence-based-larger-hosted-runners.md) 取代。更大 runner 的决策移除了其分片选择器和工作流 job;本文保留早期拓扑为何被实现的记录。
-
-[CI](../../../../.github/workflows/ci.yml) 将非 Windows job 的一分钟和 Windows job 的三分钟视为观测所得的性能目标,而非取消截止时间。托管 runner 的波动应留下完整计时证据和有用的失败日志,而不是取消本来正确的门禁。[串行跨平台 CI 参考](2026-07-21-serial-cross-platform-ci-reference.md)会在 Linux、macOS 和 Windows 上独立运行完整、未分片的主 Node 聚合,使优化后的车道清单不会成为自身完整性的唯一判据。
-
-在该拓扑中,[scripts/run-gates.ts](../../../../scripts/run-gates.ts) 是通用的有界调度器,GitHub 则为昂贵的门禁族提供显式分片名称。`scripts/static-shards.ts` 将静态门禁划分为基础、文档类型、API 契约、目录、正文、文档投影和文档构建等归属,并拒绝缺失或重复的门禁分配。Linux lint 使用互不重叠的 A-C、D-M、N-S、T-Z 包源码和包测试车道,Windows 则使用完整的包源码与包测试车道;两者都包含从 `.` 开始的仓库补集,使新增顶层目标无法消失在分片之间,并负责唯一一次跨文件重复检查。`scripts/coverage-shards.ts` 把每个 workspace 包恰好分配给一个源码覆盖率车道。目录过滤器保留尾部分隔符,因为 Vitest 位置过滤器按子字符串匹配,否则会纳入具有同名前缀的相邻项。每个覆盖率车道只包含其拥有的源码文件,重复运行穷尽式伴随拓扑测试,并且不先执行构建,因为从删除了所有生成式 `lib/` 的树开始,完整覆盖率套件仍可通过。
-
-快照重放使用两个显式多文件车道,以及大型 ACP(Agent Client Protocol)文件的八个场景分区。`scripts/snapshot-shards.ts` 拥有该清单,其测试会发现快照配置允许的每个文件。每个快照 job 在其 Linux runner 准备 Bubblewrap 的同时安装依赖,随后构建已发布运行时,并且只运行分配给它的重放表面。该套件保留五个子进程的有界并发,因为重放的大部分时间都在等待子进程协议 I/O。fixture(测试前置数据)守卫仍会在每个分区中检查完整 ACP 场景表。
-
-冷启动的独立文档类型检查会重建完整的项目引用图,因此专用文档类型车道只构建一次,再用这些声明检查 Markdown 块。Linux 文档车道使用 VitePress 的 MPA 构建,在观测所得的非 Windows 目标内保留页面渲染与死链接验证;单独的阻塞式 Windows 构建和生产站点车道保留已生成包与已发布站点检查,同时避免把两条关键路径放进同一个 job。
-
-产物使用两个车道:一个元数据车道负责 `publint`、NodeNext 声明和已编译不变量加载,另一个负责已构建二进制冒烟。每个车道都会在其消费方之前自行构建。重复短时构建会消耗 runner 分钟数,但避免了上传/下载依赖,并使每个 job 的关键路径保持有界。
-
-[scripts/publint-all.ts](../../../../scripts/publint-all.ts) 在进程内针对内存发布视图调用 publint 支持的 API;该视图由每份清单声明的文件和 npm 强制元数据文件构成。这样无需生成 103 次包管理器打包命令,也能保留 workspace 文件与已发布文件之间的区别。[scripts/verify-built-package-invariants.mjs](../../../../scripts/verify-built-package-invariants.mjs) 在真实包下暂存这些经过结构验证、由清单声明的 `lib/` 文件,再通过纯 Node 和 Cordis Loader 规范化导入已编译的自引用。若伴随项触及未声明的运行时分片,仍会失败。
-
-兼容性车道会在每条声明支持的 Node 版本线上运行源码 worker 和 Zstandard 运行时冒烟。TypeScript 在专用的主 Node 24 车道中只检查一次源码图;在运行时兼容性 job 中重复同一编译器分析只会增加耗时,不会提供运行时特有信号。
-
-工作流缓存 pnpm store,将每个不可变 ESLint 缓存的键绑定到其所属 lint 分片,为 Windows 测量保留原生 PowerShell,并保留一个聚合的 `all checks passed` 状态用于分支保护。Windows 复用三个穷尽式 lint 分区,并在共享 runner 设置后组合基础/目录/正文门禁与文档类型/API 契约门禁;只有调度方式与 Linux 分区不同。Windows 构建和生产站点验证继续阻塞,而更广泛的 Windows 静态、lint 和产物矩阵仍为观察性检查。
-
-## 曾考虑的替代方案
-
-- **保留宽车道**:最大限度减少工作流 YAML,但会保留观测到的数分钟反馈周期。
-- **让每个叶子门禁分别成为 GitHub job**:最大化扇出,但短小的生成器和正文检查准备 runner 的时间会超过检查仓库的时间。
-- **向产物消费方上传一次构建**:避免重复编译,但上传/下载和依赖调度会延长墙钟时间;干净构建足够短,可以在有界车道内重复。
-- **在两个发布门禁中保留包管理器打包**:把清单选择委托给 pnpm,但会重复启动 200 多个包管理器进程。清单结构门禁加发布视图 fixture 使优化后的清单契约显式化,并会在存在磁盘上有但未发布的依赖时失败。
-- **在覆盖率前保留构建**:提供源码套件已不再消费的生成输出;干净树覆盖率证明表明这只是纯粹的延迟。
-- **在每个 Node 版本上执行类型检查**:重复编译器工作,而兼容性冒烟已经验证实际的 Node 特有加载与压缩行为。
-
-## 后果
-
-上述分片清单和矩阵 job 不属于当前仓库契约。取而代之的更大 runner 决策在单个进程中保留完整主清单,并以串行套件作为独立完整性判据。
-
-优化后的发布验证器依赖由 `verify-package-invariants` 强制执行的清单 `files` 契约。如果发布规则超出该契约,结构门禁和两个暂存视图必须一起变化。
-
-兼容性 job 不再声称 TypeScript 本身已在每个 Node 运行时下执行。它们证明 Node 22、24 和 26 上对运行时敏感的源码加载,而主运行时负责唯一一次源码图类型检查。
diff --git a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.i18n.yaml b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.i18n.yaml
index ae5ed9b11e..a0226ea124 100644
--- a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.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
-2026-07-19-require-agent-notes-for-non-trivial-changes.md: f2645832ebcdd0b81cbff5415c7eb6f60b6fa8cf
-2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md: 659aa7cad0823fa0082be1827f8c083037376a4c
+2026-07-19-require-agent-notes-for-non-trivial-changes.md: 32d7408b3d56e6571a14a8191e9b4b0fe901f5a3
+2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md: 713845706e650b4b4acd591368d9bcff137a38b7
diff --git a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.md b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.md
index f2645832eb..32d7408b3d 100644
--- a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.md
+++ b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.md
@@ -14,6 +14,10 @@ Every non-trivial change adds or updates at least one Agent Note in the same PR.
Updating the note that already owns a decision satisfies the rule; a new note is required only when no note owns it. Purely mechanical or local edits with no behavioral, contractual, structural, process, or rationale change are exempt. The [Agent Notes README](../../README.md#when-to-write-one) owns this boundary, while root `AGENTS.md` carries the standing order.
+A fully superseded implemented note may be consolidated into the current owning note and deleted only after that owner preserves every unique rationale, alternative, consequence, verification contract, and named coverage gap. The same change repairs inbound links and removes any Chinese counterpart, consistency record, and `required` entry in `scripts/translation-pairing.manifest.json`. Partial supersession keeps both notes cross-linked and current; consolidation neither rewrites an old decision into its opposite nor leaves git history as the only copy of rationale.
+
+When a later decision removes an earlier feature completely, the removal note becomes the current owner only after the feature is absent from production code, configuration, schemas, durable or wire formats, migration, and compatibility behavior; no current documentation presents it as available; and no test exercises it as supported behavior. Removal rationale and tests that verify absence may remain. The removal owner preserves the feature's original motivation, why that motivation no longer justified the surface, alternatives to full removal, the capability given up, conditions for reintroduction, and verification of complete absence. Implementation inventories and tests that only described the deleted behavior are obsolete rather than current verification contracts. A removal limited to one transport, default, implementation, or presentation remains partial supersession.
+
Review enforces the semantic boundary. No automated gate attempts to classify a diff as trivial or non-trivial, so this policy adds no gate stage or runtime.
## Alternatives considered
@@ -22,10 +26,21 @@ Review enforces the semantic boundary. No automated gate attempts to classify a
**Require a new note for every change.** This duplicates an existing note when it already owns the decision and adds empty ceremony to purely mechanical edits.
+**Keep every fully superseded note indefinitely.** A cross-linked record is necessary while part of its decision remains current, but a wholly obsolete implemented note contradicts the current-state contract and duplicates rationale that can have one owner.
+
+**Add a `superseded/` lifecycle.** Another lifecycle would retain the obsolete record and expand the tree, format gate, and maintenance rules without reducing duplication.
+
+**Rewrite the old note into the replacement decision.** This erases the decision boundary and its rejected alternatives. Consolidation instead preserves those facts in the current owner before deleting the obsolete file.
+
+**Preserve every implementation and test detail from a removed feature.** This recreates the obsolete note inside its replacement. The removal owner keeps the rationale and verification needed to understand or revisit the current absence, while deleted mechanics remain available in git history.
+
**Add a CI diff-classification gate.** A mechanical check cannot reliably determine whether a semantic change is trivial, while another gate adds runtime and invites false positives or superficial compliance.
## Consequences
- Every substantial change preserves its rationale and rejected alternatives beside the implementation.
- Contributors maintain an existing owning note instead of creating duplicate records.
+- Fully superseded records can collapse into one current owner without losing their unique rationale or verification contract.
+- Features that were later removed can have one current owner without carrying obsolete implementation and test inventories forward.
+- Partial supersession remains explicit and cross-linked, while deletion requires link, bilingual-pair, and required-manifest cleanup in the same change.
- Mechanical edits remain lightweight, and the gate topology and runtime remain unchanged.
diff --git a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md
index 659aa7cad0..713845706e 100644
--- a/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md
+++ b/.agents/notes/implemented/process/2026-07-19-require-agent-notes-for-non-trivial-changes.zh.md
@@ -14,6 +14,10 @@ Status: implemented
更新已经持有该决策的 Agent Note 即满足规则;仅当没有 Agent Note 持有该决策时才新增记录。完全机械或局部、且不改变行为、契约、结构、流程或决策依据的编辑可豁免。[Agent Notes README](../../README.md#when-to-write-one) 持有这条边界,根目录 `AGENTS.md` 则携带常驻指令。
+只有在当前持有该决策的记录保存了所有独有的决策依据、备选方案、影响、验证契约和明确指出的覆盖缺口后,才可将被完全取代的 implemented Agent Note 合并到该记录中并删除。同一变更还要修复入站链接,并删除中文对侧文件、一致性记录,以及 `scripts/translation-pairing.manifest.json` 中对应的 `required` 条目。仅部分被取代时,两个记录仍需互相链接并保持与现状一致;合并既不将旧决策改写成与其相反的决策,也不让 git 历史成为决策依据的唯一副本。
+
+后续决策完全移除较早的功能时,只有该功能已从生产代码、配置、schema、持久化格式或协议格式、迁移和兼容行为中消失,当前文档不再将其描述为可用,且没有测试把它作为受支持行为来执行,移除记录才会成为当前持有记录。移除决策的依据和验证该功能已不存在的测试可以保留。它必须保留该功能的最初动机、为什么该动机已不足以证明继续保留该功能、完全移除之外的备选方案、放弃的能力、重新引入的条件,以及证明已彻底移除的验证。只描述已删除行为的实现清单和测试已经过时,不属于当前验证契约。仅移除一种传输、默认值、实现或展示仍属于部分取代。
+
评审负责执行这条语义边界。自动化门禁不尝试把差异分类为平凡或实质性变更,因此这项政策不会增加门禁阶段或运行时间。
## 备选方案
@@ -22,10 +26,21 @@ Status: implemented
**每项变更都必须新增 Agent Note。** 当现有 Agent Note 已经持有该决策时,这会产生重复记录,也会让纯机械编辑承担空洞的流程负担。
+**永久保留每份被完全取代的 Agent Note。** 只要旧决策仍有部分适用,就需要保留互相链接的记录;但完全失效的 implemented Agent Note 与记录当前状态的契约相矛盾,并重复保存本可由一个记录持有的决策依据。
+
+**新增 `superseded/` 生命周期。** 新增生命周期仍会保留过时记录并扩张目录树、格式门禁和维护规则,却无法减少重复内容。
+
+**将旧 Agent Note 改写为替代它的决策。** 这样会抹去决策边界及其否决的备选方案。合并做法是在删除过时文件前,先由当前持有决策的记录保存这些事实。
+
+**保留已移除功能的每一项实现与测试细节。** 这会在替代记录中重建过时记录。移除决策的持有记录只保留理解或重新审视当前已移除状态所需的决策依据与验证,已删除机制仍可从 git 历史查看。
+
**添加 CI 差异分类门禁。** 机械检查无法可靠判断语义变更是否平凡,额外门禁还会增加运行时间,并引入误报或表面合规。
## 影响
- 每项实质性变更都会在实现旁保留其决策依据和被放弃的备选方案。
- 贡献者维护现有的决策持有记录,而不是创建重复记录。
+- 被完全取代的记录可以归并到一个当前持有记录中,同时不丢失其独有的决策依据或验证契约。
+- 后来被移除的功能可以只有一个当前持有记录,而无需继续保留过时的实现与测试清单。
+- 仅部分被取代的情况仍需明确记录并互相链接;删除记录则必须在同一变更中清理链接、双语配对和 `scripts/translation-pairing.manifest.json` 的 `required` 条目。
- 机械编辑仍保持轻量,门禁拓扑和运行时间也保持不变。
diff --git a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.i18n.yaml b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.i18n.yaml
index b21353698c..ffa1e87103 100644
--- a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.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
-2026-07-20-gui-testing-system.md: e42dafcdf37e48475e7d420eaad9600e6c20891c
-2026-07-20-gui-testing-system.zh.md: e4ef6246e59e0ad6c0a3070a38c546f964e347fa
+2026-07-20-gui-testing-system.md: 546f65f065c0c2266773acc3c28b2833a094ba9b
+2026-07-20-gui-testing-system.zh.md: 6601ae0a1c2bd1671af6f02961fbda81d30ab971
diff --git a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.md b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.md
index e42dafcdf3..546f65f065 100644
--- a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.md
+++ b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.md
@@ -20,7 +20,7 @@ Cut along the architecture's natural test seams into three tiers, bottom-up:
|---|---|---|---|
| 1 Protocol isomorphism | `AbstractApiClient` + `toFetchHandler` (bidirectional data / rpcId / zod types / SSE streams / batching / timeouts) | **The full chain at the isomorphic point**: `InProcessApiClient(toFetchHandler(脚本化 impl))` skips the network but genuinely runs the wire serialization — zero browser, pure node env | `packages/host/apiproxy/tests/client-handler.spec.ts` |
| 2 Object-layer orchestration | `Session`/`SessionManager`/`ConnectionController` (state machines and timing: stitching / dedup / paging / optimistic draft clearing / pendingBuffers / reconnect / backoff) | **The "event sequence in → snapshot out" golden path**: programmable fakes + deferreds controlling timing + fake timers controlling backoff | `packages/client/{runtime,connection}/tests/` |
-| 3 Assembled presentation | Built artifacts × the real client loader and plugin composition | App-owned semantic snapshots boot all eight built client plugins under jsdom for deterministic cross-plugin state changes; bare Playwright smoke separately proves the real browser/carrier boundary, with real-host cases self-skipping without a key | `apps/web/tests/*.snapshot.ts`, `apps/web/tests/smoke-{fixture,real}.e2e.ts` |
+| 3 Assembled presentation | Built artifacts × the real client loader and plugin composition | App-owned semantic snapshots boot all eight built client plugins under jsdom for deterministic cross-plugin state changes; bare Playwright smoke separately proves the real browser/carrier boundary, with real-host cases self-skipping without a key; the keyless browser e2e lane disables the shipped model-adapter row and replays recorded session fixtures through `dsh-llm-replay` in the real in-process web assembly against conversation aria goldens ([web e2e lane](../testing/2026-07-24-web-gui-browser-e2e-lane.md)) | `apps/web/tests/*.snapshot.ts`, `apps/web/tests/smoke-{fixture,real}.e2e.ts`, `apps/web/tests/{replay-round-trip,seeded-history}.e2e.ts` |
Inter-tier discipline: **each tier tests its own layer, upper tiers never re-test lower ones** — an app semantic snapshot pins only user-visible projection across the assembled plugin boundary, while Playwright smoke proves browser and carrier liveness; wire semantics belong to tier 1 and data semantics to tier 2. Pure-function layers (lineage/partial/notifier/fold-adapter) are tested directly with zero fakes in the same package's tests/ alongside tier 2.
@@ -33,15 +33,15 @@ Inter-tier discipline: **each tier tests its own layer, upper tiers never re-tes
|---|---|---|---|
| Baseline | `pnpm run test:gui` | Tier 1+2 vitest (`packages/client packages/host`), seconds-fast, no browser, no server | Casually, after touching any GUI source |
| Semantic snapshot | `DSH_EXAMPLE_MODE=lib pnpm run test:snapshot` | Keyless assembled-application semantics plus the repo's transport-specific expected outputs | After a human-visible GUI change; before delivery |
-| Browser end-to-end | `pnpm run test:web` | Rebuilds the front-end dist first, then runs the tier-3 two-level smoke (fixture level + real-host level self-skip) | After touching the build surface/boot/carriage; before delivery |
+| Browser end-to-end | `pnpm run test:web` | Rebuilds the front-end dist first, then runs the tier-3 browser set: the two-level smoke (fixture level + real-host level self-skip) plus the keyless replayed e2e scenarios (`DSH_SNAPSHOT=record`/`refresh` re-record fixtures / rewrite goldens) | After touching the build surface/boot/carriage; before delivery |
| Gate | `pnpm run test:coverage` | The repo-wide gate (host and client GUI packages included, except annotated browser-grade exclusions) | The PR window |
**Division of labor between the browser scripts and vitest**: Playwright owns browser/carrier black-box regression and long sequential user journeys; ordinary vitest owns data-layer semantics such as reference stability, timing, and wire shapes; snapshot vitest owns stable app-level semantic output through the built composition. These lanes complement each other rather than duplicating assertions.
## Anti-regression discipline
-- **Every bug fix pins an assertion**: a browser-visible bug is pinned into the regression section of its owning verify script (one pin = one report line); a data-layer bug is pinned into the matching spec (precedent: the res-close misjudgment pinned in the webserver bridge suite — pure Node, reproduces in seconds, no longer needs the 12s browser sentinel as the only defense).
-- **All-green on fixture is not done, the real host must pass too**: what the fixture short-circuits is exactly the wire carriage chain (node:http bridge close semantics, real network timing); both empirically confirmed bugs hid there. Changes touching connection/bridge/handler/SSE must run `verify-session-real`.
+- **Every bug fix pins an assertion**: a browser-visible bug is pinned into its owning browser spec (smoke or e2e scenario); a data-layer bug is pinned into the matching spec (precedent: the res-close misjudgment pinned in the webserver bridge suite — pure Node, reproduces in seconds, no longer needs the 12s browser sentinel as the only defense).
+- **All-green on fixture is not done, the real wire must pass too**: what the fixture short-circuits is exactly the wire carriage chain (node:http bridge close semantics, real network timing); both empirically confirmed bugs hid there. Changes touching connection/bridge/handler/SSE must run the browser lane (`pnpm run test:web`) — its keyless e2e scenarios drive the real HTTP/SSE carriage, and the with-key real-host smoke remains the live-model complement.
- The code-on-disk-is-the-answer reconciliation workflow: when a behavior change lands and turns existing cases red, reconcile on the spot (fix the test or fix the code, with the RFC/contract as arbiter); no red left hanging.
## Consequences
diff --git a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.zh.md b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.zh.md
index e4ef6246e5..6601ae0a1c 100644
--- a/.agents/notes/implemented/process/2026-07-20-gui-testing-system.zh.md
+++ b/.agents/notes/implemented/process/2026-07-20-gui-testing-system.zh.md
@@ -20,7 +20,7 @@ GUI 栈需要考虑多种应用形态,同应用形态内的不同运行环境
|---|---|---|---|
| 1 协议同构层 | `AbstractApiClient` + `toFetchHandler`(双向数据/rpcId/ZOD类型/SSE 流/合批/超时) | **同构点全链**:`InProcessApiClient(toFetchHandler(脚本化 impl))` 不过网络但真跑 wire 序列化——零浏览器、纯 node env | `packages/host/apiproxy/tests/client-handler.spec.ts` |
| 2 对象层编排 | `Session`/`SessionManager`/`ConnectionController`(状态机与时序:缝合/去重/翻页/乐观清稿/pendingBuffers/重连/退避) | **「事件序列进→快照出」黄金路径**:可编程假体 + deferred 控时序 + fake timers 控退避 | `packages/client/{runtime,connection}/tests/` |
-| 3 组装呈现层 | 构建产物 × 真实 client loader 与插件组合 | 归应用所有的语义快照会在 jsdom 下启动全部 8 个已构建的 client 插件,以固定确定性的跨插件状态变化;独立使用 Playwright 裸库的冒烟测试负责验证真实浏览器/承载层边界,真 host 用例在无密钥时自行跳过 | `apps/web/tests/*.snapshot.ts`、`apps/web/tests/smoke-{fixture,real}.e2e.ts` |
+| 3 组装呈现层 | 构建产物 × 真实 client loader 与插件组合 | 归应用所有的语义快照会在 jsdom 下启动全部 8 个已构建的 client 插件,以固定确定性的跨插件状态变化;独立使用 Playwright 裸库的冒烟测试负责验证真实浏览器/承载层边界,真 host 用例在无密钥时自行跳过;无密钥浏览器 e2e 车道会禁用交付配置中的模型适配器行,并通过 `dsh-llm-replay` 在真实进程内 web 组装中回放录制的会话 fixture,与会话区 aria 期望输出比对([web e2e 车道](../testing/2026-07-24-web-gui-browser-e2e-lane.md)) | `apps/web/tests/*.snapshot.ts`、`apps/web/tests/smoke-{fixture,real}.e2e.ts`、`apps/web/tests/{replay-round-trip,seeded-history}.e2e.ts` |
层间纪律:**下层各测各的,上层不重测下层**:应用语义快照只固定组装后插件边界上的用户可见投影,Playwright 冒烟测试负责验证浏览器与承载层是否存活;wire 语义归 1 层,数据语义归 2 层。纯函数层(lineage/partial/notifier/fold-adapter)随 2 层同包 tests/ 零假体直测。
@@ -33,15 +33,15 @@ GUI 栈需要考虑多种应用形态,同应用形态内的不同运行环境
|---|---|---|---|
| 基础 | `pnpm run test:gui` | 1+2 层 vitest(`packages/client packages/host`),秒级、无浏览器无 server | 改 GUI 任意源码后随手跑 |
| 语义快照 | `DSH_EXAMPLE_MODE=lib pnpm run test:snapshot` | 无需密钥的组装应用语义,以及仓库按传输形态划分的预期输出 | 用户可见的 GUI 变更后;交付前 |
-| 浏览器端到端 | `pnpm run test:web` | 先重建前端 dist,再跑 3 层双级 smoke(fixture 级 + 真 host 级 self-skip) | 改构建面/boot/承载后;交付前 |
+| 浏览器端到端 | `pnpm run test:web` | 先重建前端 dist,再跑 3 层浏览器全集:双级 smoke(fixture 级 + 真 host 级 self-skip)加上无密钥回放 e2e 场景(`DSH_SNAPSHOT=record`/`refresh` 重录 fixture / 重写期望输出) | 改构建面/boot/承载后;交付前 |
| 门禁 | `pnpm run test:coverage` | 全仓 gate(host 与 client GUI 包均纳入,仅排除带注释的浏览器级例外) | PR 窗口 |
**浏览器脚本与 vitest 的分工**:Playwright 负责浏览器/承载层黑盒回归和较长的连续用户操作流程;普通 vitest 负责引用稳定性、时序和 wire 结构等数据层语义;快照 vitest 通过构建后的组合负责稳定的应用层语义输出。这些车道彼此互补,而不重复断言。
## 防回归纪律
-- **修一个 bug 钉一条断言**:浏览器可见的 bug 钉进所属 verify 脚本的回归节(一钉一行 report);数据层 bug 钉进对应 spec(先例:res-close 误判钉在 webserver 桥 suite——纯 Node 秒级复现,不再需要 12s 浏览器哨兵作唯一防线)。
-- **fixture 全绿不算完,真 host 也要过**:fixture 短路的恰是 wire 承载链(node:http 桥 close 语义、真网络时序),两次实证 bug 都藏在那里。改动触及连接/桥/handler/SSE 的,`verify-session-real` 必跑。
+- **修一个 bug 钉一条断言**:浏览器可见的 bug 钉进所属浏览器 spec(smoke 或 e2e 场景);数据层 bug 钉进对应 spec(先例:res-close 误判钉在 webserver 桥 suite——纯 Node 秒级复现,不再需要 12s 浏览器哨兵作唯一防线)。
+- **fixture 全绿不算完,真 wire 也要过**:fixture 短路的恰是 wire 承载链(node:http 桥 close 语义、真网络时序),两次实证 bug 都藏在那里。改动触及连接/桥/handler/SSE 的,浏览器车道(`pnpm run test:web`)必跑——其无密钥 e2e 场景驱动真实 HTTP/SSE 承载,带密钥的真 host smoke 仍是真模型侧的补充。
- 落盘代码即答案的对表工作流:行为改动落盘打红既有用例时,当场对表校准(改测试还是改代码以 RFC/契约为裁),不留悬红。
## Consequences
diff --git a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.i18n.yaml b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.i18n.yaml
index 8bd4745529..46073fc52b 100644
--- a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.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
-2026-07-21-doc-sync-through-gate-scheduler.md: b7e41ba4aeac8ea03c706acadd481eee26abd5c2
-2026-07-21-doc-sync-through-gate-scheduler.zh.md: 56699747b1ba97fd90f7d53ab0deebc73ac775ef
+2026-07-21-doc-sync-through-gate-scheduler.md: d66d9dc75ee4e8268d55e344a53c51c0bcf5f4d4
+2026-07-21-doc-sync-through-gate-scheduler.zh.md: 8c4c4595c2bc6e9439ed24cda1e70ae5a5ccd146
diff --git a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.md b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.md
index b7e41ba4ae..d66d9dc75e 100644
--- a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.md
+++ b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.md
@@ -10,7 +10,7 @@ English | [中文](2026-07-21-doc-sync-through-gate-scheduler.zh.md)
## Decision
-`doc-sync` in `package.json` delegates to the existing bounded scheduler — `tsx scripts/run-gates.ts doc-sync` — like the `check:ci:*` scripts ([parallel gate scheduling](2026-07-06-parallel-pre-push-gates.md), [parallel GitHub CI gates](2026-07-06-parallel-github-ci-gates.md)). The `doc-sync` mode expands to exactly `docSyncLeafGates()`, making the leaf list in `run-gates.ts` the single source of truth for the member set. The local mode caps default concurrency at four workers because several doc gates each build a full `ts.Program`; `DSH_GATE_CONCURRENCY` still overrides.
+`doc-sync` in `package.json` delegates to the existing bounded scheduler — `tsx scripts/run-gates.ts doc-sync` — like the `check:ci:*` scripts ([parallel gate scheduling](2026-07-06-parallel-pre-push-gates.md), [current CI topology](2026-07-22-evidence-based-larger-hosted-runners.md)). The `doc-sync` mode expands to exactly `docSyncLeafGates()`, making the leaf list in `run-gates.ts` the single source of truth for the member set. The local mode caps default concurrency at four workers because several doc gates each build a full `ts.Program`; `DSH_GATE_CONCURRENCY` still overrides.
`docSyncLeafGates` includes `verify-cordis-api`, so relevant local documentation checks and CI gate the generated runtime API catalog alongside the other generated docs.
diff --git a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.zh.md b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.zh.md
index 56699747b1..8c4c4595c2 100644
--- a/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.zh.md
+++ b/.agents/notes/implemented/process/2026-07-21-doc-sync-through-gate-scheduler.zh.md
@@ -10,7 +10,7 @@ Status: implemented
## 决策
-`package.json` 中的 `doc-sync` 委托给既有的有界调度器——`tsx scripts/run-gates.ts doc-sync`——与各 `check:ci:*` 脚本的做法一致([并行门禁调度](2026-07-06-parallel-pre-push-gates.md)、[并行 GitHub CI 门禁](2026-07-06-parallel-github-ci-gates.md))。`doc-sync` 模式恰好展开为 `docSyncLeafGates()`,使 `run-gates.ts` 里的叶子列表成为成员集合的唯一真源。本地模式把默认并发上限设为四个 worker,因为多个文档门禁各自要构建完整的 `ts.Program`;`DSH_GATE_CONCURRENCY` 仍可覆盖。
+`package.json` 中的 `doc-sync` 委托给既有的有界调度器——`tsx scripts/run-gates.ts doc-sync`——与各 `check:ci:*` 脚本的做法一致([并行门禁调度](2026-07-06-parallel-pre-push-gates.md)、[当前 CI 拓扑](2026-07-22-evidence-based-larger-hosted-runners.md))。`doc-sync` 模式恰好展开为 `docSyncLeafGates()`,使 `run-gates.ts` 里的叶子列表成为成员集合的唯一真源。本地模式把默认并发上限设为四个 worker,因为多个文档门禁各自要构建完整的 `ts.Program`;`DSH_GATE_CONCURRENCY` 仍可覆盖。
`docSyncLeafGates` 包含 `verify-cordis-api`,因此相关的本地文档检查与 CI 会同其他生成文档一起把关生成的运行时 API 目录。
diff --git a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.i18n.yaml b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.i18n.yaml
index aa0516648b..1a6a99d648 100644
--- a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.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
-2026-07-21-serial-cross-platform-ci-reference.md: b795a0aff62c20967d2c85429c0c6115c1b9585d
-2026-07-21-serial-cross-platform-ci-reference.zh.md: 223fd9cf20a1d8228cb0c6b1b2f3f95644becae6
+2026-07-21-serial-cross-platform-ci-reference.md: 3c0ae200d7dbd5b04eae6db2d6628dccc72103bf
+2026-07-21-serial-cross-platform-ci-reference.zh.md: 5c159e12739d68e0baed72aaa08331072e2c3601
diff --git a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.md b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.md
index b795a0aff6..3c0ae200d7 100644
--- a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.md
+++ b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.md
@@ -18,6 +18,10 @@ Reviewers also need a direct answer to a simpler question: what happens when the
Each reference job runs `pnpm run check:ci` without any shard selector. `DSH_GATE_CONCURRENCY=1` makes the top-level aggregate execute one ready gate at a time; coverage, snapshot replay, built-bin smoke, and publication validation also receive worker counts of one. The three operating-system jobs may run beside one another, but each host's repository gates are serial and complete. Linux installs bubblewrap before replaying snapshots, and Windows enables Developer Mode before installing the symlinked workspace.
+Platform ownership remains explicit inside that complete aggregate. `pty-local` supports Linux and macOS and therefore owns its unit and per-file coverage contract on POSIX rather than loading a backend that rejects `win32`; the Windows run still executes every portable package. Portable fixtures derive native paths through `node:path`, compare canonical identities with the same native realpath implementation as production, and use filenames legal on every host. ACP snapshot runs also pass both JavaScript and native realpath spellings of their generated cwd to the normalizer, which replaces aliases longest-first so Windows short and long paths cannot churn shared fixtures.
+
+The macOS reference runs the ordinary Vitest project in forked processes. Node 24 on macOS arm64 has aborted in its CJS lexer from a worker thread; the process boundary contains that external runtime failure without removing any test from the aggregate, while Linux and Windows retain the lower-overhead thread pool. Repository-owned races are fixed at their observation boundaries: dev bundle polling stages each candidate table, graph, and watch-baseline map before publishing a rescan, and a missing bundle remains dirty until a successful content hash. PTY readiness retains a prompt candidate while polling checks foreground ownership; the ordinary silence bound covers inherited markers from interactive children. The live-link package-manager e2e preserves the workflow-prepared Corepack home and pnpm metadata/store caches while isolating the other managers' mutable caches, so it does not discard reusable package-manager state before the install.
+
Master reference jobs are diagnostic and do not participate in the pull request's required `all checks passed` result. A pull request runs only its required jobs; a master push runs only the three serial references. Performance is evaluated from completed hosted-job timestamps and reported as a measurement; it is not encoded as a `timeout-minutes` value.
The portable reference uses GitHub's standard `ubuntu-latest`, `macos-latest`, and `windows-2025` labels. Required pull-request jobs use the same portable Linux and Windows capacity under the [required-CI decision](2026-07-23-portable-required-pull-request-ci.md). Higher-core hosted runners remain manual benchmarks because a correctness path must remain runnable without repository-external runner configuration.
@@ -36,4 +40,6 @@ The workflow contains duplicated setup steps and a master reference run can take
The reference may expose platform failures that the optimized blocking set does not yet claim to support, especially on Windows. Such a failure is evidence about current cross-platform behavior rather than a reason to weaken or silently skip the aggregate.
+The explicit `pty-local` ownership boundary means Windows does not claim coverage for a backend it cannot load, and forked macOS unit workers cost more process startup time. In return, every supported surface has an honest platform oracle, a native runtime abort cannot erase the rest of the unit result, and timing-sensitive observers start from state established before callers can mutate it.
+
Removing strict duration timeouts means a latency regression is observed rather than automatically cancelled. Hosted measurements must therefore accompany performance changes, while the completed logs retain the information needed to optimize the slow lane.
diff --git a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.zh.md b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.zh.md
index 223fd9cf20..5c159e1273 100644
--- a/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.zh.md
+++ b/.agents/notes/implemented/process/2026-07-21-serial-cross-platform-ci-reference.zh.md
@@ -18,6 +18,10 @@ Status: implemented
每个参考作业均在不设置任何分片选择器的情况下运行 `pnpm run check:ci`。`DSH_GATE_CONCURRENCY=1` 使顶层聚合每次只执行一个已经就绪的门禁;覆盖率、快照回放、built-bin 冒烟测试和发布验证的并发数也设为 1。三种操作系统的作业可以彼此并行,但每台主机上的仓库门禁都串行运行且完整执行。Linux 在回放快照前安装 bubblewrap,Windows 则在安装采用符号链接的工作区前启用开发人员模式。
+该完整聚合流程仍明确划分平台归属。`pty-local` 支持 Linux 与 macOS,因此其单元测试和逐文件覆盖率契约由 POSIX 平台负责,而不会在 Windows 上加载一个明确拒绝 `win32` 的后端;Windows 仍会执行所有可移植包(package)。可移植 fixture(测试前置数据)通过 `node:path` 派生原生路径,使用与生产代码相同的原生 realpath 实现比较规范化后的路径标识,并采用所有宿主机均允许的文件名。ACP(Agent Client Protocol)快照运行还会把生成的 cwd 分别通过 realpath 的 JavaScript 实现与原生实现得到的两种表示一并传给规范化器;规范化器按长度从长到短替换这些别名,避免 Windows 的短路径与长路径表示差异导致共享 fixture 反复变化。
+
+macOS 参考流程使用 fork 进程运行常规 Vitest 项目。macOS arm64 上的 Node 24 曾在工作线程中执行 CJS 词法分析器时异常终止;进程边界能够隔离这一外部运行时故障,且无需从聚合流程中删除任何测试,而 Linux 与 Windows 仍使用开销更低的线程池。仓库自身引入的竞态均在相应的观测边界修复:开发构建产物的轮询逻辑每次发布重新扫描结果前,都会先暂存候选表、候选图和候选监视基线映射;构建产物缺失后会一直保持脏状态,直到成功计算内容哈希。PTY 就绪检测会在轮询检查前台进程组归属期间保留提示符候选项;常规静默时限也适用于交互式子进程继承提示符标记的情况。实时链接场景下的包管理器 e2e 会保留由工作流预先准备的 Corepack 主目录、pnpm 元数据缓存和 store 缓存,同时隔离其他包管理器的可变缓存,因此不会在安装前丢弃可复用的包管理器状态。
+
master 分支的参考作业仅用于诊断,不参与拉取请求所要求的 `all checks passed` 结果。拉取请求只运行其必需作业;向 master 推送时只运行三个串行参考作业。系统根据已完成托管作业的时间戳评估性能,并将其报告为测量结果,而不是写成 `timeout-minutes` 值。
可移植的参考流程使用 GitHub 标准的 `ubuntu-latest`、`macos-latest` 和 `windows-2025` 标签。依据[必需 CI 决策](2026-07-23-portable-required-pull-request-ci.md),拉取请求必需作业使用相同的可移植 Linux 和 Windows 容量。更高核心数的托管运行器仍仅用于手动基准测试,因为正确性路径必须无需仓库外部的运行器配置即可运行。
@@ -36,4 +40,6 @@ master 分支的参考作业仅用于诊断,不参与拉取请求所要求的
参考流程可能暴露某些平台上的故障,而优化后的阻塞门禁集合尚未声明支持这些平台,Windows 尤其如此。这类失败反映了当前的跨平台行为,不应成为削弱或静默跳过该聚合流程的理由。
+明确的 `pty-local` 归属边界意味着 Windows 不会声称覆盖一个无法加载的后端,而 macOS 采用 fork 的单元测试工作进程会增加进程启动开销。这些代价换来的是:支持范围内的每项功能都有能够如实反映对应平台行为的判据,原生运行时异常终止不会抹掉其余单元测试结果,各项对时序敏感的观测逻辑也都会以调用方有机会修改状态前已建立的状态作为起点。
+
移除严格的时长超时后,系统会观测到延迟回归,而不是在发生回归时自动取消运行。因此,性能改动必须附带托管环境测量结果,已完成的日志则保留优化最慢通道所需的信息。
diff --git a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.i18n.yaml b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.i18n.yaml
index 9d87cb9ad3..3d8e6fc395 100644
--- a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.i18n.yaml
+++ b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.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
-2026-07-22-evidence-based-larger-hosted-runners.md: aaeab4ed9ae9687598f9f1d4a862120405697672
-2026-07-22-evidence-based-larger-hosted-runners.zh.md: 72b69c85908990a9f35b60f4c0a2ce213f9c8134
+2026-07-22-evidence-based-larger-hosted-runners.md: fe11e6929545923d27fbf41f5a39f7dd2b9c3fbf
+2026-07-22-evidence-based-larger-hosted-runners.zh.md: 47879284532a537cbe7e78aa2c495c4ef0be26c4
diff --git a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.md b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.md
index aaeab4ed9a..fe11e69295 100644
--- a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.md
+++ b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.md
@@ -20,6 +20,10 @@ The former gate-level and coarse primary shard jobs are absent from the workflow
Linux primary work uses three independent 32-core jobs. Coverage runs alone with its own worker bound, and the static scheduler runs alone so its result has no post-build consumer tail. After static gates finish, that job publishes its emitted `apps/*/lib`, `packages/*/*/lib`, and `vendor/*/lib` tree as a run-scoped artifact. The third job restores that exact tree, then starts lint, Node 24 runtime compatibility, build-backed snapshots, and all artifact consumers without repeating the build. Generated NodeNext consumer directories are excluded from ESLint discovery because the artifact check removes them while these processes overlap. The pnpm store and ESLint cache are restored without putting cache uploads on the pull-request critical path. Performance reports use each job's `startedAt` to `completedAt` interval; runner queue delay is capacity evidence, not repository execution time.
+The gate dependencies remain explicit. Coverage consumes source and does not wait for build. Documentation typechecking builds its complete project-reference graph once. Snapshot replay and publication consumers wait for emitted output, while Node-version compatibility jobs exercise runtime-sensitive source loading without repeating the primary source-graph typecheck. PTY and subprocess suites keep their bounded inner concurrency rather than inheriting the runner's core count.
+
+The artifact boundary remains explicit. `scripts/publint-all.ts` calls publint's supported API against an in-memory publication view formed from each manifest's declared files plus npm's mandatory metadata, avoiding one package-manager pack process per package. `scripts/verify-built-package-invariants.mjs` stages the declared `lib/` files below the real package and imports its compiled self-reference through plain Node and Cordis Loader normalization; a runtime chunk omitted from the publication contract still fails.
+
Windows shares one 32-core setup across the blocking build and production site plus observational built-artifact contracts. Linux owns the duplicate lint, coverage, and snapshot inventories because running those observational copies on Windows extends the paid critical path without adding a blocking platform claim.
An [exact-head all-size benchmark](https://github.com/deepseek-harness/deepseek-harness/actions/runs/29908491351) ran the complete unsharded primary Node aggregate on every Linux pool before the eager-build correction:
@@ -54,6 +58,10 @@ Complete serial Linux, macOS, and Windows references run only when `master` move
**Keep the former gate-level shard topology as a manual reference.** A dormant second topology kept hundreds of workflow lines, selector modules, and scenario-partition behavior alive. The all-size and serial suites provide timing and completeness controls without preserving production code that no required job exercises.
+**Return to package-manager packing in each publication validator.** Rejected because it repeats a package-manager subprocess for every package. The manifest-derived publication view and staged compiled self-reference preserve the published-file contract with one in-process inventory.
+
+**Build before coverage or typecheck on every Node version.** Rejected because coverage is source-only and compiler analysis is not runtime-specific. Build-backed consumers still wait for emitted output, and compatibility jobs exercise the runtime-sensitive paths on every advertised Node line.
+
**Use the 64-core pool for the complete primary aggregate.** Its sampled active time was three seconds lower than the 96-core result because hosted setup was nine seconds faster, but its repository gates were 5.72 seconds slower. The benchmark suite retains both pools because a sustained image or pricing change can reverse the comparison.
**Keep build behind typecheck.** This orders independent compiler invocations and turns snapshot replay into a three-stage critical chain. Build output has its own success dependency, so only snapshot and publication consumers wait for it.
diff --git a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.zh.md b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.zh.md
index 72b69c8590..4787928453 100644
--- a/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.zh.md
+++ b/.agents/notes/implemented/process/2026-07-22-evidence-based-larger-hosted-runners.zh.md
@@ -20,6 +20,10 @@ Status: implemented
Linux 主流程使用 3 个相互独立的 32 核作业。覆盖率单独运行,并设有自己的工作线程上限;静态调度器也单独运行,因此构建后的消费方不会拖延其结果。静态门禁完成后,该作业将其生成的 `apps/*/lib`、`packages/*/*/lib` 和 `vendor/*/lib` 目录树作为仅供本次运行使用的产物发布。第三个作业恢复完全相同的目录树,再让 lint、Node 24 运行时兼容性、依赖构建产物的快照和所有产物消费方基于构建完成后的工作树启动,而不重复构建。生成的 NodeNext 消费方目录不会纳入 ESLint 的文件发现范围,因为这些进程重叠执行时,产物检查会删除这些目录。pnpm store 和 ESLint 缓存会得到恢复,但缓存上传不会进入拉取请求关键路径。性能报告采用每个作业从 `startedAt` 到 `completedAt` 的区间;运行器排队延迟是容量证据,而非仓库执行时间。
+门禁依赖关系保持显式。覆盖率消费源码,不等待构建。文档类型检查只构建一次完整的 project-reference 图。快照回放和发布消费方等待生成的输出,而 Node 版本兼容性作业会验证对运行时敏感的源码加载,且不重复主源码项目图的类型检查。PTY 和子进程套件继续使用自身有界的内部并发,不继承运行器的核心数。
+
+产物边界保持显式。`scripts/publint-all.ts` 对内存中的发布视图调用 publint 支持的 API;该视图由每个 manifest(元数据清单)声明的文件和 npm 强制要求的元数据组成,从而避免为每个包启动一次包管理器 pack 进程。`scripts/verify-built-package-invariants.mjs` 将已声明的 `lib/` 文件暂存到真实包下,并通过普通 Node 和 Cordis Loader 规范化导入其已编译的自身引用;发布契约只要遗漏一个运行时分片,检查仍会失败。
+
Windows 以一次 32 核环境设置同时承载阻塞性构建、生产网站和观测性的构建产物契约。重复的 lint、覆盖率和快照清单由 Linux 承担,因为在 Windows 上运行这些观测性副本会延长付费关键路径,却不会新增任何阻塞性平台契约。
一次[分支头精确的全规格基准测试](https://github.com/deepseek-harness/deepseek-harness/actions/runs/29908491351)在修正构建尽早启动逻辑前,对每种 Linux 池都运行了完整且未分片的主 Node 聚合流程:
@@ -54,6 +58,10 @@ Windows 仓库工作在超过 16 核后收益很小,但 32 核池可以让完
**将原有的门禁级分片拓扑保留为手动参考。** 一套闲置的第二拓扑会让数百行工作流、选择器模块和场景分区行为继续存活。全规格和串行套件无需保留任何必需作业都不执行的生产代码,也能提供计时与完整性对照。
+**在每个发布校验器中恢复使用包管理器打包。** 不予采用,因为这会为每个包重复启动一个包管理器子进程。根据 manifest 构建的发布视图和已暂存的编译后自身引用,只需一份进程内清单即可保留发布文件契约。
+
+**在每个 Node 版本上先构建,再运行覆盖率或类型检查。** 不予采用,因为覆盖率只消费源码,编译器分析也不依赖运行时。依赖构建产物的消费方仍等待生成的输出,兼容性作业则在每个已声明支持的 Node 版本上验证对运行时敏感的路径。
+
**使用 64 核池运行完整主聚合流程。** 由于托管设置快了 9 秒,其采样活动耗时比 96 核结果少 3 秒,但仓库门禁慢了 5.72 秒。基准测试套件保留两种规格,因为映像或定价的持续变化可能反转比较结果。
**让构建继续等待类型检查。** 此方案会给相互独立的编译器调用排定先后顺序,并把快照回放变成 3 阶段关键链。构建输出本身有独立的成功依赖关系,因此只有快照和发布消费方需要等待它。
diff --git a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.i18n.yaml b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.i18n.yaml
index bf29a5ddca..ec8fddb9f6 100644
--- a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.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
-2026-06-19-drop-mutable-session-summary.md: 80fe043e365352b17d2a5b3efa1ab8d396d311c4
-2026-06-19-drop-mutable-session-summary.zh.md: e875d96d003367dbe8a671e6057a6b4a6a44ca74
+2026-06-19-drop-mutable-session-summary.md: f87378a1c3737950eb536be8e3f6776586eb8a01
+2026-06-19-drop-mutable-session-summary.zh.md: 05b6711b71ef87602b46706cd4340a10453e66ad
diff --git a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.md b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.md
index 80fe043e36..f87378a1c3 100644
--- a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.md
+++ b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.md
@@ -12,8 +12,8 @@ The summary was designed for a future session picker (recency ordering via `upda
- `SessionPersistence.update()` has **zero production callers** (every `.update(` hit is `createHash().update()` or a test).
- `firstPrompt` is **never read** anywhere in production.
-- `title` *is* read in the ACP bridge — but from a tool-call **presenter** (`present.title`), never from stored session metadata.
-- `updatedAt` has **no consumer**: the only production caller of `list()` reads `meta.cwd` (a `SessionHeader` field) to validate a workspace on `session/load`; resume reads `createdAt`/`cwd`/`parentSession` — all header fields.
+- Session titles come from durable `session/title` events, while tool-card titles come from tool presenters; neither reads mutable session metadata.
+- Persistence-list consumers use immutable header identity, creation, lineage, and cwd fields. Recency and previews derive from the log rather than an `updatedAt` summary.
- Decisively: the live `Session.header` was already typed `SessionHeader`, not `SessionMeta` — the summary never existed on the live session object; it lived only in the persistence layer, written and read by nothing but its own contract test.
## Decision
diff --git a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.zh.md b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.zh.md
index e875d96d00..05b6711b71 100644
--- a/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.zh.md
+++ b/.agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.zh.md
@@ -12,8 +12,8 @@ Status: implemented
- `SessionPersistence.update()` **零个生产调用方**(所有 `.update(` 匹配都是 `createHash().update()` 或测试代码)。
- `firstPrompt` 在生产代码中**从未被读取**。
-- `title` *确实*在 ACP(Agent Client Protocol)桥接层被读取过,但读的是工具调用的 **presenter**(`present.title`),从未读取存储的会话元数据。
-- `updatedAt` **没有消费方**:`list()` 唯一的生产调用方读取的是 `meta.cwd`(`SessionHeader` 字段),用于在 `session/load` 时校验工作区;恢复会话读取的是 `createdAt`/`cwd`/`parentSession`——全是 header 字段。
+- 会话标题来自持久的 `session/title` 事件,工具卡片标题来自工具 presenter;二者都不读取可变的会话元数据。
+- 持久化列表的消费方使用不可变 header 中的标识、创建、谱系和 cwd 字段。近期排序和预览派生自日志,而非某个 `updatedAt` 摘要。
- 决定性的一点:活跃的 `Session.header` 类型本来就是 `SessionHeader` 而非 `SessionMeta`——摘要从未存在于活跃会话对象上;它只存在于持久化层,除了自身的契约测试外无人写入、无人读取。
## 决策
diff --git a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.i18n.yaml b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.i18n.yaml
index e30531e02d..b40c734b2c 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.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
-2026-06-20-collapse-trace-only-session-events.md: fce5c48ef6fcb1abc6e2fbb95dc7e83d22956660
-2026-06-20-collapse-trace-only-session-events.zh.md: 5be2d77644c7b4084b892013a2c1e5b97a333d74
+2026-06-20-collapse-trace-only-session-events.md: c77062c3cd44286b43c175702c35b36f9cc31da6
+2026-06-20-collapse-trace-only-session-events.zh.md: b232b3fb60822e60b1f5767066db42b0228269a8
diff --git a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.md b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.md
index fce5c48ef6..c77062c3cd 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.md
@@ -6,7 +6,7 @@ English | [中文](2026-06-20-collapse-trace-only-session-events.zh.md)
## Problem
-The session event vocabulary includes first-class events that are not part of replayable conversation history and have little or no production consumption. `usage` is already present as a model stream chunk before the loop also appends a separate `usage` event. `error` duplicates the `turn/end { kind: 'error', message, code }` reason for loop failures; ACP settlement reads the turn-end reason, ACP rendering ignores the `error` event, and `deriveMessages()` skips it.
+The session event vocabulary includes first-class events that are not part of replayable conversation history and have little or no production consumption. `usage` is already present as a model stream chunk before the loop also appends a separate `usage` event. `error` duplicates the `turn/end { kind: 'error', message, code }` reason for loop failures; ACP settlement reads the turn-end reason, while message and UI projections skip the standalone `error` event.
These events make the canonical transcript look more useful as telemetry than it currently is. They add event variants, invariants, tests, snapshots, and persistence cases, but they are not load-bearing as separate records. The facts they carry can still be useful: token usage should remain available for accounting, and an error's step number should not silently disappear. The simplification is to fold those facts into nearby events consumers already must understand, not to record less information.
diff --git a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.zh.md b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.zh.md
index 5be2d77644..b232b3fb60 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.zh.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-会话事件词汇中包含一些一等事件,它们不属于可回放的对话历史,在生产环境中几乎没有消费方。`usage` 已经作为模型流分片存在,之后循环又追加了一个独立的 `usage` 事件。`error` 与 `turn/end { kind: 'error', message, code }` 中的循环失败原因重复;ACP(Agent Client Protocol)结算读取轮次结束原因,ACP 渲染忽略 `error` 事件,`deriveMessages()` 也跳过它。
+会话事件词汇中包含一些一等事件,它们不属于可回放的对话历史,在生产环境中几乎没有消费方。`usage` 已经作为模型流分片存在,之后循环又追加了一个独立的 `usage` 事件。`error` 与 `turn/end { kind: 'error', message, code }` 中的循环失败原因重复;ACP(Agent Client Protocol)结算读取轮次结束原因,而消息投影和 UI 投影都会跳过独立的 `error` 事件。
这些事件让规范的 transcript(文本记录)看起来比实际更像遥测数据。它们增加了事件变体、不变式、测试、快照和持久化用例,但作为独立记录并不承载实际功能。它们携带的事实仍然有用:token 用量应当保留以供计费,错误的步骤编号也不应悄然消失。简化的方式是将这些事实折叠进消费方本已必须理解的邻近事件,而非减少记录的信息量。
diff --git a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.i18n.yaml b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.i18n.yaml
index ed0b63a28d..872b3fc588 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.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
-2026-06-20-prune-dead-seam-methods.md: 70596d908bb1d7559e876d93bce0e25874ce1ff0
-2026-06-20-prune-dead-seam-methods.zh.md: 1fbc7dbf2f7abb5b17b1df10ef06e490b11335a4
+2026-06-20-prune-dead-seam-methods.md: 91a18b4c327d2f3cb636d6d9e0c26e4246a6ffd2
+2026-06-20-prune-dead-seam-methods.zh.md: b962bc45052b723eefa533b98fe85f4508d6b6a7
diff --git a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.md b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.md
index 70596d908b..91a18b4c32 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.md
@@ -12,7 +12,7 @@ A capability seam ([interface / implementation / consumer](../architecture/2026-
### `SessionPersistence.has()` and `.delete()`
-The abstract service declared its operations beyond create/append: `load`, `list`, `has`, `delete`. Production consumers of `ctx.sessionPersistence` use only two: the agent-loop resume path calls `load()` ([packages/core/agent-loop/src/index.ts:176](../../../../packages/core/agent-loop/src/index.ts)), and the ACP bridge calls `list()` for `session/list` ([packages/ui/acp/src/index.ts:494](../../../../packages/ui/acp/src/index.ts)). Grepping every `sessionPersistence.*` / `persistence.*` use across `packages/*/src` and `examples/` finds no `has(` and no `delete(` on the service. The `.has(`/`.delete(` calls in `packages/ui/acp/src/index.ts` are on the in-memory `SessionStore` and a local `Set` of loading ids, not persistence. The only callers of `has`/`delete` were the contract suites and per-backend specs.
+The abstract service declared its operations beyond create/append: `load`, `list`, `has`, `delete`. Production consumers use `load()` and `list()` for resume and session discovery, while no production caller uses persistence `has()` or `delete()`. The similarly named in-memory collection calls in protocol and UI code are unrelated. The only callers of persistence `has`/`delete` were the contract suites and per-backend specs.
`has()` was not just unused: it added a tracked-vs-untracked coordinator probe and a contract branch even though `loadStored(id)` already owns durable existence checks. `delete()` dragged the `deleteStored` backend hook that every backend had to implement. This is the [drop-mutable-session-summary](2026-06-19-drop-mutable-session-summary.md) pattern: a contract test exercised both, but no shipping code asks "is this session persisted?" or removes one.
@@ -33,7 +33,7 @@ Re-adding a seam method with a live consumer is cheap and better-designed than t
## Verification
-`has`/`delete`/`deleteStored` are gone from the persistence seam, impl, and contract suites with no new dead exports; the remaining operations (`create`/`append`/`load`/`list`) are untouched, with ACP `session/list` and crash-recovery behaving identically; and the seam README and `docs/architecture.md` list only the surviving methods.
+`has`/`delete`/`deleteStored` are gone from the persistence seam, impl, and contract suites with no new dead exports; the remaining operations (`create`/`append`/`load`/`list`) are untouched, with persistence-backed session queries and crash recovery behaving identically; and the seam README and `docs/architecture.md` list only the surviving methods.
## Consequences
diff --git a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.zh.md b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.zh.md
index 1fbc7dbf2f..b962bc4505 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.zh.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-prune-dead-seam-methods.zh.md
@@ -12,7 +12,7 @@ Status: implemented
### `SessionPersistence.has()` 与 `.delete()`
-该抽象服务在 create/append 之外声明了更多操作:`load`、`list`、`has`、`delete`。`ctx.sessionPersistence` 的生产消费方只用了两个:agent loop(智能体循环)的恢复路径调用 `load()`([packages/core/agent-loop/src/index.ts:176](../../../../packages/core/agent-loop/src/index.ts)),ACP(Agent Client Protocol)桥接层为 `session/list` 调用 `list()`([packages/ui/acp/src/index.ts:494](../../../../packages/ui/acp/src/index.ts))。在 `packages/*/src` 和 `examples/` 中 grep 所有 `sessionPersistence.*` / `persistence.*` 的使用,找不到对该服务的 `has(` 或 `delete(` 调用。`packages/ui/acp/src/index.ts` 中的 `.has(`/`.delete(` 调用作用于内存中的 `SessionStore` 和一个本地的 loading id `Set`,而非持久化。`has`/`delete` 的唯一调用者是契约测试套件和各后端的 spec。
+该抽象服务在 create/append 之外声明了更多操作:`load`、`list`、`has`、`delete`。生产消费方用 `load()` 和 `list()` 完成恢复与会话发现,而没有任何生产调用方使用持久化的 `has()` 或 `delete()`。协议和 UI 代码中名称相似的内存集合调用与此无关。持久化 `has`/`delete` 的唯一调用者是契约测试套件和各后端的 spec。
`has()` 不仅未被使用:在 `loadStored(id)` 已负责持久化存在性检查的情况下,它仍增加了协调器的已跟踪/未跟踪探测和一个契约分支。`delete()` 则拖入每个后端都必须实现的 `deleteStored` 后端钩子。这属于[删除可变会话 summary](2026-06-19-drop-mutable-session-summary.md) 的同类模式:契约测试覆盖了两者,但已发布代码从不会询问“这个会话是否已持久化?”或删除某个会话。
@@ -33,7 +33,7 @@ Status: implemented
## 验证
-`has`/`delete`/`deleteStored` 已从持久化 seam、实现和契约测试套件中移除,没有新增无用导出;剩余操作(`create`/`append`/`load`/`list`)未受影响,ACP `session/list` 和崩溃恢复行为完全一致;seam README 和 `docs/architecture.md` 仅列出存留的方法。
+`has`/`delete`/`deleteStored` 已从持久化 seam、实现和契约测试套件中移除,没有新增无用导出;剩余操作(`create`/`append`/`load`/`list`)未受影响,基于持久化的会话查询和崩溃恢复行为完全一致;seam README 和 `docs/architecture.md` 仅列出存留的方法。
## 后果
diff --git a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.i18n.yaml b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.i18n.yaml
index 7a5f7fdb48..f6cc14538d 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.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
-2026-06-20-public-agent-stop-surface.md: 81a21de30bfbc25688069efbffb21647889b1bdb
-2026-06-20-public-agent-stop-surface.zh.md: 1f5a9ea3c1c3dd6f598c258574a81e52e291f3b7
+2026-06-20-public-agent-stop-surface.md: e22c4389df18f3c9ca96763fc097eabefcc5b761
+2026-06-20-public-agent-stop-surface.zh.md: e2647b498a8c906579b4fd2b50f94d1c326fe784
diff --git a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.md b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.md
index 81a21de30b..e22c4389df 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.md
@@ -18,7 +18,7 @@ The extra surface area made the loop carry a public verb that is mostly a teardo
`cancel()` is the only public *stop* primitive on `Agent`. Lifecycle owners use `AgentHandle.dispose()` to stop and unregister an agent; non-owners use `cancel()` to abandon current and queued work. The implementation keeps a private turn cancellation holder, but it is not part of the plugin-facing `Agent` contract.
-`whenIdle()` is **retained** as the public quiescence-observation primitive (resolve once the agent settles out of `running`, resolve immediately when already idle, await the loop exit when disposed). It is not a stop verb; it is how a non-owner observes the stop *completing* without disposing the agent. Its live consumers are ACP and agent tests that await settlement through this public seam (`packages/ui/acp/tests`, `packages/core/agent-loop/tests`); the production ACP bridge owns its agents and tears them down through `AgentHandle.dispose()`, so `packages/ui/acp/src` itself has no `whenIdle()` call.
+`whenIdle()` is **retained** as the public quiescence-observation primitive (resolve once the agent settles out of `running`, resolve immediately when already idle, await the loop exit when disposed). It is not a stop verb; it is how a non-owner observes the stop *completing* without disposing the agent. Its live consumers are ACP and agent tests that await settlement through this public seam (`packages/acp/acp/tests`, `packages/core/agent-loop/tests`); the production ACP bridge owns its agents and tears them down through `AgentHandle.dispose()`, so `packages/acp/acp/src` itself has no `whenIdle()` call.
Public `abort()` is absent, and the disposer remains async and waits for the loop to stop. Tests exercise cancellation through the public typed cause and explicit signal seams rather than reaching into the holder.
diff --git a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.zh.md b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.zh.md
index 1f5a9ea3c1..e2647b498a 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.zh.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.zh.md
@@ -18,7 +18,7 @@ Status: implemented
`cancel()` 是 `Agent` 上唯一的公共*停止*原语。生命周期拥有者使用 `AgentHandle.dispose()` 停止并注销 agent;非拥有者使用 `cancel()` 放弃当前和已排队工作。实现保留一个私有轮次取消 holder,但它不属于面向插件的 `Agent` 契约。
-`whenIdle()` **保留**为公开的完全停稳观测原语(agent 从 `running` 状态稳定后 resolve,已处于 idle 时立即 resolve,dispose 后等待循环退出)。它不是停止动词;它是非所有者在不 dispose agent 的前提下观测停止*完成*的方式。它的活跃消费方是 ACP 和通过此公开 seam 等待结算的 agent 测试(`packages/ui/acp/tests`、`packages/core/agent-loop/tests`);生产环境的 ACP 桥接层拥有其 agent 并通过 `AgentHandle.dispose()` 销毁它们,因此 `packages/ui/acp/src` 本身没有 `whenIdle()` 调用。
+`whenIdle()` **保留**为公开的完全停稳观测原语(agent 从 `running` 状态稳定后 resolve,已处于 idle 时立即 resolve,dispose 后等待循环退出)。它不是停止动词;它是非所有者在不 dispose agent 的前提下观测停止*完成*的方式。它的活跃消费方是 ACP 和通过此公开 seam 等待结算的 agent 测试(`packages/acp/acp/tests`、`packages/core/agent-loop/tests`);生产环境的 ACP 桥接层拥有其 agent 并通过 `AgentHandle.dispose()` 销毁它们,因此 `packages/acp/acp/src` 本身没有 `whenIdle()` 调用。
公共 `abort()` 已不存在,disposer 仍为异步并等待循环停止。测试通过公共类型化原因和显式 signal seam 验证取消,而不会伸入 holder 内部。
diff --git a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.i18n.yaml b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.i18n.yaml
index 9e90a3a343..cfed1ac086 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.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
-2026-06-20-remove-agent-boundary-mirror-events.md: 8c5bb74f2347fe0269cbb9c6504137de761ab919
-2026-06-20-remove-agent-boundary-mirror-events.zh.md: 1bb166b7f4453b4f0605e2b1b9120bc5a05266b0
+2026-06-20-remove-agent-boundary-mirror-events.md: 31acbb122cf56adfdcfcbce602d09a35f7f13e17
+2026-06-20-remove-agent-boundary-mirror-events.zh.md: f386e8d1ccfc2df094645a3cf9ae9524091d2a4e
diff --git a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.md b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.md
index 8c5bb74f23..31acbb122c 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.md
@@ -15,7 +15,7 @@ English | [中文](2026-06-20-remove-agent-boundary-mirror-events.zh.md)
## Problem
-The loop records the canonical transcript in `SessionEvent` and also emitted a parallel set of live `agent/*` boundary mirror events: `agent/turn-start`, `agent/turn-end`, `agent/step-start`, and `agent/step-end`. The mirrors made consumers choose between two sources of truth for the SAME durable fact. ACP already chose the session log for the editor-facing transcript because it is the one durable, replayable record; consuming a live mirror would require reconciling its timing with the boundary already stored in that log. The stdio UI was the only production consumer that still rendered turn boundaries from the mirror events; it already rendered tool calls and results from `session/event`.
+The loop records the canonical transcript in `SessionEvent` and also emitted a parallel set of live `agent/*` boundary mirror events: `agent/turn-start`, `agent/turn-end`, `agent/step-start`, and `agent/step-end`. The mirrors made consumers choose between two sources of truth for the SAME durable fact. ACP already chose the session log for prompt settlement and committed output because it is the one durable, replayable record; consuming a live mirror would require reconciling its timing with the boundary already stored in that log. The stdio UI was the only production consumer that still rendered turn boundaries from the mirror events; it already rendered tool calls and results from `session/event`.
This duplication is not free. Every lifecycle change had to update the session event, the mirror event, docs, invariants, tests, and snapshot expectations. The duplicate boundary events also made failure ordering subtle: a turn can be durably closed before a live `agent/turn-end` listener runs, so a post-boundary listener failure has no valid in-log position left and must be reported out of band.
diff --git a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.zh.md b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.zh.md
index 1bb166b7f4..f386e8d1cc 100644
--- a/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.zh.md
+++ b/.agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.zh.md
@@ -15,7 +15,7 @@ Status: implemented
## 问题
-循环在 `SessionEvent` 中记录规范 transcript(文本记录),同时还发出一组并行的实时 `agent/*` 边界镜像事件:`agent/turn-start`、`agent/turn-end`、`agent/step-start` 和 `agent/step-end`。这些镜像迫使消费方在同一持久事实的两个真源之间做选择。ACP(Agent Client Protocol)已经为面向编辑器的 transcript 选择会话日志,因为它是唯一持久、可重放的记录;消费实时镜像需要把它的时序与日志中已经存储的边界进行调和。stdio UI 是唯一仍从镜像事件渲染轮次边界的生产消费方;它已经从 `session/event` 渲染工具调用和结果。
+循环在 `SessionEvent` 中记录规范 transcript(文本记录),同时还发出一组并行的实时 `agent/*` 边界镜像事件:`agent/turn-start`、`agent/turn-end`、`agent/step-start` 和 `agent/step-end`。这些镜像迫使消费方在同一持久事实的两个真源之间做选择。ACP(Agent Client Protocol)已经为提示词结算和已提交输出选择会话日志,因为它是唯一持久、可重放的记录;消费实时镜像需要把它的时序与日志中已经存储的边界进行调和。stdio UI 是唯一仍从镜像事件渲染轮次边界的生产消费方;它已经从 `session/event` 渲染工具调用和结果。
这种重复并非零成本。每次生命周期变更都需要同时更新会话事件、镜像事件、文档、不变式、测试和快照预期。重复的边界事件还使失败排序变得微妙:一个轮次可能在实时 `agent/turn-end` 监听器运行之前就已被持久化关闭,因此边界之后的监听器失败在日志中已没有合法位置可以插入,只能带外上报。
diff --git a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.i18n.yaml
index 2926d785eb..3a0fde3d06 100644
--- a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.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
-2026-07-02-remove-stream-chunk-mirror.md: 62421bfd6032d3be6e942c00abefe17ec947c986
-2026-07-02-remove-stream-chunk-mirror.zh.md: 6e006a2c1930e47a651813252d3fd2c4471e6e9e
+2026-07-02-remove-stream-chunk-mirror.md: 8b26589a9e89f83d631fa98e801a8d3e08e105d0
+2026-07-02-remove-stream-chunk-mirror.zh.md: fcd8c53b8b1b2a5b81f9f929ffd9e06ff6128e45
diff --git a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.md b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.md
index 62421bfd60..5b8e4a3381 100644
--- a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.md
+++ b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.md
@@ -27,7 +27,7 @@ The premise the deferral hinged on is settled: chunk persistence is authoritativ
Remove `agent/stream-chunk` from the agent event taxonomy. The token stream is read off `session/event` as `assistant/chunk`, the same feed persistence and replay already use — `session/event` is the single live transcript stream (assistant chunks, turn/step boundaries, tool activity, todos).
-**Consumers.** The only production consumer that mattered — the ACP bridge (`dsh-acp`), the real editor-facing streaming surface — already renders `assistant/chunk` off `session/event`, never `agent/stream-chunk`, so it is unaffected. The stdio UI (`dsh-ui-stdio`, a disposable test REPL) was the sole live consumer; it already had a `session/event` listener (from the boundary migration), so its chunk rendering folded into that listener as an `assistant/chunk` case. Consolidating to one listener also removed a latent hazard: the `inReasoning` dim-SGR flag was previously shared across two separate listeners (`agent/stream-chunk` and `session/event`), so a chunk and a boundary racing on it had no defined order; a single listener over the append order makes the interleaving deterministic.
+**Consumers.** Persistence, replay, and interactive renderers consume the authoritative session stream directly. The [automation-only ACP bridge](2026-07-23-acp-automation-only-protocol.md) emits committed `assistant/message` text rather than raw chunks, so it needs neither event. No production consumer requires an `Agent`-first token mirror.
## Scope
diff --git a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.zh.md b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.zh.md
index 6e006a2c19..9357748884 100644
--- a/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-02-remove-stream-chunk-mirror.zh.md
@@ -27,7 +27,7 @@ ctx.emit('agent/stream-chunk', agent, turn, step, chunk) // ← the mirror
从 agent 事件分类体系中移除 `agent/stream-chunk`。token 流通过 `session/event` 以 `assistant/chunk` 的形式读取——持久化与回放已经使用的正是同一个序列。`session/event` 是唯一的实时 transcript(文本记录)流(assistant 分片、轮次/步骤边界、工具活动、todo)。
-**消费方。** 唯一重要的生产消费方——ACP(Agent Client Protocol)桥接(`dsh-acp`,面向编辑器的真实流式输出接口)——已经从 `session/event` 渲染 `assistant/chunk`,从未使用 `agent/stream-chunk`,因此不受影响。stdio UI(`dsh-ui-stdio`,一个一次性的测试 REPL)是唯一的实时消费方;它在边界迁移时已经有了 `session/event` 监听器,因此其分片渲染被折叠进该监听器作为 `assistant/chunk` 分支。合并为一个监听器还消除了一个潜在隐患:`inReasoning` dim-SGR 标志此前在两个独立监听器(`agent/stream-chunk` 和 `session/event`)之间共享,分片与边界在该标志上竞争时没有确定的顺序;单一监听器按追加顺序处理,使交错变为确定性的。
+**消费方。** 持久化、回放和交互式渲染器直接消费权威的会话流。[仅面向自动化的 ACP(Agent Client Protocol)桥接层](2026-07-23-acp-automation-only-protocol.md)发出已提交的 `assistant/message` 文本而非原始分片,因此两种事件它都不需要。没有生产消费方需要一个 `Agent` 优先的 token 镜像。
## 范围
diff --git a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.i18n.yaml
index 464ffc54d4..8947371a42 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.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
-2026-07-04-drop-image-content-block.md: 566803ab5b9f213b7dc87fcf779ed7a963d988fd
-2026-07-04-drop-image-content-block.zh.md: 667ddbc1b7bce2a276fe7528f46d3e43916bad83
+2026-07-04-drop-image-content-block.md: cdedf4bd5dfe60c72cea185d88b83d3b93928ff0
+2026-07-04-drop-image-content-block.zh.md: 683fd1cdb47e3fcd68ff601c0b0ce4b46f8b06d2
diff --git a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.md b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.md
index 566803ab5b..cdedf4bd5d 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.md
+++ b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.md
@@ -6,17 +6,17 @@ English | [中文](2026-07-04-drop-image-content-block.zh.md)
## Problem
-`ImageBlock` (`packages/llm/llm/src/types.ts`) had no production producer, and every consumer on every path DROPPED it: the deepseek adapter's serializer skipped image blocks (a documented MVP limitation), the pi-ai converter skipped them as unrepresentable, the ACP codec neither advertises image prompt capability nor forwarded image blocks outbound and REJECTS image prompt content inbound, and the compaction estimator charged a flat token constant and rendered `[image]`. An `ImageBlock` constructed then would silently vanish from the wire — the vocabulary advertised a capability no path honored, which is the silent-data-loss shape AGENTS.md's defensive patterns warn against. The only constructors anywhere were tests pinning the skip/drop/estimate branches.
+`ImageBlock` (`packages/llm/llm/src/types.ts`) had no production producer, and every consumer on every path DROPPED it: the DeepSeek adapter's serializer skipped image blocks (a documented MVP limitation), the pi-ai converter skipped them as unrepresentable, and the compaction estimator charged a flat token constant and rendered `[image]`. ACP independently rejected image prompt content. An `ImageBlock` constructed then would silently vanish from the provider wire — the vocabulary advertised a capability no path honored, which is the silent-data-loss shape AGENTS.md's defensive patterns warn against. The only constructors anywhere were tests pinning the skip/drop/estimate branches.
## Decision
-Remove `ImageBlock`, its map entry, and image-specific branches from adapters, ACP rendering, and compaction. Update the owning vocabulary docs and generated references in the same change. Unknown extension blocks still exercise default branches, and ACP continues to reject inbound image prompt content independently of the harness vocabulary.
+Remove `ImageBlock`, its map entry, and image-specific branches from adapters and compaction. Update the owning vocabulary docs and generated references in the same change. Unknown extension blocks still exercise default branches, and ACP continues to reject inbound image prompt content independently of the harness vocabulary.
## Alternatives considered
### Why not keep it?
-`ContentBlockMap` can reintroduce images when adapters, ACP, and compaction all support them. Keeping a core type whose only implementation is rejection would advertise an unusable surface; absence gives producers an immediate compile-time failure instead.
+`ContentBlockMap` can reintroduce images when adapters and compaction support them. ACP may remain a text-only automation protocol. Keeping a core type whose only implementation is rejection would advertise an unusable surface; absence gives producers an immediate compile-time failure instead.
The recorded fallback, had review landed on keeping the slot: keep `ImageBlock` but replace every silent skip with a loud rejection, and document that policy in the vocabulary — the silent drop was the one state with no defender. Review landed on removal; the fallback stands as the documented alternative should the slot ever return ahead of a full feature.
@@ -26,4 +26,4 @@ No harness `ImageBlock` is constructed outside Agent Note records. ACP's indepen
## Consequences
-Re-adding a core vocabulary type later touches several packages at once — but that coordinated change is the shape a real multimodal feature needs anyway (adapter mapping, ACP advertisement, compaction pricing), and none of it existed to preserve.
+Re-adding a core vocabulary type later touches several packages at once — but that coordinated change is the shape a real multimodal feature needs anyway (adapter mapping and compaction pricing), and none of it existed to preserve.
diff --git a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.zh.md b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.zh.md
index 667ddbc1b7..683fd1cdb4 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.zh.md
@@ -6,17 +6,17 @@ Status: implemented
## 问题
-`ImageBlock`(`packages/llm/llm/src/types.ts`)没有任何生产环境的生产者,而每条路径上的每个消费方都将其丢弃:deepseek 适配器的序列化器跳过 image 块(这是文档中注明的 MVP 限制);pi-ai 转换器因无法表示而跳过;ACP(Agent Client Protocol)编解码器既不宣告图像提示词能力、也不向外转发 image 块,并且会拒绝入站的图像提示词内容;压缩(compaction)估算器对其收取一个固定 token 常量并渲染为 `[image]`。此时构造的 `ImageBlock` 会在协议格式(wire format)上静默消失——词汇宣告了一种没有任何路径兑现的能力,这正是 AGENTS.md 防御性模式所警告的静默数据丢失形态。唯一的构造调用出现在测试中,用于覆盖 skip/drop/estimate 分支。
+`ImageBlock`(`packages/llm/llm/src/types.ts`)没有任何生产环境的生产者,而每条路径上的每个消费方都将其丢弃:DeepSeek 适配器的序列化器跳过 image 块(这是文档中注明的 MVP 限制);pi-ai 转换器因无法表示而跳过;压缩(compaction)估算器对其收取一个固定 token 常量并渲染为 `[image]`。ACP(Agent Client Protocol)独立地拒绝图像提示词内容。此时构造的 `ImageBlock` 会在提供方协议格式(wire format)上静默消失——词汇宣告了一种没有任何路径兑现的能力,这正是 AGENTS.md 防御性模式所警告的静默数据丢失形态。唯一的构造调用出现在测试中,用于覆盖 skip/drop/estimate 分支。
## 决策
-移除 `ImageBlock`、其 map 条目,以及适配器、ACP 渲染和压缩中的 image 专用分支。在同一个变更中更新所属的词汇文档与生成的引用。未知扩展块仍然覆盖默认分支,ACP 继续独立于 harness 词汇拒绝入站的图像提示词内容。
+移除 `ImageBlock`、其 map 条目,以及适配器和压缩中的 image 专用分支。在同一个变更中更新所属的词汇文档与生成的引用。未知扩展块仍然覆盖默认分支,ACP 继续独立于 harness 词汇拒绝入站的图像提示词内容。
## 曾考虑的替代方案
### 为什么不保留?
-当适配器、ACP 和压缩全部支持 image 时,`ContentBlockMap` 可以重新引入。保留一个唯一实现就是拒绝的核心类型,等于宣告一个不可用的对外服务接口;移除后,生产者会立即得到编译期错误。
+当适配器和压缩支持 image 时,`ContentBlockMap` 可以重新引入。ACP 可以继续作为纯文本的自动化协议。保留一个唯一实现就是拒绝的核心类型,等于宣告一个不可用的对外服务接口;移除后,生产者会立即得到编译期错误。
评审中记录的回退方案(假如评审决定保留该槽位):保留 `ImageBlock`,但将所有静默跳过替换为显式拒绝,并在词汇文档中记录该策略——静默丢弃是唯一没有辩护者的状态。评审最终决定移除;此回退方案作为文档化的替代方案保留,以备该槽位在完整功能就绪之前回归。
@@ -26,4 +26,4 @@ Status: implemented
## 后果
-日后重新添加核心词汇类型需要同时改动多个包(package)——但这种协调变更本就是真正的多模态功能所需的形态(适配器映射、ACP 能力宣告、压缩定价),而当前并不存在需要保留的实现。
+日后重新添加核心词汇类型需要同时改动多个包(package)——但这种协调变更本就是真正的多模态功能所需的形态(适配器映射与压缩定价),而当前并不存在需要保留的实现。
diff --git a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.i18n.yaml
deleted file mode 100644
index 3e67070db5..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-04-fold-stdio-ui-helper.md: 01165df19942e8e0adf84ede3a371f53ad12e464
-2026-07-04-fold-stdio-ui-helper.zh.md: 37b316283e96d7640e1e7af5fdbb05ab487aa6fa
diff --git a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.md b/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.md
deleted file mode 100644
index 01165df199..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.md
+++ /dev/null
@@ -1,30 +0,0 @@
-# Agent Note: Fold the stdio UI helper into the stdio app
-
-Status: implemented
-
-English | [中文](2026-07-04-fold-stdio-ui-helper.zh.md)
-
-The later [redundant-agent removal](2026-07-20-remove-stdio-and-echo-agents.md) supersedes this package-placement decision and removes the folded package, app, and line-oriented surface entirely.
-
-## Problem
-
-The readline UI was a whole package (`@deepseek-ai/dsh-ui-stdio` under `packages/support/`) whose only runtime importer was the app package `@deepseek-ai/dsh-stdio-demo`. The examples reach the readline UI by loading the app, never by composing the helper themselves; every other repo reference was mechanical or descriptive surface that existed BECAUSE the package boundary existed — manifest and tsconfig entries, generated module-graph rows, dependency-graph and README rows, and doc comments naming the package. The ui group README recorded the support placement rationale ("exists chiefly for the examples and the coverage gate — `ui/` is reserved for surfaces shipped as product"), which left a standing tension: a shipped product app depending on a support package documented as NOT product surface.
-
-The boundary bought package metadata, workspace and tsconfig references, module-graph rows, README entries, and publint surface for a helper that is not independently swappable: the stdio app's front-door cluster always includes the readline UI, and nothing else can meaningfully consume it.
-
-## Decision
-
-At the time, the helper moved into `@deepseek-ai/dsh-stdio` as the terminal-channel plugin. `createStdioChat`, its `StdioRuntime` test seam, and its unit tests moved with it, keeping EOF handling, rendering, disposal, and piped-vs-TTY behavior under the per-file coverage gate without hijacking process globals. The module kept the named `name`/`inject`/`Config`/`apply` export shape consumed by the app mount, while the then-current Echo and REPL Loader smokes proved the composed tree and the plugin-shape suite pinned explicit `unwrapExports` behavior. The superseding removal note above owns the current package and example state.
-
-The earlier support helper package was removed: its manifest, tsconfig references, module-graph rows, and README rows disappeared, while the remaining documentation described the in-package module.
-
-## Alternatives considered
-
-### Why not promote it to `ui/` instead?
-
-Promotion would have resolved the support-vs-product mismatch while keeping the boundary — the right call only if the readline UI were an independently swappable integration or had a second composer, and the consumer census said neither. The structured ACP bridge stays its own package because it is the product protocol surface with its own contract and snapshot tiers; the readline helper is scaffolding for one app's front door. Re-extraction stays cheap pre-release: if a second product app wants the readline UI, split it back out then, with that consumer shaping the package contract.
-
-## Consequences
-
-- The stdio app owns its whole front door; a leaf `cordis.yml` still loads one app package and nothing changed shape for the demos.
-- A future standalone terminal UI that wants the helper as a package reintroduces it with that second consumer, rather than the repo keeping a boundary for hypothetical reuse.
diff --git a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.zh.md b/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.zh.md
deleted file mode 100644
index 37b316283e..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-04-fold-stdio-ui-helper.zh.md
+++ /dev/null
@@ -1,30 +0,0 @@
-# Agent Note: 将 stdio UI 辅助模块折入 stdio 应用
-
-Status: implemented
-
-[English](2026-07-04-fold-stdio-ui-helper.md) | 中文
-
-后来的[冗余 agent(智能体)移除](2026-07-20-remove-stdio-and-echo-agents.md)取代了这项包放置决策,并完整移除合并后的包、应用和面向行的表面。
-
-## 问题
-
-readline UI 曾是一个完整的包(`packages/support/` 下的 `@deepseek-ai/dsh-ui-stdio`),其唯一的运行时导入方是应用包 `@deepseek-ai/dsh-stdio-demo`。示例通过加载应用来使用 readline UI,从不自行组合该辅助模块;仓库中所有其他引用都是因为包边界存在而存在的机械性或描述性表面:manifest(元数据清单)与 tsconfig 条目、生成的 module-graph 行、依赖图与 README 行,以及命名该包的文档注释。ui 组 README 记录了 support 放置的理由("主要为示例和覆盖率门禁而存在,`ui/` 保留给作为产品交付的界面"),这留下了一个持续的张力:一个已交付的产品应用依赖一个被明确标注为非产品表面的 support 包。
-
-这条边界换来的是:包元数据、workspace 与 tsconfig 引用、module-graph 行、README 条目,以及 publint 表面——服务于一个并不可独立替换的辅助模块:stdio 应用的前门集群始终包含 readline UI,且没有其他消费方能有意义地使用它。
-
-## 决策
-
-当时,该辅助函数移入 `@deepseek-ai/dsh-stdio`,成为终端通道插件。`createStdioChat`、其 `StdioRuntime` 测试 seam 和单元测试随之一同迁移,使 EOF 处理、渲染、释放以及管道/TTY 行为继续受逐文件覆盖率门禁约束,而不会劫持进程全局量。该模块保留应用挂载所消费的具名 `name`/`inject`/`Config`/`apply` 导出形状;当时的 Echo 和 REPL Loader 冒烟证明组合树,插件形状套件则固定显式 `unwrapExports` 行为。上方取代本文的移除记录负责当前包和示例状态。
-
-早期的支持辅助包已移除:其清单、tsconfig 引用、模块图行和 README 行均已消失,其余文档改为描述包内模块。
-
-## 曾考虑的替代方案
-
-### 为什么不将其提升到 `ui/` 而是折入?
-
-提升可以解决 support 与 product 之间的错位,同时保留边界——只有在 readline UI 是一个可独立替换的集成或有第二个组合方时才是正确选择,而消费方普查表明两者皆非。结构化的 ACP(Agent Client Protocol)桥接保留为独立包,因为它是具有自身契约和快照层级的产品协议表面;readline 辅助模块只是一个应用前门的脚手架。在发布前重新拆分成本很低:如果将来有第二个产品应用需要 readline UI,届时再拆出来,由那个消费方来塑造包契约。
-
-## 后果
-
-- stdio 应用完整拥有自己的前门;叶子 `cordis.yml` 仍然只加载一个应用包,演示的形态没有变化。
-- 未来如果有独立的终端 UI 需要将该辅助模块作为包使用,届时由那个第二消费方驱动重新引入,而非仓库为假设性的复用保留一条边界。
diff --git a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.i18n.yaml
index 0aa39a18ff..08ee0f1c29 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.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
-2026-07-04-trim-acp-bridge-unreachable-surface.md: ce0c623930192d02c8347c3956c497ee13048904
-2026-07-04-trim-acp-bridge-unreachable-surface.zh.md: e89b0e7c9c7c8438fb71adacf1b525a0b94bf04b
+2026-07-04-trim-acp-bridge-unreachable-surface.md: 959cad37f279888fda79b1632c3553ea122803c5
+2026-07-04-trim-acp-bridge-unreachable-surface.zh.md: 9127b7f167c3e5c74cdb99266828c20e79f3cb90
diff --git a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.md b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.md
index ce0c623930..959cad37f2 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.md
+++ b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.md
@@ -4,16 +4,18 @@ Status: implemented
English | [中文](2026-07-04-trim-acp-bridge-unreachable-surface.zh.md)
+> The handshake-identity simplification remains current. The generic-card fallback was removed when [ACP became automation-only](2026-07-23-acp-automation-only-protocol.md); UI transports retain the provider-neutral presentation contract.
+
## Problem
Two pieces of `dsh-acp` surface were unreachable from any shipped configuration:
-1. **`AcpConfig.agentName` / `agentVersion`** (`packages/ui/acp/src/index.ts`). The shipped app package hands the bridge only `{ model }` (`packages/examples/acp-demo/src/index.ts`), so no leaf `cordis.yml` — the only production config surface — could set the knobs at all; they were settable solely by direct-mounting the bridge, which only a unit test did. Every snapshot expected output — the hook-matrix scenarios included — pins the schema defaults (`deepseek-harness-acp` / `0.0.1`). The pair also carried a live `TODO(double-default)`: the literals existed twice (schema `.default(...)` plus `??` fallbacks), with the TODO asking to pick one home.
+1. **`AcpConfig.agentName` / `agentVersion`** (`packages/acp/acp/src/index.ts`). The shipped app package hands the bridge only its agent target (`packages/examples/acp-demo/src/index.ts`), so no leaf `cordis.yml` — the only production config surface — could set the knobs at all; they were settable solely by direct-mounting the bridge, which only a unit test did. Every snapshot expected output — the hook-matrix scenarios included — pins the schema defaults (`deepseek-harness-acp` / `0.0.1`). The pair also carried a live `TODO(double-default)`: the literals existed twice (schema `.default(...)` plus `??` fallbacks), with the TODO asking to pick one home.
2. **The `toolKindFor` name heuristic** (same file) special-cased `bash*`/`read*`/`write`/`edit*` tool names in the generic-fallback path. Since the [render-intent union](../architecture/2026-07-02-tool-render-intent-union.md), every first-party tool those arms matched ships its own `presentCall` carrying its kind, and the presenter-less production tools (`subagent`, `subagent_fork`) fell through to `other` anyway. The arms were production-reachable only when a tool declined to present its own call — a `presentCall` that THROWS (the containment fallback), or model arguments that fail the tool's schema so `defineTool`'s `presentCall` wrapper returns `undefined` (e.g. a `bash` call missing the required `description`) — and the bridge's own module doc states the design rule the heuristic violated: "the bridge never special-cases tool names".
## Decision
-Hardcode the existing handshake identity `{ name: 'deepseek-harness-acp', version: '0.0.1' }` at initialization and remove the unreachable config fields and duplicate defaults. Replace `toolKindFor` with neutral `'other'` at both presenter fallbacks. Normal first-party presentations are unchanged; malformed or failed presentations now render an honest generic card instead of inferring a kind from the tool name. Initialize tests and snapshots pin the handshake; only the malformed calls in `hook-codex-posttool-block` change fallback card kind.
+Hardcode the existing handshake identity `{ name: 'deepseek-harness-acp', version: '0.0.1' }` at initialization and remove the unreachable config fields and duplicate defaults. The original implementation also replaced `toolKindFor` with neutral `'other'` at both presenter fallbacks; ACP no longer projects tool cards, so that fallback has left the transport entirely. Initialize tests and snapshots pin the handshake.
## Alternatives considered
@@ -23,4 +25,4 @@ Branding can return when the app package exposes it to deployments. Inferring pr
## Consequences
-Nothing beyond the fallback rendering trade described above — degenerate paths whose neutral card is more diagnosable than an inferred first-party one.
+The bridge exposes no branding knobs. UI transports own generic presentation fallback without tool-name inference, while ACP carries no tool-card surface.
diff --git a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.zh.md b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.zh.md
index e89b0e7c9c..9127b7f167 100644
--- a/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-04-trim-acp-bridge-unreachable-surface.zh.md
@@ -4,16 +4,18 @@ Status: implemented
[English](2026-07-04-trim-acp-bridge-unreachable-surface.md) | 中文
+> 握手标识简化仍然有效。通用卡片回退已随 [ACP 转为仅面向自动化](2026-07-23-acp-automation-only-protocol.md)一并移除;UI 传输层保留提供方无关的展示契约。
+
## 问题
`dsh-acp` 有两处对外表面在任何已交付的配置中都不可达:
-1. **`AcpConfig.agentName` / `agentVersion`**(`packages/ui/acp/src/index.ts`)。已发布应用包只向 bridge 传递 `{ model }`(`packages/examples/acp-demo/src/index.ts`),因此没有任何叶子 `cordis.yml`——唯一的生产配置表面——能够设置这些配置项;只有直接挂载 bridge 才能设置它们,而这种做法只存在于一个单元测试中。每份快照预期输出——包括钩子矩阵场景——都固定 schema 默认值(`deepseek-harness-acp` / `0.0.1`)。这对配置项还带有一个尚未解决的 `TODO(double-default)`:字面量存在两次(schema `.default(...)` 加 `??` 后备值),TODO 要求为它们选择一个归属。
+1. **`AcpConfig.agentName` / `agentVersion`**(`packages/acp/acp/src/index.ts`)。已发布应用包只向 bridge 传递其 agent 的提供方/模型目标(`packages/examples/acp-demo/src/index.ts`),因此没有任何叶子 `cordis.yml`——唯一的生产配置表面——能够设置这些配置项;只有直接挂载 bridge 才能设置它们,而这种做法只存在于一个单元测试中。每份快照预期输出——包括钩子矩阵场景——都固定 schema 默认值(`deepseek-harness-acp` / `0.0.1`)。这对配置项还带有一个尚未解决的 `TODO(double-default)`:字面量存在两次(schema `.default(...)` 加 `??` 后备值),TODO 要求为它们选择一个归属。
2. **`toolKindFor` 名称启发式**(同一文件)在通用回退路径中对 `bash*`/`read*`/`write`/`edit*` 工具名做了特殊处理。自[render-intent 联合类型](../architecture/2026-07-02-tool-render-intent-union.md)以来,这些分支匹配到的每个第一方工具都自带 `presentCall` 并携带其 kind,而没有 presenter 的生产工具(`subagent`、`subagent_fork`)本来就落入 `other`。这些分支只有在工具拒绝自行呈现调用时才在生产中可达:`presentCall` 抛出异常(容错回退),或模型参数未通过工具 schema 导致 `defineTool` 的 `presentCall` 包装层返回 `undefined`(例如 `bash` 调用缺少必需的 `description`)。而桥接层自身的模块文档明确声明了该启发式所违反的设计规则:"桥接层绝不对工具名做特殊处理"。
## 决策
-在初始化时硬编码现有的握手标识 `{ name: 'deepseek-harness-acp', version: '0.0.1' }`,移除不可达的配置字段与重复默认值。在两个 presenter 回退处,将 `toolKindFor` 替换为中性的 `'other'`。正常的第一方呈现不受影响;格式错误或失败的呈现现在会渲染一个诚实的通用卡片,而非从工具名推断 kind。初始化测试和快照固定握手标识;只有 `hook-codex-posttool-block` 中格式错误的调用改变了回退卡片的 kind。
+在初始化时硬编码现有的握手标识 `{ name: 'deepseek-harness-acp', version: '0.0.1' }`,移除不可达的配置字段与重复默认值。最初的实现还在两个 presenter 回退处将 `toolKindFor` 替换为中性的 `'other'`;ACP 不再投影工具卡片,因此该回退已完全离开传输层。初始化测试和快照固定握手标识。
## 曾考虑的替代方案
@@ -23,4 +25,4 @@ Status: implemented
## 后果
-除上述回退渲染的取舍外没有其他影响——退化路径下,中性卡片比推断出的第一方卡片更易于诊断。
+桥接层不暴露品牌配置项。UI 传输层拥有不做工具名推断的通用展示回退,而 ACP 不承载任何工具卡片表面。
diff --git a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.i18n.yaml
index 91e9b078ad..89b7165650 100644
--- a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.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
-2026-07-20-remove-stdio-and-echo-agents.md: 2aba8193710c96d3726b91062bfa43d039b4cabf
-2026-07-20-remove-stdio-and-echo-agents.zh.md: 2c3916683f4743384a2ce4104319da26145837fe
+2026-07-20-remove-stdio-and-echo-agents.md: d804f8e3c3de886fc48c87d6378c9f06d0e3c15a
+2026-07-20-remove-stdio-and-echo-agents.zh.md: d42e6d527231e1703153c02c66d9286fa1e73c95
diff --git a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.md b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.md
index 2aba819371..d804f8e3c3 100644
--- a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.md
+++ b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.md
@@ -18,8 +18,8 @@ The stdio and Echo agents are removed without compatibility packages, modes, com
The remaining application roles are explicit:
-- [`@deepseek-ai/dsh-tui-demo`](../../../../packages/examples/tui-demo/README.md) owns terminal-interactive execution. `examples/tui-agent` owns the complete coding composition, Code Mode overlay, PTY coverage, and terminal snapshots.
-- [`@deepseek-ai/dsh-cli-demo`](../../../../packages/examples/cli-demo/README.md) owns non-interactive execution. `examples/headless-agent` owns the real-model one-shot composition, replay snapshots, generic real-agent suites, and test-only keyless Loader fixtures.
+- [`@deepseek-ai/dsh-tui-demo`](../../../../packages/examples/tui-demo/README.md) owns terminal-interactive execution. It rejects non-TTY streams before Loader boot; `examples/tui-agent` owns the complete coding composition, Code Mode overlay, PTY coverage, and terminal snapshots.
+- [`@deepseek-ai/dsh-cli-demo`](../../../../packages/examples/cli-demo/README.md) owns non-interactive execution, including pipes. `examples/headless-agent` owns the real-model one-shot composition, replay snapshots, generic real-agent suites, and test-only keyless Loader fixtures.
- [`@deepseek-ai/dsh-acp-demo`](../../../../packages/examples/acp-demo/README.md) and `@deepseek-ai/dsh-jsonrpc` own their framed protocol integrations.
The SDK project model and create/config workflows replace the `stdio` run-interface option with `tui`; generated TUI projects compose `@deepseek-ai/dsh-tui` and create or resume one exact session. Repository-facing demo documentation requires a DeepSeek API key and leads with the real Headless or TUI agents.
@@ -28,11 +28,14 @@ Keyless validation is test-owned. The Headless Loader smoke uses a fixture adapt
## Verification
-TUI and Headless Loader coverage run the real app packages in source and built modes. TUI uses a pseudo-terminal; Headless proves its task/result and tool-call contracts. Generated graphs and repository searches reject stale package, command, leaf, and SDK-interface references.
+TUI and Headless Loader coverage run the real app packages in source and built modes. PTY-driven subprocess coverage is reserved for the TUI lifecycle; other entry-point smokes use the one-shot pipe protocol. Headless proves its task/result and tool-call contracts. Generated graphs and repository searches reject stale package, command, leaf, SDK-interface, `createStdioChat`, and `StdioRuntime` references.
+
+The TUI PTY smoke includes the Code Mode overlay composition, while `examples/cordis-agent/tests/keyless-smoke.e2e.ts` provides a minimal PTY boot over the real Cordis-agent Loader tree. The built `dsh` bin rejects a piped TUI launch before Loader boot and points at its one-shot `-p` mode; `apps/cli/tests/built-bin.e2e.ts` pins that path, while `cli-demo`'s built-bin suite runs text, JSON, and structurally parsed `stream-json` output under plain Node, persists fresh sessions, and rejects invalid arguments and missing config without contaminating stdout. Time-context integration uses the real Headless composition for two ordered turns, while its package tests own finer elapsed-time behavior.
## Alternatives considered
- **Keep the line agent only for pipes** — rejected because Headless has a bounded task contract, format-pure stdout, durable completion, and process exit status.
+- **Keep, fold, or promote the readline helper as a package** — rejected because it had one app consumer and no independently swappable contract. Folding it into the stdio app removed an unjustified support-package boundary but still retained the redundant product; a future standalone line UI needs a real second consumer before reintroducing that package.
- **Keep Echo as the keyless quick start** — rejected because the first product experience should exercise the real model and supported coding agent, not a scripted adapter with a bespoke tool.
- **Keep Echo only as a CI demo command** — rejected because test-owned Headless fixtures cover the same Loader and built-artifact boundaries without preserving a mock product leaf.
- **Remove every stdio or mock mechanism** — rejected because framed protocols, process I/O, and deterministic test adapters are independent infrastructure, not the removed agents.
@@ -43,3 +46,4 @@ TUI and Headless Loader coverage run the real app packages in source and built m
- The repository has no keyless user-facing agent demo; local agent demos require `DEEPSEEK_API_KEY`.
- CI retains keyless real-entry coverage through test fixtures rather than a product command.
- Existing stdio-agent configurations, Echo commands, and SDK `--interface=stdio` invocations fail instead of being translated.
+- Piped multi-turn interaction in one process and the readline provider for non-TTY `ask_user_question` are intentionally gone; resume covers durable multi-turn work, and a non-TTY composition must supply its own interaction provider.
diff --git a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.zh.md b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.zh.md
index 2c3916683f..d42e6d5272 100644
--- a/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.zh.md
@@ -18,8 +18,8 @@ DeepSeek Harness 在 TUI 和 Headless coding agent 之外,还提供了两个
保留的应用角色均有明确归属:
-- [`@deepseek-ai/dsh-tui-demo`](../../../../packages/examples/tui-demo/README.md) 负责终端交互式执行。`examples/tui-agent` 拥有完整 coding 组装、Code Mode 覆盖层、PTY 覆盖和终端快照。
-- [`@deepseek-ai/dsh-cli-demo`](../../../../packages/examples/cli-demo/README.md) 负责非交互式执行。`examples/headless-agent` 拥有真实模型的单次任务组装、回放快照、通用真实 agent 测试套件,以及仅供测试使用的无密钥 Loader fixture。
+- [`@deepseek-ai/dsh-tui-demo`](../../../../packages/examples/tui-demo/README.md) 负责终端交互式执行。它会在 Loader 启动前拒绝非 TTY 流;`examples/tui-agent` 拥有完整 coding 组装、Code Mode 覆盖层、PTY 覆盖和终端快照。
+- [`@deepseek-ai/dsh-cli-demo`](../../../../packages/examples/cli-demo/README.md) 负责非交互式执行,包括管道方式。`examples/headless-agent` 拥有真实模型的单次任务组装、回放快照、通用真实 agent 测试套件,以及仅供测试使用的无密钥 Loader fixture。
- [`@deepseek-ai/dsh-acp-demo`](../../../../packages/examples/acp-demo/README.md) 和 `@deepseek-ai/dsh-jsonrpc` 负责各自的分帧协议集成。
SDK 工程模型与 create/config 工作流将 `stdio` 运行接口选项替换为 `tui`;生成的 TUI 工程组合 `@deepseek-ai/dsh-tui`,并创建或恢复一个确切会话。仓库中的演示文档要求 DeepSeek API key,并优先引导到真实的 Headless 或 TUI agent。
@@ -28,11 +28,14 @@ SDK 工程模型与 create/config 工作流将 `stdio` 运行接口选项替换
## 验证
-TUI 与 Headless 的 Loader 覆盖以源码和构建产物两种模式运行真实 app 包。TUI 使用伪终端;Headless 验证任务/结果契约和工具调用契约。生成图谱与仓库搜索会拒绝陈旧的包、命令、叶节点和 SDK 接口引用。
+TUI 与 Headless 的 Loader 覆盖以源码和构建产物两种模式运行真实 app 包。由 PTY 驱动的子进程覆盖仅用于 TUI 生命周期;其他入口冒烟测试使用单次管道协议。Headless 验证任务/结果契约和工具调用契约。生成图谱与仓库搜索会拒绝陈旧的包、命令、叶节点、SDK 接口、`createStdioChat` 和 `StdioRuntime` 引用。
+
+TUI PTY 冒烟测试包含 Code Mode 覆盖层组装,而 `examples/cordis-agent/tests/keyless-smoke.e2e.ts` 会基于真实 Cordis-agent Loader 目录树执行最小 PTY 启动。构建后的 `dsh` 可执行文件会在 Loader 启动前拒绝通过管道启动 TUI,并指向其单次 `-p` 模式;`apps/cli/tests/built-bin.e2e.ts` 固定了该执行路径,而 `cli-demo` 的 built-bin 套件在普通 Node 下运行文本、JSON 和经过结构化解析的 `stream-json` 输出,持久化新建会话,并在不污染 stdout 的情况下拒绝无效参数和缺失配置。时间上下文集成通过真实 Headless 组装执行两个有序轮次,而更细粒度的耗时行为由时间上下文的包级测试负责。
## 曾考虑的替代方案
- **仅为 pipe 保留面向行 agent**:不予采纳,因为 Headless 已提供有界任务契约、格式纯净的 stdout、持久完成边界和进程退出状态。
+- **将 readline helper 作为包保留、折叠或提升**:不予采纳,因为它只有一个 app 消费方,并不存在可独立替换的契约。将它折叠进 stdio app 虽然移除了没有正当理由的支撑包边界,却仍保留了重复产品;将来要重新引入这个包,独立的面向行 UI 必须先有真正的第二个消费方。
- **保留 Echo 作为无密钥快速上手路径**:不予采纳,因为首次产品体验应使用真实模型和受支持的 coding agent,而不是带专用工具的脚本化适配器。
- **只为 CI 演示命令保留 Echo**:不予采纳,因为由测试持有的 Headless fixture 可以覆盖相同的 Loader 和构建产物边界,无需保留 mock 产品叶节点。
- **移除所有 stdio 或 mock 机制**:不予采纳,因为分帧协议、进程 I/O 和确定性测试适配器是独立基础设施,并不是被移除的 agent。
@@ -43,3 +46,4 @@ TUI 与 Headless 的 Loader 覆盖以源码和构建产物两种模式运行真
- 仓库没有面向用户的无密钥 agent 演示;本地 agent 演示需要 `DEEPSEEK_API_KEY`。
- CI 通过测试 fixture 保留针对真实入口的无密钥覆盖,而不是依赖产品命令。
- 既有 stdio agent 配置、Echo 命令和 SDK `--interface=stdio` 调用会直接失败,不会被转换。
+- 有意移除了单进程内基于管道的多轮交互,以及面向非 TTY `ask_user_question` 的 readline 提供方;恢复会话可以满足持久多轮工作,非 TTY 组装则必须自行提供交互提供方。
diff --git a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.i18n.yaml
deleted file mode 100644
index 232fec495b..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.i18n.yaml
+++ /dev/null
@@ -1,6 +0,0 @@
-# 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
-2026-07-20-retire-readline-front-door.md: 7ebcfdc246bdf6971418609c61acbd4019aa90cb
-2026-07-20-retire-readline-front-door.zh.md: cf4d03594ed3a0cf31bed96eb2133bd37959084a
diff --git a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.md b/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.md
deleted file mode 100644
index 7ebcfdc246..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.md
+++ /dev/null
@@ -1,46 +0,0 @@
-# Agent Note: Retire the readline front door and the repl-agent example
-
-Status: implemented
-
-English | [中文](2026-07-20-retire-readline-front-door.zh.md)
-
-## Problem
-
-The repo shipped two interactive terminal front doors: the line-oriented readline channel (`@deepseek-ai/dsh-stdio`) and the full-screen [`@deepseek-ai/dsh-tui`](../feature/2026-07-17-dedicated-full-screen-tui-front-door.md). After the TUI landed, readline's interactive role was redundant — `demo:tui` superseded `demo:repl` as the coding-agent experience — while its remaining real role, pipes and automation, was already served better by the one-shot `@deepseek-ai/dsh-cli-demo` app (task in, DSH-native `text`/`json`/`stream-json` out, durable persistence, signal handling).
-
-The duplication was structural, not just cosmetic: `dsh-stdio-demo` carried a `TerminalMode` (`auto`/`readline`/`tui`) selection seam, ~1,000 lines of readline unit tests, a readline transcript grammar (`[tool call] …` lines) that the CI demo smoke and two built-bin e2es grepped, and an inverted example composition where the flagship `tui-agent` leaf was defined as an include-patch over the `repl-agent` leaf it superseded.
-
-## Decision
-
-Delete the readline front door and the repl-agent example; keep exactly three front-door archetypes: **interactive TUI** (TTY-only, fails loud on pipes), **one-shot CLI** (`-p`/positional task, pipes and automation), and **servers** (ACP / JSON-RPC).
-
-- `packages/ui/stdio` and `examples/repl-agent` are gone. `packages/examples/stdio-demo` is renamed `@deepseek-ai/dsh-tui-demo` (`packages/examples/tui-demo`) and always mounts `dsh-tui`; the `TerminalMode`/`resolveTerminalMode`/`ui.mode` seam is deleted. The bin refuses non-TTY streams **before booting the Loader** (a compose-time throw inside a Loader tree is logged per-entry, not rethrown, so a piped launch would otherwise settle into an idle UI-less process instead of exiting nonzero).
-- `examples/tui-agent/cordis.yml` now owns the coding composition inline (the include-patch inversion is gone); its Code Mode overlay includes its own base. `examples/cordis-agent` moved to the TUI app.
-- `examples/echo-agent` moved to the one-shot `dsh-cli-demo` app; `dsh-cli-demo` gained `-p/--prompt` as the flag form of the single task (mutually exclusive with the positional).
-- The UI-independent with-key coding e2es (`full-loop`, `coding-task`, `resume`, `compaction`, `todo-write`, `code-mode` and their shared harness) moved verbatim from `examples/repl-agent/tests/` to `examples/tui-agent/tests/` — they assemble the stack programmatically and never touched a UI.
-- The SDK wizard's `stdio` run interface became `tui` (`RunInterface = 'acp' | 'tui' | 'embed'`), contributing a `dsh-tui` entry instead of `dsh-stdio`; the generated `index.ts` guards TTY before `startSDK` for the same pre-boot fail-loud reason as the tui-demo bin.
-
-### Testing policy: PTY only for the TUI
-
-Pipes remain the default test medium. PTY-driven subprocess tests are sanctioned **only** where the subject is the TUI itself: `examples/tui-agent/tests/tui-keyless-smoke.e2e.ts` (which gained the Code Mode overlay boot scenario, replacing repl-agent's pipe smoke as the overlay's keyless composition proof) and the minimal PTY boot smoke in `examples/cordis-agent` (whose front door IS the TUI). Everything else moved to pipes over the one-shot bin:
-
-- `examples/echo-agent/tests/echo.e2e.ts` proves the Loader boot + mock-model tool round-trip through `stream-json` records instead of readline transcript lines.
-- The CI demo-smoke gate (`scripts/run-gates.ts`, AGENTS.md) runs `demo:echo --output-format stream-json -p "echo ci smoke"` and parses the records structurally.
-- `packages/examples/tui-demo/tests/built-bin.e2e.ts` proves the built bin's piped-launch refusal (nonzero exit + pointer at `dsh-cli-demo`); the echo-round-trip-under-plain-Node and missing-config fail-loud proofs live in `cli-demo`'s built-bin suite.
-- `packages/context/time-context/tests/time-context.e2e.ts` runs one one-shot turn; multi-turn elapsed rendering stays unit-covered in its spec.
-
-## Accepted losses
-
-- **Piped multi-turn in one process** — the readline channel could script several turns over stdin; the one-shot bin runs one task per process. Multi-turn continuity is covered by `RESUME_SESSION_ID`/resume e2es and the TUI's scripted PTY conversation.
-- **Non-TTY `ask_user_question`** — the readline provider was the only non-TTY terminal implementation of `ctx.userInteraction`. A headless run whose model calls `ask_user_question` now fails that tool call (no provider); the ACP bridge remains the non-terminal provider. A future headless deployment that needs it composes its own provider.
-
-## Alternatives considered
-
-- **Keep `dsh-stdio` as a pipe/automation channel without the repl demo** — rejected: its automation role duplicated `dsh-cli-demo` with a weaker contract (unstructured transcript, EOF-exit heuristics vs. one durable turn ending and format-pure output).
-- **Rewrite the piped smokes as PTY drivers** — rejected: PTY is the flakier, more complex medium and is reserved for the one surface pipes cannot prove (real TTY takeover/restore).
-
-## Consequences
-
-- One interactive front door (TUI), one automation front door (one-shot CLI), two servers; no mode-selection seam in the terminal app.
-- ~1,000 lines of readline unit tests deleted with their behavior; the readline transcript grammar is gone from all gates.
-- This supersedes the packaging half of [fold the stdio UI helper](2026-07-04-fold-stdio-ui-helper.md) (the folded package is now deleted) and amends the composition described in [the TUI front-door note](../feature/2026-07-17-dedicated-full-screen-tui-front-door.md) (no `auto` selection; `tui-agent` owns the coding composition).
diff --git a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.zh.md b/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.zh.md
deleted file mode 100644
index cf4d03594e..0000000000
--- a/.agents/notes/implemented/simplification/2026-07-20-retire-readline-front-door.zh.md
+++ /dev/null
@@ -1,46 +0,0 @@
-# Agent Note: 退役 readline 前端与 repl-agent 示例
-
-Status: implemented
-
-[English](2026-07-20-retire-readline-front-door.md) | 中文
-
-## 问题
-
-仓库同时提供两个交互式终端前端:面向行的 readline 通道(`@deepseek-ai/dsh-stdio`)和全屏的 [`@deepseek-ai/dsh-tui`](../feature/2026-07-17-dedicated-full-screen-tui-front-door.md)。TUI 落地之后,readline 的交互角色已经冗余——`demo:tui` 作为编码 agent 体验取代了 `demo:repl`——而它剩下的真实角色(管道与自动化)已由单次任务的 `@deepseek-ai/dsh-cli-demo` 应用以更好的方式承担(任务输入、DSH 原生 `text`/`json`/`stream-json` 输出、持久化、信号处理)。
-
-这种重复是结构性的,不只是表面问题:`dsh-stdio-demo` 携带一个 `TerminalMode`(`auto`/`readline`/`tui`)选择接缝、约 1,000 行 readline 单元测试、一套被 CI 演示冒烟测试和两个 built-bin e2e 用 grep 匹配的 readline 文本记录语法(`[tool call] …` 行),以及一个倒置的示例组合:旗舰 `tui-agent` 叶节点被定义为对它所取代的 `repl-agent` 叶节点的 include patch。
-
-## 决定
-
-删除 readline 前端和 repl-agent 示例;只保留三类前端原型:**交互式 TUI**(仅 TTY,管道下快速失败)、**单次任务 CLI**(`-p`/位置参数任务,服务管道与自动化)以及**服务器**(ACP / JSON-RPC)。
-
-- `packages/ui/stdio` 与 `examples/repl-agent` 已删除。`packages/examples/stdio-demo` 更名为 `@deepseek-ai/dsh-tui-demo`(`packages/examples/tui-demo`)并始终挂载 `dsh-tui`;`TerminalMode`/`resolveTerminalMode`/`ui.mode` 接缝随之删除。bin 在**启动 loader 之前**就拒绝非 TTY 流(Loader 树内组合期抛出的异常按条目记录日志而不会重新抛出,管道启动否则会沉降为一个空闲的无 UI 进程而不是以非零码退出)。
-- `examples/tui-agent/cordis.yml` 现在内联拥有编码组合(include patch 倒置消失);其 Code Mode 覆盖层 include 自己的基础配置。`examples/cordis-agent` 迁移到 TUI 应用。
-- `examples/echo-agent` 迁移到单次任务的 `dsh-cli-demo` 应用;`dsh-cli-demo` 新增 `-p/--prompt` 作为单个任务的旗标形式(与位置参数互斥)。
-- 与 UI 无关的带密钥编码 e2e(`full-loop`、`coding-task`、`resume`、`compaction`、`todo-write`、`code-mode` 及其共享 harness)原样从 `examples/repl-agent/tests/` 移入 `examples/tui-agent/tests/`——它们以编程方式组装整个栈,从不接触任何 UI。
-- SDK 向导的 `stdio` 运行接口改为 `tui`(`RunInterface = 'acp' | 'tui' | 'embed'`),贡献 `dsh-tui` 配置项而不是 `dsh-stdio`;生成的 `index.ts` 在 `startSDK` 之前检查 TTY,理由与 tui-demo bin 的启动前快速失败相同。
-
-### 测试策略:PTY 仅用于 TUI
-
-管道仍是默认测试介质。PTY 驱动的子进程测试**仅**在被测对象就是 TUI 本身时获准使用:`examples/tui-agent/tests/tui-keyless-smoke.e2e.ts`(新增 Code Mode 覆盖层启动场景,取代 repl-agent 的管道冒烟测试成为该覆盖层的无密钥组合证明)和 `examples/cordis-agent` 中最小的 PTY 启动冒烟测试(其前端就是 TUI)。其余全部改为通过单次任务 bin 走管道:
-
-- `examples/echo-agent/tests/echo.e2e.ts` 通过 `stream-json` 记录证明 Loader 启动 + mock 模型的工具往返,而不是匹配 readline 文本记录行。
-- CI 演示冒烟门禁(`scripts/run-gates.ts`、AGENTS.md)运行 `demo:echo --output-format stream-json -p "echo ci smoke"` 并结构化解析记录。
-- `packages/examples/tui-demo/tests/built-bin.e2e.ts` 证明构建产物 bin 对管道启动的拒绝(非零退出 + 指向 `dsh-cli-demo` 的提示);纯 Node 下的 echo 往返证明与缺失配置的快速失败证明位于 `cli-demo` 的 built-bin 套件。
-- `packages/context/time-context/tests/time-context.e2e.ts` 运行一个单次任务轮次;多轮 elapsed 渲染仍由其单元测试覆盖。
-
-## 接受的损失
-
-- **单进程内的管道多轮对话**——readline 通道可以通过 stdin 脚本化多个轮次;单次任务 bin 每个进程只运行一个任务。多轮连续性由 `RESUME_SESSION_ID`/resume e2e 和 TUI 的脚本化 PTY 对话覆盖。
-- **非 TTY 的 `ask_user_question`**——readline 提供方是 `ctx.userInteraction` 唯一的非 TTY 终端实现。模型调用 `ask_user_question` 的 headless 运行现在会让该工具调用失败(没有提供方);ACP 桥接仍是非终端提供方。未来需要它的 headless 部署自行组合提供方。
-
-## 曾考虑的替代方案
-
-- **保留 `dsh-stdio` 作为纯管道/自动化通道而只删 repl 演示**——不予采纳:它的自动化角色以更弱的契约重复了 `dsh-cli-demo`(非结构化文本记录、EOF 退出的启发式判断,对比后者的一次持久轮次结束和格式纯净输出)。
-- **把管道冒烟测试改写为 PTY 驱动**——不予采纳:PTY 是更易波动、更复杂的介质,仅保留给管道无法证明的那一个表面(真实 TTY 的接管/恢复)。
-
-## 后果
-
-- 一个交互式前端(TUI)、一个自动化前端(单次任务 CLI)、两个服务器;终端应用不再有模式选择接缝。
-- 约 1,000 行 readline 单元测试随其行为一起删除;readline 文本记录语法从所有门禁中消失。
-- 本决定取代 [fold the stdio UI helper](2026-07-04-fold-stdio-ui-helper.md) 的打包部分(被折叠的包现已删除),并修订 [TUI 前端 Agent Note](../feature/2026-07-17-dedicated-full-screen-tui-front-door.md) 描述的组合(不再有 `auto` 选择;`tui-agent` 拥有编码组合)。
diff --git a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.i18n.yaml
index c6b8ad0680..dbd47ad90e 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.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
-2026-07-22-plan-specific-collaboration-state.md: 2fc163213ca0ee1de5633e4d7db14a814b2f7bb2
-2026-07-22-plan-specific-collaboration-state.zh.md: 811f657bf31c96dde88e400fc25fe2fe6df1f157
+2026-07-22-plan-specific-collaboration-state.md: fb26d15238f0eb1b63fdccc7e48a6c49a44236cf
+2026-07-22-plan-specific-collaboration-state.zh.md: 93186eebb263458bc99e7f7562d065fbf9e5d4bf
diff --git a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.md b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.md
index 2fc163213c..fb26d15238 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.md
+++ b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.md
@@ -1,4 +1,4 @@
-# Agent Note: Collapse named session modes into plan mode
+# Agent Note: Plan-specific collaboration state
Status: implemented
@@ -8,7 +8,9 @@ English | [中文](2026-07-22-plan-specific-collaboration-state.zh.md)
The first plan-mode implementation introduced a generic named-mode registry even though the product shipped only `plan`. `ModeConfig.modes`, definition-name validation, `ctx.modes.list()`, retired-definition fallback, and a synthetic `review` mode in tests existed only to support hypothetical future collaboration modes. The production-specific behavior—plan guidance, `/plan`, and `exit_plan_mode`—still lived in the same package, so the generic API did not isolate a reusable mechanism from plan policy.
-The word “mode” also spans unrelated domains. Sandbox mode is an enforcing policy owned by `ctx.sandboxPolicy` and logged as `sandbox/mode`; plan mode is a collaboration stance that contributes guidance and a reviewed exit. Treating both as instances of one named-mode abstraction would obscure their independent ownership. ACP's protocol happens to expose a generic mode picker, but that is an adapter vocabulary rather than evidence that the harness needs a generic mode domain.
+The word “mode” also spans unrelated domains. Sandbox mode is an enforcing policy owned by `ctx.sandboxPolicy` and logged as `sandbox/mode`; plan mode is a collaboration stance that contributes guidance and a reviewed exit. Treating both as instances of one named-mode abstraction would obscure their independent ownership. A transport's generic vocabulary is not evidence that the harness needs a generic mode domain.
+
+Plan mode also needs a durable stance, a reviewable plan artifact, an explicit human boundary, and request reconstruction across resume and fork. Those requirements belong to the plan feature even after the generic registry and interactive ACP projections are removed.
## Decision
@@ -16,10 +18,22 @@ Plan mode owns a plan-specific product package: `@deepseek-ai/dsh-plan-mode` at
Configuration is exactly `{ section: string }`. The package registers the fixed `plan:policy` section, `/plan [message]`, the exact `/plan off` direct-exit form, and `exit_plan_mode` itself. Bare `/plan` selects active; another non-empty argument selects it first and then sends the trimmed text through `agent.steer()`, making the text an ordinary logged user message in the affected step. `/plan off` selects inactive without model input and can cancel an entry that is still pending at the boundary. The exit tool remains registered while plan mode is inactive so the request tool catalog stays stable.
-ACP keeps its protocol-level `default` and `plan` ids. The bridge maps those two ids to the boolean service, advertises only that fixed pair, rejects every other id at the adapter boundary, and maps committed `plan/mode` events back to `current_mode_update`. The protocol remains generic without forcing genericity into the product domain.
+Human-facing compositions own plan selection and review. This note originally kept ACP's protocol-level `default`/`plan` picker as an adapter over the boolean service; [ACP as an automation-only protocol](2026-07-23-acp-automation-only-protocol.md) supersedes that wire projection, so the ACP composition now mounts neither plan mode nor a mode-selection protocol.
Sandbox mode and approval policy remain separate enforcement axes. Plan mode neither reads nor writes them, and the simplification introduces no shared base type, registry, or preset abstraction across those concepts.
+### Boundary and model contract
+
+`plan/mode` is log-only and non-surface, so resume, fork, and compaction recover the state without a live mirror. A spawned agent begins inactive because there is no creation-time plan option. Pending user selections flush before the affected request assembly on prompt submission, ordinary continuation, or a request-recovery retry; a failed durable append leaves the intent pending for a later boundary.
+
+The active state contributes the deployment's section at prompt order 50. Inactive state contributes no section, while `exit_plan_mode` remains registered in both states, so a transition changes the logged request header but not native tool schemas or the Code Mode SDK. A user-driven transition appends one plugin-sourced notice only when the last request header described the opposite state; a pre-first-request or net-zero selection adds none, and an approved tool exit relies on its tool result instead of a second notice.
+
+### Reviewed exit
+
+`exit_plan_mode` requires a calling agent in active plan mode and a non-empty markdown plan beginning with a heading. The user-interaction question carries that exact plan as detail and offers `Approve` or `Keep planning` plus free-text feedback. Only one `Approve` selection with no custom text consents; every other answer stays in plan mode and returns corrective feedback to the model. An approved exit becomes a silent pending selection, leaving plan guidance active for the rest of the current tool batch and removing it before the next request.
+
+The tool renders the submitted plan as a generic card titled by its first heading. An absent or failed user-interaction provider, a failed review, or plugin disposal while review is pending fails closed and leaves manual `/plan off` as the human escape path.
+
## Deleted surface
- The arbitrary definition map, mode-name regular expression, reserved-name rules, and per-definition command loop.
@@ -31,17 +45,27 @@ Sandbox mode and approval policy remain separate enforcement axes. Plan mode nei
**Keep a private generic registry and expose only plan today.** Rejected because the unused name/config machinery would still be maintained and tested without a second production consumer. A future collaboration state can establish the right shared seam from two concrete cases.
-**Fold sandbox mode into the same service.** Rejected because collaboration guidance and execution confinement have different owners, lifecycle semantics, and consumers. Their shared English noun is not a domain relationship.
+**Fold sandbox or approval policy into plan state.** Rejected because collaboration guidance, execution confinement, and permission decisions have different owners, lifecycle semantics, and consumers. A mode-owned sandbox cap also makes a user's explicit sandbox selection appear to succeed while silently doing nothing.
-**Let ACP own plan state.** Rejected because TUI, resume, fork, prompt assembly, and the exit tool need the same logged fact independently of ACP. ACP owns only the wire projection.
+**Let one presentation transport own plan state.** Rejected because TUI, Web, resume, fork, prompt assembly, and the exit tool need the same logged fact independently of any one transport. Presentation adapters own only their projections.
+
+**Split a capability-seam trio or put the state in the agent loop.** Rejected because plan mode has no swappable backend, while existing session, prompt, tool, command, and lifecycle seams already provide every required hook.
+
+**Put flips in surface messages or store plans in files.** Rejected because the stance is a log-only fact and the tool argument already records the reviewable plan. Surface duplication spends model context, while a plan directory creates a second durable home.
+
+**Filter tools by a per-plan name allowlist or a global policy stack.** Rejected because mutability is a property of each tool, including future and MCP tools, rather than a list that every plan deployment must maintain. Effects metadata can establish a shared policy only when a concrete consumer exists; until then plan mode is guidance, not a security boundary.
+
+**Review through the approval seam or prose.** Rejected because a plan review is not a permission decision, needs the exact artifact and corrective free text, and must have a logged tool call as its structured transition. The user-interaction seam supplies that contract.
## Verification
- Package tests retain boundary ordering, retry, append-failure, HMR disposal, prompt assembly, stable native and Code Mode schemas, review outcomes, and invariant coverage through the boolean service.
- Command tests cover bare `/plan`, `/plan `, active `/plan off`, pending-entry cancellation, inactive idempotence, absence of `/mode` and `/review`, and effect-scoped removal.
-- ACP tests cover fixed advertisement, both ids, unknown-id rejection, optimistic updates, committed exits, and load replay.
- The keyless TUI scenarios enter through `/plan `, leave through `/plan off`, and prove that each committed `plan/mode` precedes the request header it changes, the entry message is logged under plan guidance, and the post-exit request omits that guidance.
+- The complete `exit_plan_mode` review arc is package-tested but has no assembled-application snapshot after the interactive ACP scenarios were retired; current keyless TUI scenarios cover command entry and direct exit only.
## Consequences
-The implementation has one vocabulary for one shipped feature. Adding another collaboration stance is now an explicit design decision instead of a config entry, while ACP clients continue to see their standard mode picker. The migration intentionally rejects old `mode/set` logs and old `modes.plan.section` configuration under the repository's pre-release format policy.
+The implementation has one vocabulary for one shipped feature. Adding another collaboration stance is an explicit design decision instead of a config entry, and automation clients do not acquire human mode controls through ACP. The migration intentionally rejects old `mode/set` logs and old `modes.plan.section` configuration under the repository's pre-release format policy.
+
+Plan state remains reconstructable and tool schemas remain stable, but an idle pending selection is lost if the process exits before the next boundary. Entering or leaving plan mode changes the prompt from order 50 onward, and a model that ignores the guidance can still mutate unless the deployment independently configures sandbox, approval, or filesystem policy.
diff --git a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.zh.md b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.zh.md
index 811f657bf3..93186eebb2 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.zh.md
@@ -1,4 +1,4 @@
-# Agent Note: 将具名会话模式收敛为 plan mode
+# Agent Note: plan 专用协作状态
Status: implemented
@@ -8,7 +8,9 @@ Status: implemented
产品只交付了 `plan`,首个 plan mode 实现却引入了通用的具名模式注册表。`ModeConfig.modes`、定义名称校验、`ctx.modes.list()`、已退役定义的回退逻辑,以及测试中合成的 `review` 模式,都只为支持假想中的未来协作模式而存在。plan 引导、`/plan` 和 `exit_plan_mode` 这些生产专用行为仍位于同一个包(package)内,因此通用 API 并未将可复用机制与 plan 策略隔离开来。
-「mode」一词还横跨互不相关的领域。沙箱模式是由 `ctx.sandboxPolicy` 拥有、以 `sandbox/mode` 记录日志的强制执行策略;plan mode 则是一种协作方式,会贡献引导内容和经评审的退出路径。若把两者都视为同一个具名模式抽象的实例,就会掩盖二者各自独立的归属关系。ACP(Agent Client Protocol)协议恰好暴露了通用模式选择器,但这只是适配器词汇,并不能证明 harness 需要通用模式领域。
+「mode」一词还横跨互不相关的领域。沙箱模式是由 `ctx.sandboxPolicy` 拥有、以 `sandbox/mode` 记录日志的强制执行策略;plan mode 则是一种协作方式,会贡献引导内容和经评审的退出路径。若把两者都视为同一个具名模式抽象的实例,就会掩盖二者各自独立的归属关系。传输协议的通用词汇并不能证明 harness 需要通用模式领域。
+
+Plan mode 还需要持久协作状态、可评审的计划产物、显式人工决策边界,以及跨恢复与 fork 的请求重建。即使移除通用注册表和 ACP 交互投影,这些要求仍归 plan 功能所有。
## 决策
@@ -16,10 +18,22 @@ Plan mode 拥有一个 plan 专用产品包:位于 `packages/plan/plan-mode/`
配置严格为 `{ section: string }`。该包自行注册固定的 `plan:policy` 段、`/plan [message]`、精确匹配的 `/plan off` 主动退出形式,以及 `exit_plan_mode`。不带参数的 `/plan` 选择激活;其他非空参数则先选择激活,再通过 `agent.steer()` 发送去除首尾空白后的文本,使该文本在受影响的步骤中成为一条记录到日志的普通用户消息。`/plan off` 选择未激活,不产生模型输入,并可取消仍待在边界生效的进入选择。即使 plan mode 未激活,退出工具仍保持注册,以确保请求工具目录稳定。
-ACP 保留协议层的 `default` 和 `plan` id。桥接层把这两个 id 映射到布尔服务,只公布这组固定选项,在适配器边界拒绝其他所有 id,并把已提交的 `plan/mode` 事件映射回 `current_mode_update`。协议仍保持通用性,但不会迫使产品领域也采用通用抽象。
+面向人类的组合拥有 plan 选择与评审。本笔记最初把 ACP 协议级的 `default`/`plan` 选择器保留为布尔服务之上的适配器;[ACP 作为仅面向自动化的协议](2026-07-23-acp-automation-only-protocol.md)取代了那个线上投影,因此 ACP 组合现在既不挂载 plan mode,也不提供模式选择协议。
沙箱模式与审批策略仍是彼此独立的强制约束轴。Plan mode 既不读取也不写入二者;此次简化也没有为这些概念引入共享基类型、注册表或预设抽象。
+### 边界与模型契约
+
+`plan/mode` 仅记录到日志且不进入表层,因此恢复、fork 和压缩都能恢复该状态,无需实时镜像。spawn 出的 agent 初始处于未激活状态,因为创建时没有 plan 选项。待生效的用户选择会在提示词提交、普通 continuation 或请求恢复重试时,于受影响的请求组装前写入日志;持久追加失败会让意图保持待定,留到后续边界处理。
+
+激活状态在提示词顺序 50 处贡献部署提供的区段。未激活状态不贡献区段,但 `exit_plan_mode` 在两种状态下都保持注册,因此状态转换会改变已记录的请求头,却不改变原生工具 schema 或 Code Mode SDK。用户发起的转换只会在上一条请求头描述相反状态时追加一条来源为插件的通知;第一次请求前的选择或最终状态未变化的选择不会追加通知,经批准的工具退出则依赖其工具结果,不再追加第二条通知。
+
+### 经评审的退出
+
+`exit_plan_mode` 要求调用方 agent 处于激活的 plan mode,并提交一份非空、以标题开头的 markdown 计划。用户交互问题将这份原样计划作为详情,并提供 `Approve`、`Keep planning` 和自由文本反馈。仅当唯一选择为 `Approve` 且没有自定义文本时才视为同意;其他所有回答都会留在 plan mode,并向模型返回纠正性反馈。经批准的退出会成为一项静默的待生效选择,使 plan 引导在当前工具批次的剩余部分继续有效,并在下一次请求前移除。
+
+工具将提交的计划渲染为 generic 卡片,标题取自第一个标题。用户交互提供方缺失或失败、评审失败,或评审待定期间插件被 dispose,都会失败关闭,并保留手动 `/plan off` 作为人类退出路径。
+
## 删除的接口
- 任意定义映射、模式名正则表达式、保留名称规则以及逐定义命令循环。
@@ -31,17 +45,27 @@ ACP 保留协议层的 `default` 和 `plan` id。桥接层把这两个 id 映射
**保留私有的通用注册表,目前只暴露 plan。** 不予采纳,因为没有第二个生产消费方时,仍需维护和测试未使用的名称与配置机制。未来若出现另一种协作状态,可以从两个具体案例出发建立合适的共享 seam。
-**将沙箱模式折叠进同一服务。** 不予采纳,因为协作引导与执行约束有不同的归属方、生命周期语义和消费方。二者的英文名称都含「mode」,不代表存在领域关系。
+**将沙箱或审批策略折叠进 plan 状态。** 不予采纳,因为协作引导、执行约束和权限决策有不同的归属方、生命周期语义和消费方。由 mode 拥有的沙箱上限还会让用户显式选择沙箱看似成功,实际却被静默忽略。
-**让 ACP 拥有 plan 状态。** 不予采纳,因为 TUI、恢复、fork、提示词组装和退出工具都需要在 ACP 之外独立使用同一项已记录事实。ACP 只拥有协议投影。
+**让一种呈现传输拥有 plan 状态。** 不予采纳,因为 TUI、Web、恢复、fork、提示词组装和退出工具都需要独立于任何单一传输使用同一项已记录事实。呈现适配器只拥有各自的投影。
+
+**拆成能力 seam 三包,或把状态放进 agent loop。** 不予采纳,因为 plan mode 没有可替换后端,而现有的会话、提示词、工具、命令和生命周期 seam 已经提供所需的全部钩子。
+
+**将状态切换写入表层消息,或把计划存入文件。** 不予采纳,因为协作状态是仅日志事实,工具参数已经记录了可评审的计划。重复写入表层会消耗模型上下文,而计划目录会形成第二个持久归属。
+
+**按 plan 专用名称允许列表或全局策略栈筛选工具。** 不予采纳,因为可变性是每个工具自身的属性,包括未来工具和 MCP 工具,而不应由每个 plan 部署维护一份列表。只有出现具体消费方后,effects 元数据才能建立共享策略;在此之前,plan mode 是引导机制,不是安全边界。
+
+**通过审批 seam 或普通文本完成评审。** 不予采纳,因为计划评审不是权限决策,需要精确的计划产物和纠正性自由文本,而且必须以已记录的工具调用作为结构化转换。用户交互 seam 提供了这项契约。
## 验证
- 包测试通过布尔服务继续覆盖边界顺序、重试、追加失败、HMR(热模块替换)资源释放、提示词组装、稳定的原生 schema 与 Code Mode schema、评审结果和不变式。
- 命令测试覆盖不带参数的 `/plan`、`/plan `、激活状态下的 `/plan off`、取消待生效的进入选择、未激活状态下的幂等性、不存在 `/mode` 和 `/review`,以及随 effect 作用域移除。
-- ACP 测试覆盖固定模式列表公布、两个 id、未知 id 拒绝、乐观更新、已提交退出和加载回放。
- 无密钥 TUI 场景通过 `/plan ` 进入、通过 `/plan off` 退出,并证明每个已提交的 `plan/mode` 都先于其所改变的请求头,进入消息在 plan 引导下记录到日志,且退出后的请求不含该引导。
+- 完整的 `exit_plan_mode` 评审流程有包测试,但交互式 ACP 场景退役后没有组装应用快照;当前无密钥 TUI 场景只覆盖命令进入和直接退出。
## 后果
-该实现只用一套词汇描述一项已交付功能。若要添加另一种协作方式,必须显式作出设计决策,而不能只增加配置项;ACP 客户端仍可看到标准模式选择器。根据仓库的预发布格式策略,本次迁移有意拒绝旧的 `mode/set` 日志与 `modes.plan.section` 配置。
+该实现只用一套词汇描述一项已交付功能。若要添加另一种协作方式,必须显式作出设计决策,而不能只增加配置项;自动化客户端不会通过 ACP 获得面向人类的模式控制。根据仓库的预发布格式策略,本次迁移有意拒绝旧的 `mode/set` 日志与 `modes.plan.section` 配置。
+
+Plan 状态仍可重建,工具 schema 仍保持稳定,但如果进程在下一边界前退出,空闲状态下待生效的选择会丢失。进入或离开 plan mode 会改变提示词顺序 50 处及其后的内容;如果模型忽略引导,仍可能执行修改,除非部署另行配置沙箱、审批或文件系统策略。
diff --git a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.i18n.yaml
index 72e2341385..fbe06d209f 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.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
-2026-07-22-tui-titles-from-session-title-service.md: b54b99647230255cf241415f94aa21b2630c44cd
-2026-07-22-tui-titles-from-session-title-service.zh.md: 67cc3332f0694887d5af0d71997d140b74669f46
+2026-07-22-tui-titles-from-session-title-service.md: 04355b9c426af423dec347997f3b8ac62483eb7f
+2026-07-22-tui-titles-from-session-title-service.zh.md: 5fc783c5ca08baba12a60f2aa6b5e286307aa9f0
diff --git a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.md b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.md
index b54b996472..04355b9c42 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.md
+++ b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.md
@@ -6,11 +6,11 @@ English | [中文](2026-07-22-tui-titles-from-session-title-service.zh.md)
## Problem
-Two model-title implementations coexisted after the tui-staging line merged onto master. The TUI carried its own `autoTitle` feature: a fire-and-forget `ctx.llm.stream` call after the first user message that set the terminal window title via OSC 0, with a one-shot latch, its own prompt, its own 40-character cap, and its own resume re-derivation ([auto-title Agent Note](../feature/2026-07-21-tui-auto-pane-title.md), [default-on Agent Note](../feature/2026-07-21-tui-auto-title-default-on.md)). Master had meanwhile landed [log-backed session titles](../feature/2026-07-21-log-backed-session-titles.md): a `sessionTitle` capability whose accepted revisions are durable `session/title` events, with a deterministic fallback and optional model providers. The TUI already consumed `session/title` for its header subtitle and window title, so a session could be titled twice by different strategies, and the TUI's process-local title was invisible to every other consumer (ACP, resume listings, forks).
+A per-session title makes terminal panes and tabs distinguishable, but a TUI-local model call would create a second title pipeline beside [log-backed session titles](../feature/2026-07-21-log-backed-session-titles.md). The local path needs its own prompt, cap, one-shot latch, resume derivation, cancellation, and failure fallback, while its process-local result remains invisible to session listings, forks, Web consumers, and replay. If both paths run, one session can also be titled twice by different strategies.
## Decision
-The TUI-local generation is removed; the session-title service is the one title source. `TuiConfig.autoTitle`, the latch, the abort controller, the title prompt, and `titleLine` are gone from `dsh-tui`. The terminal rename stays: the TUI folds the latest logged title on mount (`foldSessionTitle`), renders it as the banner subtitle, and sets the terminal window title to ` — ` on every accepted `session/title` event — including resumed sessions, whose titles now replay from the log instead of being re-generated.
+The session-title service is the one title source. The TUI contains no `autoTitle` config, title-model request, latch, abort controller, prompt, or output cap. It folds the latest logged title on mount (`foldSessionTitle`), renders it as the banner subtitle, and calls `runtime.terminal.setTitle` with ` — ` on every accepted `session/title` event. The same terminal-safe OSC 0 path handles the configured fallback title, resumed sessions, and live revisions without renaming tmux windows or adding another terminal-control surface.
Model-made titles are a composition choice: `examples/tui-agent/cordis.yml` (and the scripted PTY fixture) mount `@deepseek-ai/dsh-session-title-first-message-llm`, which inherits the main request's route and replaces the spine's deterministic fallback with a short model summary. Deployments without the provider keep the fallback title from `dsh-agent-spine-demo`'s bundled `SessionTitleService`.
@@ -20,6 +20,16 @@ Model-made titles are a composition choice: `examples/tui-agent/cordis.yml` (and
**Port auto-title's prompt and cap into the service as a third provider.** The first-message-llm provider already exists with the same cadence, a reviewed prompt contract, durable request records, and supersession fencing; a second near-identical provider would be pure duplication.
+**Use only a truncated first prompt or only a model title.** A deterministic fallback provides an immediate, free title, while an optional model provider improves quality without delaying the main turn. Forcing either strategy removes that deployment choice.
+
+**Make model titles a TUI default or block the first turn for them.** The cost and route belong to composition, and auxiliary title latency must stay off the interaction critical path. The TUI consumes accepted state instead of owning generation policy.
+
+**Rename a tmux window or use a separate terminal escape.** Rejected because the existing terminal adapter's OSC 0 path labels the pane or tab without acquiring tmux ownership or adding a second control API.
+
+## Verification
+
+TUI tests pin restored and live `session/title` consumption, terminal-safe title rendering, the configured fallback, and the absence of a TUI-owned model path. The keyless PTY smoke boots the real composition, accepts a logged provider title, and observes the resulting terminal title. The [log-backed title decision](../feature/2026-07-21-log-backed-session-titles.md) owns provider, persistence, resume, fork, cancellation, and stale-completion coverage.
+
## Consequences
-One title pipeline: durable, replayable, visible to every consumer, and fenced against stale completions by the service. The TUI sheds ~90 lines and its `llm`-streaming path. The cost is that a title now requires the provider plugin in the composition for model quality — a leaf choice, not a TUI default — and the terminal title changes shape from the bare model summary to the suffixed ` — ` form the log-backed path always used. The superseded auto-title Agent Notes carry pointers here.
+One title pipeline is durable, replayable, visible to every consumer, and fenced against stale completions by the service. The TUI has no `llm`-streaming title path. Model quality requires a provider plugin in the composition, while deployments without one keep the deterministic fallback; the terminal title consistently uses the suffixed ` — ` shape.
diff --git a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.zh.md b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.zh.md
index 67cc3332f0..5fc783c5ca 100644
--- a/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.zh.md
+++ b/.agents/notes/implemented/simplification/2026-07-22-tui-titles-from-session-title-service.zh.md
@@ -6,11 +6,11 @@ Status: implemented
## 问题
-tui-staging 分支合入 master 后,两套模型标题实现并存。TUI 自带 `autoTitle` 特性:在首条用户消息后发起一次 fire-and-forget 的 `ctx.llm.stream` 调用,通过 OSC 0 设置终端窗口标题,带有一次性闩锁、自己的提示词、自己的 40 字符截断和自己的恢复重推导([auto-title Agent Note](../feature/2026-07-21-tui-auto-pane-title.md)、[default-on Agent Note](../feature/2026-07-21-tui-auto-title-default-on.md))。而 master 已落地[日志承载的会话标题](../feature/2026-07-21-log-backed-session-titles.md):一个 `sessionTitle` 能力,其被接受的修订是持久的 `session/title` 事件,带确定性回退和可选的模型 provider。TUI 已经消费 `session/title` 作为横幅副标题和窗口标题,于是一个会话可能被两种策略各标题一次,且 TUI 的进程本地标题对其他所有消费者(ACP、恢复列表、fork)不可见。
+每会话标题让终端窗格和标签页易于区分,但 TUI 本地模型调用会在[日志承载的会话标题](../feature/2026-07-21-log-backed-session-titles.md)旁形成第二条标题管线。本地路径需要自己的提示词、截断上限、一次性闩锁、恢复推导、取消和失败回退,而其进程本地结果仍对会话列表、fork、Web 消费方和回放不可见。若两条路径同时运行,同一会话还可能被不同策略命名两次。
## 决策
-移除 TUI 本地生成;session-title 服务是唯一的标题来源。`TuiConfig.autoTitle`、闩锁、abort controller、标题提示词和 `titleLine` 全部从 `dsh-tui` 删除。终端重命名保留:TUI 在挂载时折叠最新的已记录标题(`foldSessionTitle`),将其渲染为横幅副标题,并在每个被接受的 `session/title` 事件上把终端窗口标题设为 `<会话标题> — <配置标题>` —— 包括恢复的会话,其标题现在从日志回放而不是重新生成。
+session-title 服务是唯一的标题来源。TUI 不包含 `autoTitle` 配置、标题模型请求、闩锁、abort controller、提示词或输出上限。TUI 在挂载时折叠最新的已记录标题(`foldSessionTitle`),将其渲染为横幅副标题,并在每个被接受的 `session/title` 事件上调用 `runtime.terminal.setTitle`,传入 ` — `。同一条终端安全的 OSC 0 路径会处理配置的回退标题、恢复的会话和实时修订,既不重命名 tmux 窗口,也不增加另一套终端控制接口。
模型生成的标题是组合选择:`examples/tui-agent/cordis.yml`(以及脚本化 PTY fixture)挂载 `@deepseek-ai/dsh-session-title-first-message-llm`,它继承主请求的确切路由,用简短的模型摘要替换 spine 的确定性回退。未挂载该 provider 的部署保留 `dsh-agent-spine-demo` 内置 `SessionTitleService` 的回退标题。
@@ -20,6 +20,16 @@ tui-staging 分支合入 master 后,两套模型标题实现并存。TUI 自
**把 auto-title 的提示词和截断移植为服务的第三个 provider。** first-message-llm provider 已经存在,节奏相同,且有经过评审的提示词契约、持久的请求记录和替换围栏;再造一个近乎相同的 provider 纯属重复。
+**只使用截断后的首条提示词,或只使用模型标题。** 确定性回退可以立即且免费地提供标题,而可选模型 provider 可以提升质量,不会延迟主轮次。强制采用任一种策略都会移除这项部署选择。
+
+**让模型标题成为 TUI 默认行为,或为此阻塞第一个轮次。** 成本与路由归组合所有,辅助标题的延迟不得进入交互关键路径。TUI 只消费已接受的状态,不拥有生成策略。
+
+**重命名 tmux 窗口,或使用另一种终端转义序列。** 不予采纳,因为现有终端适配器的 OSC 0 路径可以标记窗格或标签页,无需取得 tmux 归属,也无需增加第二套控制 API。
+
+## 验证
+
+TUI 测试锁定恢复后和实时的 `session/title` 消费、终端安全的标题渲染、配置的回退标题,以及不存在 TUI 自有模型路径。无密钥 PTY 冒烟测试启动真实组合,接收已记录的 provider 标题,并观察由此产生的终端标题。[日志承载标题决策](../feature/2026-07-21-log-backed-session-titles.md)拥有 provider、持久化、恢复、fork、取消和陈旧完成结果的覆盖。
+
## 影响
-标题管线归一:持久、可回放、对所有消费者可见,并由服务对过期完成设防。TUI 削减约 90 行及其 `llm` 流式路径。代价是模型质量的标题现在需要在组合中挂载 provider 插件 —— 这是叶配置选择,不是 TUI 默认值 —— 且终端标题形状从裸模型摘要变为日志路径一贯使用的 `<标题> — <产品>` 后缀形式。被取代的 auto-title Agent Note 携带指向本文的指针。
+唯一的标题管线持久、可回放、对所有消费方可见,并由服务防止陈旧完成结果生效。TUI 不再有 `llm` 流式标题路径。若要提升模型标题质量,组合中必须挂载 provider 插件;未挂载的部署保留确定性回退。终端标题始终采用 ` — ` 后缀形式。
diff --git a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.i18n.yaml b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.i18n.yaml
similarity index 62%
rename from .agents/notes/implemented/feature/2026-07-14-time-context-plugin.i18n.yaml
rename to .agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.i18n.yaml
index 28ecd2a765..fe203d8f0f 100644
--- a/.agents/notes/implemented/feature/2026-07-14-time-context-plugin.i18n.yaml
+++ b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.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
-2026-07-14-time-context-plugin.md: 189f75fc12fe12e9dec56fc71ea901ec2eaa8b19
-2026-07-14-time-context-plugin.zh.md: 12671cb891531627fffabb7bd91a1532bc3de6b9
+2026-07-23-acp-automation-only-protocol.md: 0fe2fc27a963d21e8a24c1682359ab3bc9e7af48
+2026-07-23-acp-automation-only-protocol.zh.md: 0a471f0bf1b12e835660cdce2d2a2acd761e1b89
diff --git a/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.md b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.md
new file mode 100644
index 0000000000..0fe2fc27a9
--- /dev/null
+++ b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.md
@@ -0,0 +1,53 @@
+# Agent Note: ACP as an automation-only protocol
+
+Status: implemented
+
+English | [中文](2026-07-23-acp-automation-only-protocol.zh.md)
+
+## Problem
+
+The ACP bridge had become a second interactive product UI. It translated durable events into editor cards, terminal metadata, diffs, plans, titles, reasoning, commands, modes, model and permission pickers, session navigation, and human elicitation. Those responsibilities duplicated the TUI and the Web client while coupling an automation transport to UI services, persistence queries, presentation policy, and editor-specific conventions.
+
+ACP still has one useful role: another agent or automated controller can start a harness process, create an isolated session, send text, receive the committed answer, cancel work, and answer a permission request. The out-of-process ACP subagent backend depends on that standard protocol boundary.
+
+The snapshot suite complicates removal. Most ACP scenarios exercise the assembled agent backend rather than ACP presentation, so deleting the suite with the editor bridge would discard broad keyless behavioral coverage.
+
+## Decision
+
+`@deepseek-ai/dsh-acp` is an automation transport under [`packages/acp/acp`](../../../../packages/acp/acp/README.md), outside the `ui` package group. Its public protocol is intentionally small: version negotiation, fresh text sessions with one in-flight prompt each, committed assistant text updates, per-session cancellation, concurrent sessions, and connection-owned teardown. Prompts carry the spec-required baseline only — text plus resource links flattened to bracketed textual references; the bridge rejects additional directories, MCP servers, beyond-baseline prompt content (image, audio, embedded resources), empty prompts, unknown sessions, and overlapping prompts.
+
+The bridge emits only committed `assistant/message` text. Reasoning, raw chunks, tool activity, todos, plans, titles, retry markers, terminal metadata, diffs, locations, and resource links remain in the durable session log or in UI-specific transports. It does not provide session load/list/delete, commands, modes, configuration selectors, model switching, plan review, or human elicitation.
+
+One-shot `session/request_permission` remains. It is a machine policy channel for bridge-owned agents, not a human approval UI: the answerer accepts only an exact agent object in the bridge's live session map, delegates foreign or call-less requests, and maps failed RPCs to the fail-closed unavailable outcome. The client chooses allow once, reject once, or cancel, and the bridge never turns that response into a durable grant. Asking policy stays in the approval seam and its producers; [`dsh-subagent-acp`](../../../../packages/subagent/subagent-acp/README.md) uses this channel programmatically.
+
+The app composition contains the agent spine, persistence, checkpoint policy, and ACP transport. It does not mount command, session-query, session-reference, plan-mode, permission-picker, or user-interaction services for ACP. SDK scaffolding likewise treats `ask_user_question` as TUI-only.
+
+The transport programs interface-level agent, session, and approval services rather than the concrete agent loop. Tool execution stays inside the harness; ACP never delegates shell execution to an editor. stdout carries framed JSON-RPC only, so the app mounts no stdout logger and the bridge does not monkey-patch process output.
+
+Disconnect and plugin disposal share one memoized quiescence boundary. Both successful and failed transport closure settle pending prompts as cancelled, dispose every bridge-owned agent, and await loop and session cleanup. A create that loses the close race disposes its unpublished handle.
+
+## Snapshot boundary
+
+The ACP snapshot suite still boots the assembled ACP example and retains scenarios that pin backend behavior. Only scenarios driven through deleted UI methods leave the suite; semantic-checkpoint recovery runs through the headless `stream-json` example because ACP no longer loads sessions.
+
+Protocol and lifecycle tests pin stop-reason and prompt codecs, version negotiation, fresh-session creation, text and resource-link flattening, rejection of empty or unsupported prompts, exact-agent permission ownership, multi-session isolation, prompt settlement, per-session cancellation, failed transport closure, ACP-only reload cleanup, and teardown quiescence. Built and real-stdio smokes reject stray stdout. The `session/new` branch that loses a real stdio close race remains coverage-exempt because the in-memory transport cannot reproduce that ordering; it disposes the unpublished handle, while the surrounding disposal tests pin the no-orphan invariant.
+
+## Alternatives considered
+
+**Keep ACP as an editor UI until Web reaches parity.** Rejected because it leaves two interactive contracts to evolve and keeps editor conventions in the automation boundary.
+
+**Keep the earlier editor bridge behind disciplined seams.** Rejected even though that bridge correctly used interface services, tool-owned render intents, approval and user-interaction answerers, harness-owned execution, and a stdout-pure composition. Its terminal cards were capability-gated, display-only Zed `_meta` projections with a text fallback rather than ACP `terminal/create`, so shell execution never left the harness. The projection derived each display terminal id from the stable per-call id to prevent collisions and recovered exit code or signal from the rendered status markers because the pure result presenter received content blocks rather than a structured exit; marker round-trip tests and an explicit no-capability `console` fallback test pinned both contracts. Those boundaries were coherent but could not make editor cards, session navigation, configuration pickers, and human elicitation belong in an automation protocol.
+
+**Replace ACP with a private subagent RPC.** Rejected because ACP already supplies a typed, interoperable process protocol and is used by the out-of-process subagent backend.
+
+**Remove machine permission requests with the other interaction features.** Rejected because an automated parent must answer a child agent's one-shot policy decision; this is control flow between agents, not presentation.
+
+**Delete the ACP snapshot suite or migrate every scenario in this change.** Rejected because most scenarios test the backend and remain valuable, while a full harness migration is an independent testing change. Only scenarios whose driver was a deleted UI method leave this suite.
+
+## Consequences
+
+ACP has a narrow contract suitable for agents and automation, while TUI and Web own human interaction and presentation. The package has fewer injected services, dependencies, protocol branches, and lifecycle states, and it no longer claims compatibility as a general editor front door.
+
+Automation clients receive complete committed text rather than token deltas or structured tool UI. They inspect durable logs or another API when they need reasoning, tool traces, titles, or richer state. Fresh-session-only operation also means callers that need durable browsing or resume use a host API rather than ACP.
+
+Backend snapshot coverage therefore remains transport-coupled to ACP even though that transport is incidental to the behavior under test.
diff --git a/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.zh.md b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.zh.md
new file mode 100644
index 0000000000..0a471f0bf1
--- /dev/null
+++ b/.agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.zh.md
@@ -0,0 +1,53 @@
+# Agent Note:ACP 作为仅面向自动化的协议
+
+Status: implemented
+
+[English](2026-07-23-acp-automation-only-protocol.md) | 中文
+
+## 问题
+
+ACP(Agent Client Protocol)桥接层已经变成第二套交互式产品 UI。它将持久事件转换为编辑器卡片、终端元数据、diff、计划、标题、推理、命令、模式、模型和权限选择器、会话导航以及面向人类的询问。这些职责与 TUI 和 Web 客户端重复,同时将自动化传输层与 UI 服务、持久化查询、展示策略和编辑器特定约定耦合在一起。
+
+ACP 仍有一个有用的职责:另一个 agent(智能体)或自动化控制器可以启动 harness 进程、创建隔离会话、发送文本、接收已提交的回答、取消工作并回答权限请求。跨进程 ACP subagent 后端依赖这个标准协议边界。
+
+快照套件使移除工作更复杂。大多数 ACP 场景测试的是组装后的 agent 后端,而不是 ACP 展示层;如果随编辑器桥接层一起删除整个套件,就会丢失大量无密钥行为覆盖。
+
+## 决策
+
+`@deepseek-ai/dsh-acp` 是位于 [`packages/acp/acp`](../../../../packages/acp/acp/README.md) 下、独立于 `ui` 包组的自动化传输层。其公开协议特意保持精简:版本协商、全新文本会话(每个会话最多允许一个进行中的提示词)、已提交的助手文本更新、按会话取消、并发会话,以及由连接负责的资源清理。提示词只承载规范要求的基线内容——文本,加上被展平为方括号文本引用的资源链接;桥接层会拒绝附加目录、MCP 服务器、超出基线的提示词内容(图片、音频、内嵌资源)、空提示词、未知会话和重叠提示词。
+
+桥接层只发出已提交的 `assistant/message` 文本。推理、原始分片、工具活动、待办事项、计划、标题、重试标记、终端元数据、diff、位置和资源链接仍保留在持久会话日志或 UI 专用传输层中。它不提供会话加载、列出与删除、命令、模式、配置选择器、模型切换、plan 评审或面向人类的询问。
+
+保留一次性 `session/request_permission`。它是为桥接层拥有的 agent 提供的机器策略通道,而不是面向人类的审批 UI:应答者只接受桥接层当前会话映射中的精确 agent 对象;外部请求或缺少调用标识的请求会继续委派;RPC 失败则映射为失败关闭的 `unavailable` 结果。客户端可选择允许一次、拒绝一次或取消,桥接层绝不会将该响应转换为持久授权。询问策略仍归审批 seam 及其生产者所有;[`dsh-subagent-acp`](../../../../packages/subagent/subagent-acp/README.md) 会以程序化方式使用该通道。
+
+应用组装包含 agent 主干、持久化、检查点策略和 ACP 传输层。它不会为 ACP 挂载命令、会话查询、会话引用、plan mode、权限选择器或用户交互服务。SDK 脚手架同样将 `ask_user_question` 视为 TUI 专属功能。
+
+传输层调用 agent、会话和审批的接口服务,而不依赖具体的 agent loop。工具执行仍留在 harness 内;ACP 绝不会把 shell 执行委派给编辑器。stdout 只承载分帧 JSON-RPC,因此 app 不挂载 stdout logger,桥接层也不会 monkey-patch 进程输出。
+
+断开连接与插件 dispose(资源释放)共享同一个经记忆化处理的静止边界。传输关闭无论成功还是失败,都会将待处理提示词以已取消状态结算,dispose 每个由桥接层拥有的 agent,并等待循环和会话清理完成。创建流程如果在与关闭的竞态中落败,就会 dispose 其尚未发布的 handle。
+
+## 快照边界
+
+ACP 快照套件仍会启动组装后的 ACP 示例,并保留用于锁定后端行为的场景。从该套件移出的只有通过已删除的 UI 方法驱动的场景;由于 ACP 不再加载会话,语义检查点恢复通过 headless `stream-json` 示例执行。
+
+协议与生命周期测试会锁定停止原因编解码器和提示词编解码器、版本协商、新会话创建、文本与资源链接展平、拒绝空提示词或不受支持的提示词、精确 agent 权限归属、多会话隔离、提示词结算、按会话取消、传输关闭失败、ACP 专属重载清理,以及拆卸完全停稳。构建产物冒烟测试与真实 stdio 冒烟测试会拒绝混入 stdout 的额外输出。`session/new` 中在真实 stdio 关闭竞态中落败的分支仍属于覆盖豁免,因为内存传输层无法复现这一顺序;该分支会 dispose 尚未发布的 handle,而周边 dispose 测试会锁定无遗留资源不变式。
+
+## 考虑过的替代方案
+
+**在 Web 达到同等能力前,继续将 ACP 作为编辑器 UI。** 不予采用,因为这会留下两套需要演进的交互契约,并使编辑器约定继续存在于自动化边界中。
+
+**通过严格的 seam 保留早期编辑器桥接层。** 不予采用,尽管该桥接层正确使用了接口服务、工具自有的 render intent、审批与用户交互应答者、harness 自有执行,以及保持 stdout 纯净的组装。其终端卡片是经过能力门控、仅用于展示的 Zed `_meta` 投影,并提供文本回退,而非使用 ACP `terminal/create`,因此 shell 执行从未离开 harness。该投影从稳定的逐调用 id 派生每个展示用终端 id,以避免冲突;由于纯结果展示器接收的是内容块,而不是结构化退出信息,它会从渲染后的状态标记中恢复退出码或信号。标记往返测试和显式的无能力 `console` 回退测试锁定了这两项契约。这些边界保持一致,却无法让编辑器卡片、会话导航、配置选择器和面向人类的询问成为自动化协议应有的职责。
+
+**用私有 subagent RPC 替换 ACP。** 不予采用,因为 ACP 已经提供类型化、可互操作的进程协议,并由跨进程 subagent 后端使用。
+
+**随其他交互功能一起移除机器权限请求。** 不予采用,因为自动化父 agent 必须回答子 agent 的一次性策略决策;这是 agent 之间的控制流,而不是展示层。
+
+**删除 ACP 快照套件,或在本次变更中迁移每个场景。** 不予采用,因为大多数场景测试后端且仍有价值,而完整的 harness 迁移是一项独立的测试变更。只有驱动脚本依赖已删除 UI 方法的场景才离开该套件。
+
+## 结果
+
+ACP 具有适合 agent 与自动化的精简契约,而 TUI 和 Web 拥有面向人类的交互与展示。该包注入的服务、依赖、协议分支和生命周期状态更少,也不再将自身定位为通用编辑器入口。
+
+自动化客户端收到完整的已提交文本,而不是 token 增量或结构化工具 UI。当它们需要推理、工具跟踪信息、标题或更丰富的状态时,需要查看持久日志或其他 API。只支持全新会话也意味着,需要浏览持久会话或恢复会话的调用方必须使用 host API,而不是 ACP。
+
+因此,后端快照覆盖仍与 ACP 传输层耦合,尽管对于受测行为而言,该传输层只是附带因素。
diff --git a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.i18n.yaml b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.i18n.yaml
index 15551560ba..feeadfed91 100644
--- a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.i18n.yaml
+++ b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.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
-2026-06-19-acp-snapshot-tests.md: 43900632c4e5e3904c2d0e3b75f2a3fe3b3a50cc
-2026-06-19-acp-snapshot-tests.zh.md: a4613b19001afe87c1f6bdb61196fabbfa947a4c
+2026-06-19-acp-snapshot-tests.md: b4cda8f32fe7a84a977bcbdbe5db0671cb9a7083
+2026-06-19-acp-snapshot-tests.zh.md: 5337c3852b524af4e8c556e93ec80084b30a6d0b
diff --git a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md
index 43900632c4..b4cda8f32f 100644
--- a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md
+++ b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.md
@@ -6,7 +6,7 @@ English | [中文](2026-06-19-acp-snapshot-tests.zh.md)
## Problem
-Unit tests do not exercise the complete ACP subprocess transcript, while real-API tests are nondeterministic and key-gated. Editor-facing `session/update` output can therefore regress despite green unit coverage, as the [default-export postmortem](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md) demonstrated.
+Unit tests do not exercise the complete assembled-agent subprocess or its ACP automation wire, while real-API tests are nondeterministic and key-gated. Loader wiring, backend behavior, and protocol output can therefore regress despite green unit coverage, as the [default-export postmortem](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md) demonstrated.
The blocker for a full-transcript test is the model: the agent's output is driven by a non-deterministic LLM, and a key-gated test that hits the real API on every run is neither deterministic nor CI-runnable. We want the fidelity of a real run with the determinism of a fixture.
@@ -52,10 +52,10 @@ Replay uses a `cordis.snapshot.yml` overlay that replaces the real adapter with
A snapshot run asserts **two** normalized surfaces, because the harness's external surfaces are distinct:
-1. The **stdout transcript** — the framed `session/update` JSON-RPC the editor sees. Catches regressions in the ACP bridge's event→update translation (`streamSessionEventUpdate`). Compared against a committed `stdout.expected.jsonl`.
+1. The **stdout transcript** — the framed ACP JSON-RPC responses and committed-message updates an automation client receives. It catches regressions in the transport contract and is compared against a committed `stdout.expected.jsonl`.
2. The **re-persisted session JSONL**, normalized and compared with `session.jsonl`. The same fixture is both replay source and expected log. Prompt text is scrubbed; one scenario per header class pins readable prompt and tool content as described in the [header-pinning Agent Note](2026-07-06-pin-request-header-content-in-one-scenario.md). Override scenarios derive model behavior solely from their sidecar.
-The surfaces are complementary: stdout covers bridge projection, while JSONL covers loop, tool, and boundary structure that the projection omits.
+The surfaces are complementary: stdout covers the minimal automation wire, while JSONL covers loop, tool, and boundary structure that the wire intentionally omits.
Normalization replaces session, cwd, protocol-id, timestamp, path, and process volatility while preserving deterministic sequence numbers. Scenarios constrain real bash use to stable commands. The stdout expected output remains wire-shaped JSONL and every raw line must parse as JSON. Vitest updates only the stdout expected output; normalized session equality never overwrites the replay fixture.
@@ -79,6 +79,6 @@ Tool determinism comes from a generated cwd, scrubbed environment, fresh non-log
## Consequences
-The new tier adds reviewed per-scenario input, session, stdout, optional override, and optional workspace fixtures. Workspace seeds are copied into the generated cwd for both record and replay. In return the tier provides deterministic keyless transcript coverage through the real Loader and tool composition. The subprocess, input, workspace, normalization, and replay harness can support examples beyond ACP.
+The tier adds reviewed per-scenario input, session, stdout, optional override, and optional workspace fixtures. Workspace seeds are copied into the generated cwd for both record and replay. In return the tier provides deterministic keyless coverage through the real Loader and tool composition. Most retained scenarios exercise the assembled backend rather than ACP; the [automation-only ACP decision](../simplification/2026-07-23-acp-automation-only-protocol.md#snapshot-boundary) keeps that corpus here and defers any move to a transport-neutral headless suite as an independent testing change (the suite-level FIXME marks it).
-This Agent Note relates to but does not supersede the [proposed determinism Agent Note](../../proposed/testing/2026-06-11-deterministic-and-stress-testing.md): that proposal's "universal replay fixture" re-derives session *message history* after every test (an internal-consistency invariant), whereas snapshot tests pin the *external protocol output*. They are complementary — one guards the event-sourcing invariant, the other guards the editor-facing contract.
+This Agent Note relates to but does not supersede the [proposed determinism Agent Note](../../proposed/testing/2026-06-11-deterministic-and-stress-testing.md): that proposal's "universal replay fixture" re-derives session *message history* after every test (an internal-consistency invariant), whereas these snapshots pin assembled behavior plus the external automation output. They are complementary until the backend corpus moves off ACP.
diff --git a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.zh.md b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.zh.md
index a4613b1900..5337c3852b 100644
--- a/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.zh.md
+++ b/.agents/notes/implemented/testing/2026-06-19-acp-snapshot-tests.zh.md
@@ -6,9 +6,9 @@ Status: implemented
## 问题
-单元测试不会覆盖完整的 ACP(Agent Client Protocol)子进程 transcript(文本记录),而真实 API 测试不具确定性且受密钥门控。因此,即使单元覆盖率为绿色,面向编辑器的 `session/update` 输出仍可能回归,[默认导出事后分析](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md)已经证明了这一点。
+单元测试不会覆盖组装后的完整 agent(智能体)子进程及其 ACP(Agent Client Protocol)自动化线协议,而真实 API 测试不具确定性且受密钥门控。因此,即使单元覆盖率为绿色,Loader 接线、后端行为和协议输出仍可能回归,[默认导出事后分析](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md)已经证明了这一点。
-全 transcript 测试的阻塞因素在于模型:agent(智能体)的输出由非确定性的 LLM(大语言模型)驱动,而每次运行都命中真实 API 的密钥门控测试既不确定也无法在 CI 中运行。我们需要真实运行的保真度与 fixture(测试前置数据)的确定性兼得。
+全 transcript(文本记录)测试的阻塞因素在于模型:agent 的输出由非确定性的 LLM(大语言模型)驱动,而每次运行都命中真实 API 的密钥门控测试既不确定也无法在 CI 中运行。我们需要真实运行的保真度与 fixture(测试前置数据)的确定性兼得。
本 Agent Note(agent 决策记录)记下了新增第三层测试——**快照测试**——的决策,以及让它具备确定性、在 CI 中无需密钥、且维护成本低廉的设计选择。
@@ -52,10 +52,10 @@ Status: implemented
快照运行断言**两个**归一化后的表面,因为 harness 的外部表面是不同的:
-1. **stdout transcript**——编辑器看到的、经过 framing 的 `session/update` JSON-RPC。用于捕获 ACP bridge 中事件→更新转换(`streamSessionEventUpdate`)的回归。与已提交的 `stdout.expected.jsonl` 比较。
+1. **stdout transcript**——自动化客户端收到的、经过 framing 的 ACP JSON-RPC 响应与已提交的消息更新。它捕获传输契约的回归,与已提交的 `stdout.expected.jsonl` 比较。
2. **重新持久化的会话 JSONL**,经过规范化后与 `session.jsonl` 比较。同一 fixture 同时作为重放来源和预期日志。提示词文本会被清理;按照[请求头固定 Agent Note](2026-07-06-pin-request-header-content-in-one-scenario.md)所述,每种请求头类别由一个场景固定可读提示词与工具内容。Override 场景仅从其 sidecar 派生模型行为。
-两个表面互补:stdout 覆盖 bridge 投影,JSONL 覆盖投影所省略的 loop、工具和 boundary 结构。
+两个表面互补:stdout 覆盖精简的自动化线协议,JSONL 覆盖线协议有意省略的 loop、工具和 boundary 结构。
规范化会替换会话、cwd、协议 id、时间戳、路径和进程易变值,同时保留确定性序号。场景把真实 bash 使用限制在稳定命令上。stdout 预期输出仍是线协议形状的 JSONL,每个原始行都必须可解析为 JSON。Vitest 只更新 stdout 预期输出;规范化会话相等性检查从不覆盖重放 fixture。
@@ -79,6 +79,6 @@ Status: implemented
## 后果
-新测试层为每个场景增加经过评审的输入、会话、stdout、可选 override 和可选 workspace fixture。记录与重放都会把 workspace seed 复制到生成的 cwd。作为回报,该层通过真实 Loader 和工具组合提供确定性的无密钥 transcript 覆盖。子进程、输入、workspace、规范化和重放 harness 也可以支持 ACP 之外的示例。
+该测试层为每个场景增加经过评审的输入、会话、stdout、可选 override 和可选 workspace fixture。记录与重放都会把 workspace seed 复制到生成的 cwd。作为回报,该层通过真实 Loader 和工具组合提供确定性的无密钥覆盖。保留下来的大多数场景测试的是组装后的后端而非 ACP;[仅面向自动化的 ACP 决策](../simplification/2026-07-23-acp-automation-only-protocol.md#snapshot-boundary)将该语料保留在此处,并把向传输无关 headless 套件的任何迁移推迟为一项独立的测试变更(套件级 FIXME 标记了这一点)。
-本 Agent Note 与[拟议的确定性 Agent Note](../../proposed/testing/2026-06-11-deterministic-and-stress-testing.md)相关,但不取代它:该提案的“通用重放 fixture”在每次测试后重新派生会话*消息历史*(内部一致性不变量),而快照测试固定*外部协议输出*。两者相互补充——一个守护事件溯源不变量,另一个守护面向编辑器的契约。
+本 Agent Note 与[拟议的确定性 Agent Note](../../proposed/testing/2026-06-11-deterministic-and-stress-testing.md)相关,但不取代它:该提案的“通用重放 fixture”在每次测试后重新派生会话*消息历史*(内部一致性不变量),而这些快照固定组装后的行为与外部自动化输出。在后端语料迁出 ACP 之前,两者相互补充。
diff --git a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.i18n.yaml b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.i18n.yaml
index 0ca80ac88b..97313f0333 100644
--- a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.i18n.yaml
+++ b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.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
-2026-06-19-real-api-e2e-ci.md: a9289980ada8ab6390b5daa488f07ec258db1bd5
-2026-06-19-real-api-e2e-ci.zh.md: dcb5759d26e7680d392a4855665b2eb88551ba79
+2026-06-19-real-api-e2e-ci.md: 935664fd01df4844ee19be7b4f2f297ebf5bd29b
+2026-06-19-real-api-e2e-ci.zh.md: 9c10614f5a7e6b38f6850b29fab87d0e09806c5f
diff --git a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.md b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.md
index a9289980ad..935664fd01 100644
--- a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.md
+++ b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.md
@@ -6,7 +6,7 @@ English | [中文](2026-06-19-real-api-e2e-ci.zh.md)
## Problem
-The harness leans hard on real-API tests by policy: [docs/testing.md](../../../../docs/testing.md) argues that a no-key suite proves the plumbing but not the product, and the [ACP inject postmortem](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md) is the standing proof — 178 keyless tests stayed green while a real editor session crashed instantly. The real-API e2e suite (`pnpm run test:e2e`, the `*.e2e.ts` files) exists precisely to close that gap: it drives the agent against the live DeepSeek API — real model calls, real bash tools, multi-turn, resume, ACP-over-stdio.
+The harness leans hard on real-API tests by policy: [docs/testing.md](../../../../docs/testing.md) argues that a no-key suite proves the plumbing but not the product, and the [ACP inject postmortem](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md) is the standing proof — 178 keyless tests stayed green while a real ACP client session crashed instantly. The real-API e2e suite (`pnpm run test:e2e`, the `*.e2e.ts` files) exists precisely to close that gap: it drives the agent against the live DeepSeek API — real model calls, real bash tools, multi-turn, resume, ACP-over-stdio.
The default gate ([.github/workflows/ci.yml](../../../../.github/workflows/ci.yml)) is deliberately keyless: it carries no secret and runs for forks. `test:e2e` self-skips without a key (`describe.skipIf(!process.env.DEEPSEEK_API_KEY)`), so adding it there would report green without exercising the real suite. A separate secret-bearing workflow is required to make real-API coverage a merge signal.
diff --git a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.zh.md b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.zh.md
index dcb5759d26..9c10614f5a 100644
--- a/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.zh.md
+++ b/.agents/notes/implemented/testing/2026-06-19-real-api-e2e-ci.zh.md
@@ -6,7 +6,7 @@ Status: implemented
## 问题
-根据策略,harness 高度依赖真实 API 测试:[docs/testing.md](../../../../docs/testing.md) 指出,无密钥套件证明的是管线,而非产品;[ACP(Agent Client Protocol)inject 事后分析](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md)则是常设证据——178 项无密钥测试保持绿色时,真实编辑器会话却立即崩溃。真实 API e2e 套件(`pnpm run test:e2e`,即 `*.e2e.ts` 文件)的存在正是为了弥合这一缺口:它针对实时 DeepSeek API 驱动 agent(智能体)——真实模型调用、真实 bash 工具、多轮次、恢复、ACP-over-stdio。
+根据策略,harness 高度依赖真实 API 测试:[docs/testing.md](../../../../docs/testing.md) 指出,无密钥套件证明的是管线,而非产品;[ACP(Agent Client Protocol)inject 事后分析](../../../../docs/postmortem/0001-acp-default-export-drops-inject.md)则是常设证据——178 项无密钥测试保持绿色时,真实 ACP 客户端会话却立即崩溃。真实 API e2e 套件(`pnpm run test:e2e`,即 `*.e2e.ts` 文件)的存在正是为了弥合这一缺口:它针对实时 DeepSeek API 驱动 agent(智能体)——真实模型调用、真实 bash 工具、多轮次、恢复、ACP-over-stdio。
默认门禁([.github/workflows/ci.yml](../../../../.github/workflows/ci.yml))刻意无密钥:不携带 secret,可供 fork 运行。`test:e2e` 在无密钥时自动跳过(`describe.skipIf(!process.env.DEEPSEEK_API_KEY)`),因此将其加入该工作流只会报绿而不会真正执行真实套件。要让真实 API 覆盖率成为合并信号,需要一个独立的、携带 secret 的工作流。
diff --git a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.i18n.yaml b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.i18n.yaml
index 379054599b..65e49bd193 100644
--- a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.i18n.yaml
+++ b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.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
-2026-06-22-subagent-snapshot-replay.md: 4aad8b6ddd3e4f8b8ad5565e10b263953d81ea31
-2026-06-22-subagent-snapshot-replay.zh.md: 96890bc98a64357d9785b8081c8ba123bb3c716e
+2026-06-22-subagent-snapshot-replay.md: 8cd7bc86e07af9ed274c18574b575b9070854e88
+2026-06-22-subagent-snapshot-replay.zh.md: eae78129405fedd03c2c579845c07c6e5694cc30
diff --git a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.md b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.md
index 4aad8b6ddd..8cd7bc86e0 100644
--- a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.md
+++ b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.md
@@ -6,12 +6,12 @@ English | [中文](2026-06-22-subagent-snapshot-replay.zh.md)
## Problem
-The snapshot tier (`pnpm run test:snapshot`) boots the real `acp-agent` subprocess, replays a recorded session through [`dsh-llm-replay`](../../../../packages/support/llm-replay), and diffs the normalized stdout transcript + re-persisted session log against committed expected outputs. It is the only tier that exercises the full editor-facing transcript end to end.
+The snapshot tier (`pnpm run test:snapshot`) boots the real `acp-agent` subprocess, replays a recorded session through [`dsh-llm-replay`](../../../../packages/support/llm-replay), and diffs the normalized automation wire + re-persisted session log against committed expected outputs. Most scenarios exercise assembled backend behavior through that real process boundary.
It was built for ONE session per process, and that assumption is wired into two places:
- **`dsh-llm-replay` keyed nothing.** It served the Nth `llm/stream` call the Nth recorded entry from a single global cursor. With a parent agent AND an in-process subagent both streaming on one context, the calls interleave and the single cursor hands the child the parent's script (and vice versa).
-- **The harness harvested one log.** `findSessionLog` walked the sessions root and returned the FIRST `.jsonl` it found. A subagent runs as a second `Session` with its own log in the same cwd bucket, so the child's transcript was silently dropped.
+- **The harness harvested one log.** `findSessionLog` walked the sessions root and returned the FIRST `.jsonl` it found. A subagent runs as a second `Session` with its own log, so the child's transcript was silently dropped.
This was the `TODO(subagent-snapshots)` deferral recorded in the [subagent seam Agent Note](../feature/2026-06-21-subagent-capability-seam.md): the in-process backends (PR2) shipped with unit + e2e coverage, but the full-transcript snapshot tier could not express a nested-agent shape until this infrastructure landed. This Agent Note is that stacked follow-up.
@@ -39,7 +39,7 @@ The alternative considered and rejected was a **call-ordered merge of the parent
### 3. The harness harvests every log, primary-first
-`harvestSessionLogs` collects every `.jsonl` across every cwd bucket under the sessions root (the JSONL backend puts a parent and its same-cwd child in the same bucket), parses each header, and orders them primary-first: the top-level session (no `parentSession`) leads, then each child by ascending `createdAt`. `RunResult.sessionLogs` is the plural result; the spec writes each back to its fixture on record (`session.jsonl` + `session..jsonl`) and diffs each harvested log against its fixture on replay. The normalizer already accepted plural session ids and collapses any stray UUID, so no normalizer change was needed.
+`harvestSessionLogs` recursively collects every fixed `session.jsonl` transcript under the sessions root (the JSONL backend gives each parent and child its own project/session directory), parses each header, and orders them primary-first: the top-level session (no `parentSession`) leads, then each child by ascending `createdAt`. `RunResult.sessionLogs` is the plural result; the spec writes each back to its fixture on record (`session.jsonl` + `session..jsonl`) and diffs each harvested log against its fixture on replay. The normalizer already accepted plural session ids and collapses any stray UUID, so no normalizer change was needed.
### 4. Scenarios
diff --git a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.zh.md b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.zh.md
index 96890bc98a..eae7812940 100644
--- a/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.zh.md
+++ b/.agents/notes/implemented/testing/2026-06-22-subagent-snapshot-replay.zh.md
@@ -6,12 +6,12 @@ Status: implemented
## 问题
-快照层(`pnpm run test:snapshot`)会启动真实 `acp-agent` 子进程,通过 [`dsh-llm-replay`](../../../../packages/support/llm-replay) 重放已记录会话,并将规范化 stdout transcript(文本记录)+ 重新持久化的会话日志与已提交预期输出进行 diff。它是唯一端到端覆盖完整面向编辑器 transcript 的测试层。
+快照层(`pnpm run test:snapshot`)会启动真实 `acp-agent` 子进程,通过 [`dsh-llm-replay`](../../../../packages/support/llm-replay) 重放已记录会话,并将规范化后的自动化线协议 + 重新持久化的会话日志与已提交预期输出进行 diff。大多数场景通过这条真实进程边界测试组装后的后端行为。
该层最初为每个进程只有一个会话而构建,这一假设硬编码在两处:
- **`dsh-llm-replay` 没有做任何键控。** 它用一个全局游标,将第 N 次 `llm/stream` 调用对应到单一录制序列的第 N 条。当父 agent(智能体)和一个进程内 subagent 在同一个上下文上同时流式输出时,调用交错,单一游标会把子 agent 的脚本发给父 agent(反之亦然)。
-- **harness 只收集一份日志。** `findSessionLog` 遍历 sessions 根目录,返回找到的第一个 `.jsonl`。subagent 作为第二个 `Session` 运行,在同一个 cwd bucket 下有自己的日志,因此子 agent 的 transcript 被静默丢弃。
+- **harness 只收集一份日志。** `findSessionLog` 遍历 sessions 根目录,返回找到的第一个 `.jsonl`。subagent 作为第二个 `Session` 运行并拥有自己的日志,因此子 agent 的 transcript(文本记录)被静默丢弃。
这就是 [subagent seam Agent Note(agent 决策记录)](../feature/2026-06-21-subagent-capability-seam.md)中通过 `TODO(subagent-snapshots)` 推迟的工作:进程内后端(PR2)落地时已有单元 + e2e 覆盖,但在这套基础设施落地前,完整 transcript 快照层无法表达嵌套 agent 形状。本 Agent Note 就是该堆叠式后续工作。
@@ -39,7 +39,7 @@ Status: implemented
### 3. harness 收集所有日志,主会话优先
-`harvestSessionLogs` 收集 sessions 根目录下每个 cwd bucket 中的所有 `.jsonl`(JSONL 后端将父会话与同 cwd 的子会话放在同一个 bucket),解析各自的 header,并按主会话优先排序:顶层会话(无 `parentSession`)在前,各子会话按 `createdAt` 升序排列。`RunResult.sessionLogs` 是复数结果;spec 在录制时将每份日志写回对应 fixture(`session.jsonl` + `session..jsonl`),在回放时将每份收集到的日志与其 fixture 做 diff。归一化器已支持复数会话 id 并会折叠任何游离 UUID,因此无需修改归一化器。
+`harvestSessionLogs` 递归收集 sessions 根目录下所有固定命名为 `session.jsonl` 的 transcript(JSONL 后端为每个父会话和子会话分别提供独立的项目/会话目录),解析各自的 header,并按主会话优先排序:顶层会话(无 `parentSession`)在前,各子会话按 `createdAt` 升序排列。`RunResult.sessionLogs` 是复数结果;spec 在录制时将每份日志写回对应 fixture(`session.jsonl` + `session..jsonl`),在回放时将每份收集到的日志与其 fixture 做 diff。归一化器已支持复数会话 id 并会折叠任何游离 UUID,因此无需修改归一化器。
### 4. 场景
diff --git a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.i18n.yaml b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.i18n.yaml
index 743d7ecbf0..83bd9197d8 100644
--- a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.i18n.yaml
+++ b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.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
-2026-07-04-hook-snapshot-matrix.md: ceb91a70e1f9582cf2a7cb4cec0ba8aaf5699b8e
-2026-07-04-hook-snapshot-matrix.zh.md: 250995dd1aae566aa5b9d40a14ed281b14fbfe0b
+2026-07-04-hook-snapshot-matrix.md: 98f9db27fd8afaaa99c9985dff4d148ef4eb926b
+2026-07-04-hook-snapshot-matrix.zh.md: 40b9ee84ad7232ac806096375e8da42bd02bfddf
diff --git a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.md b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.md
index ceb91a70e1..98f9db27fd 100644
--- a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.md
+++ b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.md
@@ -8,7 +8,7 @@ English | [中文](2026-07-04-hook-snapshot-matrix.zh.md)
The hook bridges — [`dsh-hooks-claude`](../../../../packages/hooks/hooks-claude) (7 Claude Code hook points) and [`dsh-hooks-codex`](../../../../packages/hooks/hooks-codex) (5 Codex points) — map external hook commands onto the harness interception seams. They carry deep unit and coverage-spec coverage (every decision arm, every payload dialect, driven against a mocked seam) plus one key-gated e2e (`hooks.e2e.ts`, a live `PreToolUse` block). But the full-transcript snapshot tier — the one net that boots the real `acp-agent` subprocess, replays a recorded session keyless, and diffs the normalized ACP stdout + re-persisted log against committed expected outputs — covered exactly ONE hook: a Claude `UserPromptSubmit` block (`hook-cc-promptsubmit-block`).
-That is the tier a mocked unit test structurally cannot be: it exercises the REAL bridge translating a REAL hook process's outcome into the REAL seam decision, then the REAL loop's reaction, rendered exactly as an editor sees it. A bridge-translation or loop-structure regression that left every unit green would still escape it for every hook point but one — and for the Codex bridge, the ACP example did not even LOAD it, so no Codex hook could fire end-to-end at all.
+That is the tier a mocked unit test structurally cannot be: it exercises the REAL bridge translating a REAL hook process's outcome into the REAL seam decision, then the REAL loop's reaction through the automation wire and persisted log. A bridge-translation or loop-structure regression that left every unit green would still escape it for every hook point but one — and for the Codex bridge, the ACP example did not even LOAD it, so no Codex hook could fire end-to-end at all.
## Decision
diff --git a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.zh.md b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.zh.md
index 250995dd1a..40b9ee84ad 100644
--- a/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.zh.md
+++ b/.agents/notes/implemented/testing/2026-07-04-hook-snapshot-matrix.zh.md
@@ -8,7 +8,7 @@ Status: implemented
钩子 bridge——[`dsh-hooks-claude`](../../../../packages/hooks/hooks-claude)(7 个 Claude Code 钩子点)和 [`dsh-hooks-codex`](../../../../packages/hooks/hooks-codex)(5 个 Codex 点)——把外部钩子命令映射到 harness 拦截 seam。它们有深入的单元与覆盖率规格覆盖(每个决策分支、每种 payload dialect,针对 mock seam 驱动),外加一个受密钥门控的 e2e(`hooks.e2e.ts`,实时 `PreToolUse` 阻止)。但完整 transcript(文本记录)快照层——会启动真实 `acp-agent` 子进程、无需密钥重放已记录会话,并将规范化 ACP(Agent Client Protocol)stdout + 重新持久化日志与已提交预期输出进行 diff 的那张网——只覆盖了一个钩子:Claude `UserPromptSubmit` 阻止(`hook-cc-promptsubmit-block`)。
-这正是 mock 单元测试在结构上无法替代的层级:它验证的是真实 bridge 将真实钩子进程的结果翻译到真实 seam 决策,再到真实 agent loop(智能体循环)的反应,渲染结果与编辑器看到的完全一致。一个 bridge 翻译或 loop 结构的回归,即使让所有单元测试保持绿色,也会在除那一个钩子点之外的所有点上逃逸;而对于 Codex bridge,ACP 示例甚至没有加载它,因此没有任何 Codex 钩子能端到端触发。
+这正是 mock 单元测试在结构上无法替代的层级:它验证的是真实 bridge 将真实钩子进程的结果翻译到真实 seam 决策,再经由自动化线协议和持久化日志检验真实 agent loop(智能体循环)的反应。一个 bridge 翻译或 loop 结构的回归,即使让所有单元测试保持绿色,也会在除那一个钩子点之外的所有点上逃逸;而对于 Codex bridge,ACP 示例甚至没有加载它,因此没有任何 Codex 钩子能端到端触发。
## 决策
diff --git a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.i18n.yaml b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.i18n.yaml
similarity index 63%
rename from .agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.i18n.yaml
rename to .agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.i18n.yaml
index 41265a5b0f..ef389fecb3 100644
--- a/.agents/notes/implemented/architecture/2026-06-11-custom-schema-dsl.i18n.yaml
+++ b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.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
-2026-06-11-custom-schema-dsl.md: 947d53555df078bfa9f3dac48eab4b8c0074007c
-2026-06-11-custom-schema-dsl.zh.md: 26ebfe2fb15a6c034e809b3f51187342fa500193
+2026-07-24-web-gui-browser-e2e-lane.md: fb870c4bb2c85d8be7ac11f9f29f05bf24f446a4
+2026-07-24-web-gui-browser-e2e-lane.zh.md: c43e526d27fd4d8cf4f774e8a03480930682041d
diff --git a/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.md b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.md
new file mode 100644
index 0000000000..fb870c4bb2
--- /dev/null
+++ b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.md
@@ -0,0 +1,90 @@
+# Agent Note: Keyless browser e2e lane for the web GUI
+
+Status: implemented
+
+English | [中文](2026-07-24-web-gui-browser-e2e-lane.zh.md)
+
+## Problem
+
+The web GUI ships as a real assembled chain — chromium page → client plugin bundles → HTTP unary RPC + two SSE streams → `toFetchHandler`/apiproxy → the host agent loop, tools, and JSONL persistence — and no test exercised that chain keylessly and deterministically. The [GUI testing system](../process/2026-07-20-gui-testing-system.md) covers tier 1 (wire isomorphism in node), tier 2 (object-layer state machines), and tier-3 smokes, but the keyless smoke drives `FixtureApiClient` — no host, no wire, no agent loop — while the full-chain smoke needs `DEEPSEEK_API_KEY` and a live model, so it is nondeterministic and self-skips in keyless CI. The snapshot philosophy of [docs/testing.md](../../../../docs/testing.md) — record once with a key, replay forever keyless, refresh on format churn — already covers the ACP, headless `stream-json`, and TUI transcript surfaces; the web surface was the one assembled product shape without it. The gap is exactly where the two confirmed GUI P0s hid: the wire carriage chain the fixture client short-circuits.
+
+## Decision
+
+`pnpm run test:web` carries a keyless, deterministic browser e2e lane under `apps/web/tests/`: recorded session-log fixtures replayed through `@deepseek-ai/dsh-llm-replay` against the real in-process web composition, asserting a normalized conversation aria golden plus in-process world state. No new package; the product deltas are two additive `dsh-llm-replay` surfaces (`paceMs`, `ReplayHandle`).
+
+### Scaffold: `apps/web/tests/scaffold.ts`
+
+A plain shared-fixture module (the [testing-policy sanctioned shape](../../../../docs/testing.md)), not a package: the gate-worthy logic — replay derivation, session parsing, log scrubbing, persistence — lives in the gated packages `dsh-llm-replay`, `dsh-acp-snapshot`, and `dsh-session-persistence-jsonl`; what remains is boot wiring and browser glue, and chromium-driving code cannot hold per-file 100% coverage on the browserless coverage runners.
+
+`launchWebScaffold()` boots the real web composition from the shipped `apps/cli/cordis.yml` through the vendored Loader's include mechanism — the same tree and mechanism `AppCLIEntry` drives for `dsh web`. Divergences ride include patches over that tree, the ACP `cordis.snapshot.yml` pattern expressed in-process: temp `persistenceRoot`, `workspace-context` disabled (recorded fixtures must not embed this repo's AGENTS.md), `session-title-llm` disabled (its fire-and-forget title call would race the loop for the session's replay cursor), the webserver row pinned to port 0 with the built dist, and in keyless modes `llm-deepseek` disabled. A patch id that stops matching a row fails the boot sweep loudly instead of drifting. The boot runs `chdir`'d to the temp workspace so the api-gateway's `process.cwd()` session default, tool cwds, and fixtures agree; the `dsh web` bin's own glue (argv, profile json, AppCLIEntry) stays held by the keyless CLI smokes in `smoke-real.e2e.ts`. Setup rollback and ordinary close both dispose the Cordis tree before removing the two owned temp roots, attempt every cleanup independently, and report cleanup failures without masking the setup failure.
+
+Keyless model displacement is the disabled adapter row plus `installLlmReplay` filling the open seam on the settled root ctx in providers-catalog mode — never catch-all: with the adapter row disabled no adapter exists, so catch-all would leave `resolveModelContext` unroutable and `compact-basic`'s post-step pressure check would warn every step instead of being provably inert (the published 128k `contextWindow` keeps it inert for small fixtures). The direct install rather than an inserted replay plugin row is deliberate: it returns the `ReplayHandle` the teardown consumption check needs. A scenario with no fixture leaves the seam empty, so a stray stream fails loud with NO_ADAPTER.
+
+`seedSession()` seeds cold sessions through the real persistence API — a throwaway `Context` mounting `SessionStore` + `SessionPersistenceJsonl` against the host's root, `create()` + `append()`, one `utimes` backdate for deterministic sidebar order (the `semantic-checkpoint.snapshot.ts` precedent) — never raw file writes, so the seeder knows nothing of bucket hashing, filename encoding, or compression, and the host's zstd default needs no boot knob. Seeds are validated at seed time (parseable, ending in `turn/end` — an open final turn would be mutated by resume's crash repair).
+
+### Determinism rules
+
+The barrier stack for replay-mode browser assertions is, in order: (1) host-side `await agent.whenIdle()` under a timeout, keyed off the in-process `turn/end` — the idle flip follows the persistence flush, so one await covers turn completion and durability; (2) browser settled poll (streaming detached, final text visible). Record-mode log harvest runs after `whenIdle()` and before scaffold disposal while the live session remains available. An in-process `turn/end` listener alone is a wrong barrier (it fires before the SSE frame reaches the browser and before the fsync); file polling is banned (slow on NFS, superseded by `whenIdle`); `networkidle` is banned outright (never resolves while an SSE stream is open).
+
+No single-shot transient-DOM assertions: every hop from replay yield to React commit can coalesce chunks, so sampling `[data-streaming]` is a race by construction. Streaming incrementality is asserted from the persisted `assistant/chunk` events (model-visible ⟺ logged makes the log the authoritative proof). `dsh-llm-replay`'s opt-in `paceMs` (default absent = burst) is a realism knob so the browser observes genuinely incremental SSE; correctness never leans on it, and abort during a pace wait cancels promptly.
+
+Every scenario fails on any pageerror and on the client's connection-loss/gap-repair console warnings: the reconnect machine plus history resync would otherwise self-heal a dead SSE path and the suite would certify a broken wire. Scaffold `close()` calls the `ReplayHandle.assertConsumed()` teardown check (every recorded script bound, every cursor drained), converting silent underruns and shifted bindings into crisp diagnostics. No vitest retry on the lane; one chromium per file, fresh context per scenario, one host per scenario; viewport pinned; interaction selectors anchor on roles, `data-*` attributes, and visible text, while the frame and conversation-region captures use the existing CSS-module local-name anchors.
+
+### Expected outputs
+
+One committed golden per scenario: a normalized `ariaSnapshot()` of the conversation region (`ui.expected.md`) — uuid/cwd/workspace-basename/duration tokens normalized, captured poll-until-equal at the settled milestone — plus a few role/text anchor assertions that stay green under a semantics-preserving component rewrite while the golden churns reviewably. The aria tree is the mechanization of the client rule "assert what the user would see, never class names". World-state assertions ride root-context session events inline (which tool call produced which durable result, whether `turn/end` completed) instead of a second committed log golden: the persisted-log surface is pinned by the ACP/headless/TUI suites through the same loop and persistence, and re-pinning it here would double refresh cost against the tier discipline. `refresh` is the sole golden writer — a missing golden in replay mode fails with the healing command rather than self-bootstrapping.
+
+The typecheck plane split is structural: the three files that boot the host spine (`scaffold`, `replay-round-trip.e2e`, and `seeded-history.e2e`) are excluded from the client-registered `apps/web` project. Those files and their shared `support.ts` are included file-by-file in `tsconfig.host.json` — one program cannot hold both sides of the cordis `Context` merges.
+
+### Modes and fixtures
+
+`DSH_SNAPSHOT` selects replay (default, keyless), record (with key), or refresh (keyless) as inline spec branches — the TUI shape, not a suite factory: at two scenarios the acp-snapshot factory machinery has no owner, and the genuinely shared parts are already exported (`scrubRequestHeaders`, `parseSessionLog`, `installLlmReplay`). Each spec splits into drive steps (type, send, `whenTurnSettled` — run in all modes, never waiting on model-content selectors, so record cannot hang on a live model answering differently) and assertion steps (replay/refresh only). Record = drive live through the real composer + harvest the in-memory `session.header`/`session.events` (the TUI `rawSessionLog` shape — no file decompression) + `scrubRequestHeaders` + `{{sessionId}}`/`{{cwd}}`/`{{rpcId}}` tokenization; a follow-up keyless refresh regenerates `ui.expected.md`. Both scenarios' fixtures were recorded against this assembly through this flow. A drift guard ties each spec's drive prompt to the fixture's recorded `user/message`. A fixture-inventory guard holds each scenario directory closed (exact file set, every JSONL a scrub fixed-point with no run-local `rpcId`). Web fixtures scrub headers everywhere and pin no header class, following the TUI precedent over the strict [pinned-header](2026-07-06-pin-request-header-content-in-one-scenario.md) reading — see Deferred.
+
+### Scenarios
+
+1. **`replay-round-trip`** — new session, prompt through the real composer, replay streams reasoning + a `bash` tool call that really executes in the temp workspace + final text (paced 15ms). Asserts settled markdown, the aria golden, and inline world state (the bash call's durable result is exactly `WEB_E2E_OK\n`, completed `turn/end`, >10 chunk events).
+2. **`seeded-history`** — a recorded session seeded cold; the sidebar lists it (group row → session row, collapsed by default), opening renders tool cards and text purely from the log through the implicit cold-resume attach inside `session.history` — zero model calls in replay, so no binding constraints; record mode drives the same turn live (real `read` tool against seeded workspace files) to produce the seed.
+
+### CI stance
+
+The lane ships gate-exempt inside `pnpm run test:web`, exactly as that config's header records. Adding chromium to CI would reverse the "no browser infrastructure in CI" premise in the [GUI testing note](../process/2026-07-20-gui-testing-system.md) and therefore requires its own Agent Note cross-linked from there, staged as a non-required job first with measured promotion criteria (consecutive green runs, wall time, zero-retry flake budget, runner browser-cache strategy). `TODO(ci-browser)` marks the seam. Scenarios are POSIX-oriented (the lane is not in the Windows matrix).
+
+## Prior art
+
+Surveyed AI-chat/agent web UIs and mocking layers (LibreChat, vercel/ai-chatbot + AI SDK, lobe-chat, open-webui, OpenHands, Chainlit, continue, cline, langfuse, gradio/streamlit; Playwright HAR/route, MSW, Polly/nock, WireMock, aimock). The dominant proven architecture for apps that own their backend is an in-process fake/replay model behind the real backend seam with everything downstream real (LibreChat's `LIBRECHAT_TEST_RUN_HOOK` fake model; ai-chatbot's `MockLanguageModelV3` + `simulateReadableStream`; continue's scripted mock provider classes) — which is what `dsh-llm-replay` already is. Browser-level SSE interception cannot exercise incremental rendering (`route.fulfill` delivers the whole body at once; playwright#33564) and leaves the server SSE stack untested, so projects use it only for edge cases. Chunk pacing as a fixture parameter recurs everywhere (LibreChat 10ms default with slow profiles; ai-chatbot 500ms); real models in CI rot (open-webui's suite grew 120-second timeouts, was disabled, then deleted); sessions are seeded at the persistence layer with controlled timestamps (LibreChat inserts backdated Mongo documents; langfuse seeds its DB). No surveyed project replays a recorded agent-event log through the real backend for UI tests — the closest are provider-level recorded fixtures (aimock) and frontend-level socket history emission (OpenHands MSW) — so the session-log-as-fixture design goes one step beyond prior art along the axis this repo's model-visible ⟺ logged invariant makes natural.
+
+## Alternatives considered
+
+**Browser-network SSE interception (`page.route`).** Rejected: `route.fulfill` cannot stream, so incremental token rendering is unexercisable and the server-side SSE/backpressure/close path — where both confirmed P0s hid — goes untested.
+
+**Mock HTTP provider at `DEEPSEEK_BASE_URL`.** Rejected as the lane's mechanism (kept for the one existing workspace-probe smoke): fixtures become hand-authored OpenAI SSE byte scripts, a second fixture format that drifts from the session-log format the rest of the repo records and replays; the adapter's real HTTP path is with-key e2e's job.
+
+**Growing the `?fixture` client.** Rejected: tier separation — `FixtureApiClient` exists to test the client shell without a server; everything below the client API seam stays untested by construction.
+
+**Placeholder `DEEPSEEK_API_KEY` + replay interception instead of disabling the adapter row.** Rejected despite zero composition change and two in-tree precedents: it satisfies `llm-deepseek`'s fail-loud key check with a lie and leaves a dead adapter mounted-but-intercepted; the disabled row (the ACP overlay's move) is honest keylessness and fails loud at the earliest resolvable point.
+
+**A `packages/support/web-snapshot` package with a `defineWebSnapshotSuite` factory.** Rejected: chromium-driving source cannot honestly hold per-file 100% coverage on browserless coverage runners, and at two scenarios a factory generalizes from one consumer while the genuinely shared logic is already exported from gated packages. Re-entry trigger: a second web-shaped consumer or ≥6 scenarios with demonstrably drifting inline branches; the package boundary would then be drawn browser-free.
+
+**A committed normalized-session-log golden as a second expected surface.** Rejected: the log surface is pinned by the ACP/headless/TUI suites through the same loop and persistence; here it would double refresh cost and re-test lower tiers. Inline world-state assertions on root-context events keep the world-verification duty.
+
+**Spawning the `dsh web` bin with a `DSH_SNAPSHOT` replay branch.** Rejected: it needs a test-only replay branch plus environment plumbing in the shipped CLI. The in-process scaffold already loads the same `apps/cli/cordis.yml`; only argv, profile JSON, and `AppCLIEntry` glue remain outside it, and the keyless CLI smokes cover those paths.
+
+**Changing the wire protocol for testability.** Rejected: the contract already has a first-class keyless isomorphic seam (`InProcessApiClient(toFetchHandler(api))`), the per-event unbatched SSE is exactly what makes replay observable in a browser, and testing a wire we no longer ship would invert the tier's purpose.
+
+**Real-model browser tests as the keyless lane.** Rejected: nondeterministic by construction; the surveyed cautionary case (open-webui) grew unbounded timeouts and was deleted. The with-key W5 smoke stays as the live-model complement.
+
+**A client `data-dsh-busy` settled signal.** Deferred: the multi-condition settled polls proved sufficient at two scenarios and the host-side `whenIdle` barrier does the heavy lifting. Re-entry trigger: the first settled-poll flake, or a scenario needing a state the DOM does not expose.
+
+## Testing
+
+The lane itself: `pnpm run test:web` runs both scenarios keylessly alongside the existing smoke pair; `DSH_SNAPSHOT=record pnpm exec vitest run --config vitest.web.config.ts apps/web/tests/` re-records a scenario's fixture against the live model; `DSH_SNAPSHOT=refresh` rewrites both aria goldens keylessly. `paceMs` validation, pacing floor, abort-during-pace, and both `assertConsumed` failure shapes are pinned in `packages/support/llm-replay/tests/llm-replay.spec.ts`.
+
+## Deferred
+
+- **Web header-class pin**: web fixtures tokenize `{{system}}`/`{{tools}}` everywhere and no scenario pins the web composition's prompt/tool schemas (`TODO(web-header-pin)` — the scaffold `recordFixture` JSDoc marks it). Following the TUI scrub-everywhere precedent; revisit when the web assembly's header diverges from the repl composition it mirrors.
+- **CI browser provisioning**: reversal of the no-browser-in-CI ruling, staged criteria above (`TODO(ci-browser)`).
+- **Follow-up-prompt-after-resume scenario**: the history/live stitch path over the real wire; add as its own scenario when that code changes or regresses.
+
+## Consequences
+
+The web surface gains its record-once/replay-forever tier: the real chromium → SSE → apiproxy → loop → tools → persistence chain runs keylessly in ~10-30s, deterministic across repeat runs, with fixtures owned and re-recordable by the lane itself. Costs accepted: every intentional conversation-UI change ends with a keyless `DSH_SNAPSHOT=refresh` (golden churn is reviewed diff, anchors keep semantic green); the aria format is Playwright-owned — the one committed snapshot format the repo does not control — so playwright version bumps must be deliberate bump-and-refresh commits (the dependency floats `^1.49.0` in `apps/web/package.json`; pin exactly if churn bites); replay's first-call-order binding constrains scenarios to one prompting session each, with the consumption assertion as the tripwire; `compact-basic` shares the session's replay cursor and stays inert only under the published 128k catalog window; and the lane guards regressions only where it runs (locally, `test:web`) until the CI reversal is separately decided.
diff --git a/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.zh.md b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.zh.md
new file mode 100644
index 0000000000..c43e526d27
--- /dev/null
+++ b/.agents/notes/implemented/testing/2026-07-24-web-gui-browser-e2e-lane.zh.md
@@ -0,0 +1,90 @@
+# Agent Note: Web GUI 的无密钥浏览器 e2e 车道
+
+Status: implemented
+
+[English](2026-07-24-web-gui-browser-e2e-lane.md) | 中文
+
+## 问题
+
+Web GUI 以一条真实组装链交付——chromium 页面 → client 插件 bundle → HTTP 单次 RPC + 两条 SSE(Server-Sent Events)流 → `toFetchHandler`/apiproxy → host 端的 agent loop(智能体循环)、工具与 JSONL 持久化——却没有任何测试无密钥且确定性地检验这条链。[GUI 测试体系](../process/2026-07-20-gui-testing-system.md)覆盖第 1 层(Node 中的协议同构)、第 2 层(对象层状态机)与第 3 层冒烟测试,但无密钥冒烟驱动的是 `FixtureApiClient`——没有 host、没有 wire、没有 agent loop——而全链路冒烟需要 `DEEPSEEK_API_KEY` 和真实模型,因此不确定、在无密钥 CI 中自行跳过。[docs/testing.md](../../../../docs/testing.md) 的快照哲学——带密钥录制一次、永久无密钥回放、格式变动时刷新——已覆盖 ACP(Agent Client Protocol)、headless `stream-json` 与 TUI 三个文本记录(transcript)表面;web 表面是唯一没有这层保障的组装形态。而缺口恰恰是两起已实证 GUI P0 藏身之处:fixture(测试前置数据)客户端短路掉的 wire 承载链。
+
+## 决策
+
+`pnpm run test:web` 携带 `apps/web/tests/` 下的无密钥、确定性浏览器 e2e 车道:录制的会话日志 fixture 经 `@deepseek-ai/dsh-llm-replay` 对真实进程内 web 组合回放,断言规范化后的会话区 aria 预期输出加进程内世界状态。不新增包(package);产品侧增量只有 `dsh-llm-replay` 的两处增量接口(`paceMs`、`ReplayHandle`)。
+
+### Scaffold:`apps/web/tests/scaffold.ts`
+
+一个普通的共享 fixture 模块([测试政策认可的形态](../../../../docs/testing.md)),不是包:值得门禁把守的逻辑——回放推导、会话解析、日志脱敏、持久化——都在已受门禁的包 `dsh-llm-replay`、`dsh-acp-snapshot`、`dsh-session-persistence-jsonl` 中;剩下的只是启动接线和浏览器胶水,而驱动 chromium 的源码在无浏览器的覆盖率 runner 上无法诚实保持逐文件 100% 覆盖率。
+
+`launchWebScaffold()` 通过 vendored Loader 的 include 机制,从交付的 `apps/cli/cordis.yml` 启动真实 web 组合——与 `AppCLIEntry` 为 `dsh web` 驱动的是同一棵树、同一套机制。差异全部经 include patch 覆盖在这棵树上,即 ACP `cordis.snapshot.yml` 模式的进程内表达:临时 `persistenceRoot`;禁用 `workspace-context`(录制的 fixture 不得嵌入本仓库的 AGENTS.md);禁用 `session-title-llm`(其发后不管的标题调用会与循环争抢会话的回放游标);webserver 行钉到端口 0 加已构建 dist;无密钥模式下禁用 `llm-deepseek`。patch 的 id 一旦不再匹配任何行,boot 扫描会大声失败而不是漂移。boot 在临时工作区 `chdir` 下运行,使 api-gateway 的 `process.cwd()` 会话默认值、工具 cwd 与 fixture 一致;`dsh web` bin 自身的胶水(argv、profile json、AppCLIEntry)仍由 `smoke-real.e2e.ts` 中的无密钥 CLI 冒烟把守。初始化回滚和正常关闭都会先对 Cordis 树执行 dispose(资源释放),再删除 scaffold 持有的两个临时根目录;每项清理都会独立尝试,并会报告清理失败而不掩盖初始化失败。
+
+无密钥的模型替换 = 禁用适配器行的 patch 加 `installLlmReplay` 在停稳的根 ctx 上以提供方目录(providers-catalog)模式填充开放的 seam——绝不用 catch-all:适配器行被禁用后不存在任何适配器,catch-all 会让 `resolveModelContext` 无路由可走,`compact-basic` 的步后压力检查将步步告警,而不是被可证明地闲置(发布的 128k `contextWindow` 使该路径对小 fixture 保持闲置)。选择直接安装而非插入回放插件行是刻意的:直接安装返回收尾消费检查所需的 `ReplayHandle`。没有 fixture 的场景让 seam 保持空置,任何离群的流式调用都会以 NO_ADAPTER 大声失败。
+
+`seedSession()` 通过真实持久化 API 播种冷会话——一次性 `Context` 挂载 `SessionStore` + `SessionPersistenceJsonl` 指向 host 的根目录,`create()` + `append()`,一次 `utimes` 回拨保证侧栏顺序确定(`semantic-checkpoint.snapshot.ts` 先例)——绝不裸写文件,因此播种器对桶哈希、文件名编码、压缩一无所知,host 的 zstd 默认值也无需任何启动开关。种子在播种时即校验(可解析、以 `turn/end` 结尾——未闭合的最终轮次会被恢复(resume)的崩溃修复改写)。
+
+### 确定性规则
+
+回放模式下浏览器断言的屏障栈,按序:(1)host 侧 `await agent.whenIdle()` 加超时,以进程内 `turn/end` 为锚——空闲翻转发生在持久化落盘之后,一次等待同时覆盖轮次完成与持久性;(2)浏览器安定轮询(流式输出节点已卸载、最终文本可见)。录制模式下,日志采收在 `whenIdle()` 之后、scaffold 释放之前进行,此时运行中的会话仍然可用。单独监听进程内 `turn/end` 是错误屏障(它先于 SSE 帧到达浏览器、先于 fsync 触发);文件轮询被禁止(NFS 上慢,且被 `whenIdle` 取代);`networkidle` 被彻底禁止(SSE 流保持打开时它永不解析)。
+
+不做单次瞬态 DOM 断言:从回放产出到 React 提交的每一跳都可能合并分片,采样 `[data-streaming]` 天然就是竞态。流式输出的增量性由持久化的 `assistant/chunk` 事件断言(模型可见 ⟺ 已记录,使日志成为权威证据)。`dsh-llm-replay` 的可选 `paceMs`(默认缺省 = 突发)只是让浏览器观察到真正增量 SSE 的真实感旋钮;正确性绝不依赖它,且节奏等待期间中止会即时取消。
+
+每个场景都会因任何 pageerror 或客户端的连接丢失/间隙修复控制台警告而失败:否则重连机制加历史重同步会把一条死掉的 SSE 通路自愈掉,套件反而认证了坏 wire。Scaffold 的 `close()` 调用 `ReplayHandle.assertConsumed()` 收尾检查(每个已录脚本都被绑定、每个游标都耗尽),把静默的少放与错绑变成清晰诊断。车道不设 vitest 重试;每文件一个 chromium、每场景一个新 context、每场景一个 host;视口固定;交互选择器锚定 role、`data-*` 属性和可见文本,而 frame 与会话区采集则使用既有的 CSS 模块局部类名锚点。
+
+### 预期输出
+
+每场景一份提交的预期输出:会话区规范化 `ariaSnapshot()`(`ui.expected.md`)——uuid/cwd/工作区目录名/时长归一为稳定 token,在安定里程碑处轮询至两次相等再采集——外加几条 role/文本锚断言,让保语义的组件重写在预期输出可评审地变动时仍保持绿色锚点。aria 树是 client 规则「断言用户所见,绝不断言类名」的机械化。世界状态断言内联在根上下文的会话事件上(哪次工具调用产生了哪项已持久化的工具结果、`turn/end` 是否完成)而不是第二份提交的日志预期输出:持久化日志表面已由 ACP/headless/TUI 套件经同一循环和持久化钉住,在此重复钉住会违背分层纪律、翻倍刷新成本。`refresh` 是预期输出的唯一写入者——回放模式下预期输出缺失会连同修复命令一起报错,而不是静默自举。
+
+类型检查平面切分是结构性的:启动 host 主干的三个文件(`scaffold`、`replay-round-trip.e2e` 和 `seeded-history.e2e`)被排除出注册在 client 侧的 `apps/web` 工程。这三个文件及其共享的 `support.ts` 逐文件纳入 `tsconfig.host.json`——一个程序不能同时持有 cordis `Context` 合并的两侧。
+
+### 模式与 fixture
+
+`DSH_SNAPSHOT` 以内联 spec 分支选择 replay(默认,无密钥)、record(带密钥)或 refresh(无密钥)——TUI 的形态,不是套件工厂:两个场景撑不起 acp-snapshot 工厂机制,且真正共享的部分已被导出(`scrubRequestHeaders`、`parseSessionLog`、`installLlmReplay`)。每个 spec 切分为驱动步骤(输入、发送、`whenTurnSettled`——所有模式都执行,绝不等待模型内容选择器,因此 record 不会因真实模型答法不同而挂起)与断言步骤(仅 replay/refresh)。Record = 经真实输入框实时驱动 + 采收内存中的 `session.header`/`session.events`(TUI 的 `rawSessionLog` 形态——无需文件解压)+ `scrubRequestHeaders` + `{{sessionId}}`/`{{cwd}}`/`{{rpcId}}` token 化;随后一次无密钥 refresh 重新生成 `ui.expected.md`。两个场景的 fixture 都经此流程对本组装录制。一条漂移防线把每个 spec 的驱动提示词与 fixture 录制的 `user/message` 绑定。fixture 清单防线保持每个场景目录封闭(精确文件集合,每个 JSONL 都是脱敏不动点,不含当次运行的 `rpcId`)。Web fixture 全部脱敏请求头且不钉任何头类别,沿用 TUI 先例而非[钉住请求头](2026-07-06-pin-request-header-content-in-one-scenario.md)的严格读法——见「暂缓」。
+
+### 场景
+
+1. **`replay-round-trip`**——新会话,经真实输入框发送提示词,回放流式输出推理(reasoning)+ 一次在临时工作区真实执行的 `bash` 工具调用 + 最终文本(15ms 节奏)。断言安定后的 markdown、aria 预期输出与内联世界状态(这次 bash 调用的已持久化工具结果严格等于 `WEB_E2E_OK\n`、完成的 `turn/end`、>10 个分片事件)。
+2. **`seeded-history`**——冷播种一份已录会话;侧栏列出它(分组行 → 会话行,默认折叠),打开后纯凭日志经 `session.history` 内的隐式冷恢复挂载渲染工具卡片与文本——replay 下零模型调用,因此没有任何绑定约束;record 模式实时驱动同一轮(真实 `read` 工具读取播种的工作区文件)来产出种子。
+
+### CI 立场
+
+车道随 `pnpm run test:web` 交付、豁免门禁,与该配置头部注释所记一致。往 CI 加 chromium 会推翻 [GUI 测试笔记](../process/2026-07-20-gui-testing-system.md)中「CI 无浏览器基础设施」的前提,因此需要自己的 Agent Note 并从那里交叉链接,分阶段推进:先作为非必需任务,再以量化标准晋升(连续绿色运行次数、耗时、零重试的抖动预算、runner 浏览器缓存策略)。`TODO(ci-browser)` 标记该接缝。场景目前面向 POSIX(车道不在 Windows 矩阵中)。
+
+## 业界先例
+
+调研了 AI 聊天/agent web UI 与 mock 层(LibreChat、vercel/ai-chatbot + AI SDK、lobe-chat、open-webui、OpenHands、Chainlit、continue、cline、langfuse、gradio/streamlit;Playwright HAR/route、MSW、Polly/nock、WireMock、aimock)。自有后端的应用的主流成熟架构是:真实后端 seam 后放一个进程内伪造/回放模型,下游全部真实(LibreChat 的 `LIBRECHAT_TEST_RUN_HOOK` 伪模型;ai-chatbot 的 `MockLanguageModelV3` + `simulateReadableStream`;continue 的脚本化 mock 提供方类)——这正是 `dsh-llm-replay` 已然所是。浏览器层 SSE 拦截无法检验增量渲染(`route.fulfill` 一次性交付整个响应体;playwright#33564),且服务端 SSE 栈完全失测,因此各项目只把它用于边缘用例。分片节奏作为 fixture 参数反复出现(LibreChat 默认 10ms 附慢速档;ai-chatbot 500ms);CI 里的真实模型会腐烂(open-webui 的套件长出 120 秒超时,先被禁用后被删除);会话在持久化层以受控时间戳播种(LibreChat 直插回拨时间的 Mongo 文档;langfuse 播种其数据库)。没有任何被调研项目为 UI 测试把录制的 agent 事件日志经真实后端回放——最接近的是提供方层录制 fixture(aimock)与前端层 socket 历史发射(OpenHands MSW)——因此会话日志即 fixture 的设计沿着本仓库「模型可见 ⟺ 已记录」不变式所指的方向比业界先例多走了一步。
+
+## 曾考虑的替代方案
+
+**浏览器网络层 SSE 拦截(`page.route`)。** 已否决:`route.fulfill` 无法流式输出,增量 token 渲染无从检验,且服务端 SSE/背压/关闭路径——两起已实证 P0 的藏身处——完全失测。
+
+**`DEEPSEEK_BASE_URL` 处的 mock HTTP 提供方。** 作为本车道机制已否决(仅保留给既有的工作区探针冒烟):fixture 会变成手写的 OpenAI SSE 字节脚本,一种与仓库其余部分录制回放的会话日志格式渐行渐远的第二 fixture 格式;适配器的真实 HTTP 路径归带密钥 e2e 管。
+
+**扩展 `?fixture` 客户端。** 已否决:分层纪律——`FixtureApiClient` 的存在意义就是脱离服务器测试客户端 shell;client API seam 以下按构造即失测。
+
+**用占位 `DEEPSEEK_API_KEY` + 回放拦截替代禁用适配器行。** 尽管零组合改动且树内有两处先例仍被否决:它用谎言满足 `llm-deepseek` 的快速失败密钥检查,还留下一个挂载却被拦截的死适配器;禁用行(ACP overlay 的同款做法)是诚实的无密钥,并在最早可解析点快速失败。
+
+**`packages/support/web-snapshot` 包 + `defineWebSnapshotSuite` 工厂。** 已否决:驱动 chromium 的源码在无浏览器的覆盖率 runner 上无法诚实保持逐文件 100%,且两个场景就上工厂是从单一消费方过度泛化,真正共享的逻辑已从受门禁的包中导出。重启条件:出现第二个 web 形态消费方,或 ≥6 个场景的内联分支被证实各自漂移;届时包边界将画在无浏览器一侧。
+
+**第二份提交的规范化会话日志预期输出。** 已否决:日志表面已由 ACP/headless/TUI 套件经同一循环与持久化钉住;在此只会翻倍刷新成本并重复测试下层。内联在根上下文事件上的世界状态断言保住了验证世界的义务。
+
+**以 `DSH_SNAPSHOT` 回放分支拉起 `dsh web` bin。** 已否决:它需要在交付的 CLI 中增加测试专用回放分支和环境变量管道。进程内 scaffold 已加载同一份 `apps/cli/cordis.yml`;只剩 argv、profile JSON 和 `AppCLIEntry` 胶水不在其覆盖范围内,而这些路径已由无密钥 CLI 冒烟覆盖。
+
+**为可测试性改 wire 协议。** 已否决:契约已有第一等的无密钥同构 seam(`InProcessApiClient(toFetchHandler(api))`),逐事件不合批的 SSE 恰是回放在浏览器中可观测的原因,测试一条不再交付的 wire 会颠倒该层的存在意义。
+
+**以真实模型浏览器测试充当无密钥车道。** 已否决:按构造即不确定;被调研的前车之鉴(open-webui)长出无界超时后被删除。带密钥的 W5 冒烟仍是真实模型侧的补充。
+
+**客户端 `data-dsh-busy` 安定信号。** 暂缓:两个场景下多条件安定轮询已经够用,host 侧 `whenIdle` 屏障承担了重活。重启条件:第一次安定轮询抖动,或某场景需要等待 DOM 不暴露的状态。
+
+## Testing
+
+车道自身:`pnpm run test:web` 与既有冒烟对一起无密钥运行两个场景;`DSH_SNAPSHOT=record pnpm exec vitest run --config vitest.web.config.ts apps/web/tests/` 对真实模型重录某场景的 fixture;`DSH_SNAPSHOT=refresh` 无密钥重写两份 aria 预期输出。`paceMs` 校验、节奏下限、节奏中中止、`assertConsumed` 的两种失败形态钉在 `packages/support/llm-replay/tests/llm-replay.spec.ts`。
+
+## 暂缓
+
+- **Web 头类别钉住**:web fixture 处处 token 化 `{{system}}`/`{{tools}}`,没有场景钉住 web 组合的提示词/工具 schema(`TODO(web-header-pin)`——scaffold 的 `recordFixture` JSDoc 有标记)。沿用 TUI 处处脱敏先例;当 web 组装的请求头与其镜像的 repl 组合进一步分叉时重审。
+- **CI 浏览器供给**:推翻 CI 无浏览器裁定,分阶段标准见上(`TODO(ci-browser)`)。
+- **恢复后追问场景**:真实 wire 上的历史/实时缝合路径;当该代码变更或回归时作为独立场景补充。
+
+## 后果
+
+Web 表面获得了录制一次/永久回放的层级:真实 chromium → SSE → apiproxy → 循环 → 工具 → 持久化的链路以约 10-30 秒无密钥运行,重复运行结果确定,fixture 由车道自身持有并可重录。接受的成本:每次有意的会话 UI 变更都以一次无密钥 `DSH_SNAPSHOT=refresh` 收尾(预期输出变动是受评审的 diff,锚断言保住语义绿色);aria 格式归 Playwright 所有——仓库唯一不受自己控制的提交快照格式——因此 playwright 版本升级必须是刻意的升级加刷新提交(依赖在 `apps/web/package.json` 中浮动为 `^1.49.0`;若变动伤人则改为精确锁定);回放的首次调用顺序绑定把每个场景限制为至多一个发起提示的会话,消费断言是绊线;`compact-basic` 与会话共享回放游标,仅在发布的 128k 目录窗口下保持闲置;在 CI 反转被单独决策之前,车道只在其运行之处(本地,`test:web`)把守回归。
diff --git a/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.i18n.yaml b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.i18n.yaml
new file mode 100644
index 0000000000..c9cebc9db9
--- /dev/null
+++ b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.i18n.yaml
@@ -0,0 +1,6 @@
+# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
+# side as of the last confirmed-consistent state. Both languages carry equal authority;
+# after editing either side, bring the other along and re-record with:
+# pnpm run verify-translation-pairing --write
+2026-07-25-scriptable-llm-wire-fault-server.md: 0795f71f0eab1a107740aaa8cba6fa04b1fbd306
+2026-07-25-scriptable-llm-wire-fault-server.zh.md: a0e3c98d729adcc74e6d98933ab2beb538f71cb3
diff --git a/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.md b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.md
new file mode 100644
index 0000000000..0795f71f0e
--- /dev/null
+++ b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.md
@@ -0,0 +1,41 @@
+# Agent Note: Scriptable LLM wire fault server
+
+Status: implemented
+
+English | [中文](2026-07-25-scriptable-llm-wire-fault-server.zh.md)
+
+## Problem
+
+Adapter unit tests use local HTTP servers to classify individual provider failures, while retry tests use an in-process scripted `LlmAdapter` to prove closed-step recovery. Neither boundary provides a reusable server for running the shipping HTTP adapter, agent loop, and retry policy together, and neither lets a developer point an existing app at deterministic transport faults by changing only its base URL and API key.
+
+Connection refusal, a reset before the first event, clean EOF without `[DONE]`, a valid content-less completion, and a reset after partial output have different adapter and recovery outcomes. Treating them as one generic mock failure hides whether the provider boundary preserved the distinction and whether failed chunks remained outside committed model history.
+
+## Decision
+
+`@deepseek-ai/dsh-llm-mock-server` is a support package with an importable Node HTTP server and a standalone CLI. It accepts OpenAI-compatible root and `/v1` chat-completions paths, validates an optional bearer token, captures requests, and consumes one explicit behavior per accepted request. Script exhaustion fails loud; repetition requires `repeatLast`.
+
+Request behaviors cover socket reset, post-header disconnect, partial disconnect, stall, valid empty completion, clean truncated streams, malformed payloads, representative HTTP failures, complete text/reasoning/tool-call responses, slow streaming, and max-token completion. A true `connection_refused` is a CLI listener-lifecycle phase because a bound request handler cannot refuse its own TCP connection.
+
+The `random` script entry performs a new weighted selection for every request. The server exposes and logs its unsigned 32-bit seed, accepts caller-supplied relative weights, and ships a success-heavy stress profile that mixes transport, protocol, provider, timeout, and semantic-empty outcomes. The profile is configurable test pressure rather than an estimate of production incident frequency; `connection_refused` remains outside the request-level pool.
+
+The server reports wire facts only and does not classify retryability. Real-composition tests route it through `dsh-llm-deepseek`, `dsh-agent-loop`, and `dsh-llm-retry`: connection refusal, hard disconnect, partial reset, idle timeout, and a valid content-less completion recover under the existing default policy; clean partial EOF remains `STREAM_CLOSED` and is not retried by default. The package does not change those policies.
+
+## Verification
+
+Package tests exercise every request behavior, split UTF-8 request decoding, HTTP validation without script consumption, script exhaustion/repetition, stalled-connection teardown, CLI parsing and delay bounds, IPv6 base URLs, random seed reproducibility, weight validation, single-result telemetry, lifecycle cleanup, and the invariant companion under the per-file coverage gate. The retry integration suite proves exact request counts, numbered retry steps, request-body identity, failed partial-chunk isolation, semantic-empty recovery, clean-EOF classification, timeout recovery, true refused-connection recovery after delayed listener startup, and bounded exhaustion through the real HTTP/SSE adapter.
+
+## Alternatives considered
+
+**Implement the server in Python** — rejected because Node's standard HTTP and socket APIs expose every required fault, while TypeScript keeps the server, CLI parser, tests, package build, lint, and coverage inside the repository's existing toolchain. A second runtime would add environment and subprocess dependencies without increasing wire isolation.
+
+**Keep separate inline mock servers in adapter tests** — rejected because those fixtures cannot be launched by an existing app and would duplicate behavior sequencing, randomization, telemetry, and connection cleanup across suites. A support package gives tests a shared implementation without promoting it to product API.
+
+**Use only an in-process `LlmAdapter` mock** — rejected because it bypasses fetch, HTTP status/header parsing, SSE framing, socket termination, and the adapter idle watchdog: the exact boundaries this test infrastructure exists to exercise.
+
+**Change retry defaults with the server** — rejected because the server reveals existing semantics rather than deciding policy. Extending recovery to `STREAM_CLOSED` requires a separate decision with its own cost, latency, and duplicate-generation trade-offs.
+
+## Consequences
+
+Developers can reproduce fault sequences by changing only provider URL/key configuration, and automated tests can keep socket-level failures deterministic through explicit scripts and seeds. The same wire fixture now exposes gaps between hard resets, clean truncation, and recovered empty completions without splicing attempts or modifying model history.
+
+The server adds a support package, executable build entry, and behavior vocabulary that must remain compatible with both direct tests and CLI examples. Arrival-ordered scripts are intentionally shared across clients, random defaults are stress weights rather than operational truth, and exact connection refusal requires coordinating the client attempt with the pre-listen interval.
diff --git a/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.zh.md b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.zh.md
new file mode 100644
index 0000000000..a0e3c98d72
--- /dev/null
+++ b/.agents/notes/implemented/testing/2026-07-25-scriptable-llm-wire-fault-server.zh.md
@@ -0,0 +1,41 @@
+# Agent Note: 可脚本控制的 LLM(大语言模型)协议层故障服务器
+
+Status: implemented
+
+[English](2026-07-25-scriptable-llm-wire-fault-server.md) | 中文
+
+## 问题
+
+适配器单元测试使用本地 HTTP 服务器对各类提供方故障逐一分类,重试测试则使用进程内的脚本化 `LlmAdapter` 证明已关闭步骤的恢复能力。这两个边界都无法提供可复用的服务器,以便同时运行交付版本的 HTTP 适配器、agent loop(智能体循环)和重试策略;开发者也无法仅修改现有应用的 base URL 与 API key,就让应用连接到确定性的传输故障。
+
+连接遭拒、首个事件前连接被重置、未收到 `[DONE]` 即正常 EOF、合法但无内容的完成,以及输出部分内容后连接被重置,会产生不同的适配器与恢复结果。把它们统一视为普通 mock 故障,会掩盖提供方边界是否保留了这些区别,以及失败请求的分片是否确实没有进入已提交的模型历史。
+
+## 决策
+
+`@deepseek-ai/dsh-llm-mock-server` 是一个支持包(package),提供可导入的 Node HTTP 服务器和独立 CLI(命令行界面)。它接受兼容 OpenAI 的根路径和 `/v1` chat-completions 路径,校验可选的 bearer token,捕获请求,并对每个已接受请求消耗一个显式行为。脚本耗尽时快速失败;只有设置 `repeatLast` 才会重复最后一个行为。
+
+请求行为覆盖 socket 重置、发送 header 后断开、发送部分内容后断开、停滞、合法空完成、正常关闭但被截断的流、畸形 payload、典型 HTTP 故障、完整的文本/推理/工具调用响应、慢速流式输出以及达到 token 上限的完成。真正的 `connection_refused` 由 CLI 的监听器生命周期阶段实现,因为已经绑定端口的请求处理器无法拒绝自身的 TCP 连接。
+
+脚本项 `random` 会为每个请求重新执行一次加权选择。服务器公开并记录其无符号 32 位 seed,允许调用方提供相对权重,并内置一套偏重成功结果的压力测试配置,将传输、协议、提供方、超时和语义空结果混合在一起。该配置用于提供可调的测试压力,并非对生产事故发生频率的估算;`connection_refused` 仍不进入请求级随机池。
+
+服务器只报告协议层事实,不判断是否可重试。真实组合测试让请求依次经过 `dsh-llm-deepseek`、`dsh-agent-loop` 和 `dsh-llm-retry`:在现有默认策略下,连接遭拒、硬断开、部分输出后重置、空闲超时以及合法的无内容完成均可恢复;正常关闭的部分输出 EOF 仍归类为 `STREAM_CLOSED`,默认不重试。该包不会改变这些策略。
+
+## 验证
+
+包测试覆盖所有请求行为、跨分片 UTF-8 请求解码、不消耗脚本的 HTTP 校验、脚本耗尽与重复、停滞连接清理、CLI 解析及延迟边界、IPv6 base URL、随机 seed 可复现性、权重校验、单结果遥测、生命周期清理,以及逐文件覆盖率门禁下的配套不变式插件。重试集成套件通过真实 HTTP/SSE(Server-Sent Events)适配器,验证准确的请求次数、带编号的重试步骤、请求体完全一致、失败的部分分片不会泄漏、语义空结果恢复、正常 EOF 分类、超时恢复、监听器延迟启动后从真实连接遭拒中恢复,以及有界重试耗尽。
+
+## 曾考虑的替代方案
+
+**使用 Python 实现服务器**:不予采纳。Node 的标准 HTTP 与 socket API 足以暴露所有所需故障,而 TypeScript 可以让服务器、CLI 解析器、测试、包构建、lint 和覆盖率全部留在仓库现有工具链中。引入第二套运行时会增加环境与子进程依赖,却不能增强协议隔离。
+
+**在适配器测试中继续使用各自独立的内联 mock 服务器**:不予采纳。这些 fixture(测试前置数据)无法作为独立服务器启动并供现有应用连接,还会让不同测试套件重复实现行为编排、随机化、遥测和连接清理。支持包让测试共享同一套实现,又不会将其提升为产品 API。
+
+**仅使用进程内的 `LlmAdapter` mock**:不予采纳。它会绕过 fetch、HTTP 状态与 header 解析、SSE 分帧、socket 终止以及适配器的空闲看门狗,而这正是这套测试基础设施要覆盖的边界。
+
+**随服务器一起修改默认重试策略**:不予采纳。服务器用于揭示既有语义,而非决定策略。是否将恢复能力扩展到 `STREAM_CLOSED`,需要单独决策,并权衡成本、延迟和重复生成风险。
+
+## 后果
+
+开发者只需修改提供方 URL/key 配置即可复现故障序列;自动化测试则可通过显式脚本和 seed,让 socket 层故障保持确定性。同一套协议 fixture 现在可以暴露硬重置、正常截断与恢复后的空完成之间的差异,而不会拼接多次尝试的内容或修改模型历史。
+
+服务器新增了一个支持包、可执行构建入口和行为词汇,三者必须同时兼容直接测试与 CLI 示例。按请求到达顺序执行的脚本有意由所有客户端共享;随机模式的默认值代表压力测试权重,而非实际运行规律;精确模拟连接遭拒时,需要让客户端尝试与监听开始前的时间区间协调一致。
diff --git a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.i18n.yaml b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.i18n.yaml
similarity index 61%
rename from .agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.i18n.yaml
rename to .agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.i18n.yaml
index 0852385897..5ab7a8229d 100644
--- a/.agents/notes/implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.i18n.yaml
+++ b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.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
-2026-06-18-acp-terminal-and-tool-rendering.md: 166e52f8e659a78b1347279c61956ca1001e4939
-2026-06-18-acp-terminal-and-tool-rendering.zh.md: 1623c8f51f9a1e0aacf374d182ab1c74dfa407cc
+2026-07-24-domain-kv-storage-and-workspace.md: cd666a47a3cba4dea8846cd0f1373224e6fc456f
+2026-07-24-domain-kv-storage-and-workspace.zh.md: 81adf1eb6bc32aa3ca8b9ef4c352fb94f95ace91
diff --git a/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.md b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.md
new file mode 100644
index 0000000000..cd666a47a3
--- /dev/null
+++ b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.md
@@ -0,0 +1,329 @@
+# Agent Note: Domain KV storage capability seam and the workspace entity
+
+Status: proposed
+
+English | [中文](2026-07-24-domain-kv-storage-and-workspace.zh.md)
+
+## Problem
+
+The host's only persistence surface is the session event log (`packages/session-persistence`: append-only, one file per session). Anything that does not belong to a single session has nowhere to live, and two real needs exist today:
+
+- **The workspace entity.** The GUI needs workspace as a real object: path, title, and the list of owned sessions. Ownership belongs to the workspace — "which sessions belong to this workspace" is not any single session's fact, so writing it into the session log is semantically wrong. Until now workspace was only a sidebar visual grouping derived from cwd, with no entity (that conclusion has been overturned).
+- **Dynamic session metadata** (the foreseeable second consumer). Cold session listings read only the first log line (an immutable creation-time snapshot); title, terminal status, and anything that evolves with the session is unavailable. The fix direction is a sidecar metadata table — exactly a KV table with high-frequency per-key updates.
+
+Separately, workspace deletion will eventually need to delete its owned sessions, and `SessionPersistence` has no delete primitive nor does the host expose a `session.delete` endpoint — that gap's design is settled in this note, but its implementation is marked future work: this phase touches no session-side code.
+
+## Proposal
+
+Create the `packages/storage/` group — the `ctx.storage` hub (backend registry + data-form mounts), two backends, the domain data form — plus the workspace consumer package; extend `SessionPersistence` with a delete primitive.
+
+| Package | Path | ctx surface | This phase |
+| --- | --- | --- | --- |
+| `@deepseek-ai/dsh-storage` | `packages/storage/storage/` | `ctx.storage` (the hub) | ✓ |
+| `@deepseek-ai/dsh-storage-json` | `packages/storage/storage-json/` | registers backend `json` | ✓ |
+| `@deepseek-ai/dsh-storage-sqlite` | `packages/storage/storage-sqlite/` | registers backend `sqlite` | ✓ |
+| `@deepseek-ai/dsh-storage-domain` | `packages/storage/storage-domain/` | mounts `ctx.storage.domain` | ✓ |
+| `@deepseek-ai/dsh-workspace` | `packages/workspace/workspace/` | `ctx.workspace` | ✓ |
+| `SessionPersistence.delete` extension + cascade orchestration | `packages/session-persistence/*` | new method on the existing seam | ✗ future work (session side untouched this phase) |
+| `workspace.*` / `session.delete` RPC, GUI wiring, boot assembly | — | — | ✗ next phase |
+
+(workspace lives in its own group rather than `packages/host/`: the host group's naming rule requires the `dsh-host-*` prefix while this package is named `dsh-workspace`; and the workspace entity is a domain concept, not bound to the host assembly tier. Unrelated to the existing `workspace-context` package — that is an AGENTS.md instruction loader.)
+
+Dependency direction: `dsh-workspace` → `dsh-domain` → `dsh-storage` ← the two backends. `dsh-workspace` additionally depends on the read-only face of `ctx.sessionPersistence` (attach's cwd check reads the session header; when the service is absent, attach rejects outright — no verification, no bookkeeping). The `ctx.sessions` running-check for session deletion moves into future work together with the cascade.
+
+### `dsh-storage`: the storage hub
+
+A pure registration hub, no IO of its own, no Config. The `Storage` service mounts at `ctx.storage` with two faces: `backend` (a `BackendRegistry`: `register(name, backend)` returns the disposer, duplicate names throw; `get(name)` throws `backend-not-found` for unknown names) and data-form mounting (`mount(form, facility)` over the merge-extensible `StorageForms` map, into which `dsh-domain` merges the `domain` key; unmounted access throws `form-not-mounted`). The signature text lives in `packages/storage/storage/src/index.ts` and `src/registry.ts`.
+
+**Multiple backends stay mounted side by side**; which backend serves a domain is `dsh-domain`'s configuration (below), never a global either-or. Disposer semantics = remove the name from the table; closing the backend itself belongs to the backend package's effect closure, unregister first then close.
+
+A backend is one **medium owner** (a file-tree root / one db file) exposing primitives through **data-shape facets** — only `kv` this phase; the session migration adds `log` (see the migration section). A facet is an optional member: absence means the backend cannot serve that shape, and resolution fails loud. The `kv` facet's primitive surface: `open(descriptor)` (descriptor = name/version/table list/global flag, with names and table names restricted to `^[a-z][a-z0-9_]*$` doubling as file-name and SQL-identifier segments) returns a unit exposing `loadAll` / `putRecord` / `deleteRecord` (missing key is a no-op) / `setGlobal` / `close` (idempotent); values are opaque JSON to the backend. The normative text (with per-method JSDoc) is `packages/storage/storage/src/backend.ts`.
+
+The backend contract (asserted clause by clause by the shared conformance suite, one suite for both backends):
+
+1. `open` creates when the medium holds nothing (lazy materialization allowed: may defer to the first write, but `loadAll` must immediately serve empty tables); loads when the medium exists.
+2. A stored version ≠ descriptor.version → `StorageError('version-mismatch')`; no migration, no rebuild.
+3. Durability: after a write primitive resolves, a process crash followed by a re-open must observe the write in `loadAll`.
+4. The backend does not promise write ordering within a unit — **the caller serializes**; the backend only guarantees each single call is atomic (JSON whole-file replace / SQLite single statement).
+5. `deleteRecord` is idempotent; `putRecord` overwrites.
+6. Any string key / any JSON value is safe (keys never reach file paths, a structural property).
+7. `close` is idempotent; any operation after close → `StorageError('closed')`.
+
+The error vocabulary is `StorageError` with a code discriminant: `backend-not-found` / `form-not-mounted` / `duplicate-backend` / `duplicate-mount` / `version-mismatch` / `malformed-medium` / `closed` (`packages/storage/storage/src/error.ts`).
+
+### `dsh-storage-json`
+
+Config is `root` only (required, no default, schemastery); apply registers backend `json` inside `ctx.effect()`, and the disposer unregisters the name before `backend.close()`.
+
+- Layout `/.json`, one file per unit; directory 0o700, files 0o600.
+- File format (version stamp in the header; the file is always the current net state, `JSON.stringify(…, null, 2)` human-readable — that legibility is this backend's reason to exist):
+
+```json
+{
+ "unit": { "name": "workspace", "version": 1 },
+ "global": null,
+ "tables": { "workspaces": { "": {} } }
+}
+```
+
+- Writes: every write primitive = full serialization of the in-memory state → temp write + fsync → atomic rename publish (the Windows variant follows session-persistence-jsonl's win32 path). Memory is authoritative, disk is its projection.
+- `loadAll`: parse the whole file at open; a missing `unit` header, non-object tables, etc. → `malformed-medium`. A missing file = an empty unit, materialized on first write.
+
+### `dsh-storage-sqlite`
+
+Config is `path` (required, `':memory:'` allowed) plus `journalMode` (enum, default `wal`); apply mirrors json, registering backend `sqlite`.
+
+- `node:sqlite` `DatabaseSync`; the open sequence follows session-persistence-sqlite: mkdir 0o700 → `open(path,'wx',0o600)` exclusive create when missing → `PRAGMA foreign_keys=ON` → journal_mode → version check → create tables.
+- Physical layout version `STORAGE_SQLITE_SCHEMA_VERSION = 1` in `PRAGMA user_version`: 0 → stamp; ≠ → `version-mismatch`.
+- DDL (all STRICT; table names concatenated from the restricted character set with the `u_` prefix, no external input ever reaches DDL):
+
+```sql
+CREATE TABLE IF NOT EXISTS units (name TEXT PRIMARY KEY, version INTEGER NOT NULL) STRICT;
+CREATE TABLE IF NOT EXISTS unit_globals (
+ unit TEXT PRIMARY KEY REFERENCES units(name), value TEXT NOT NULL) STRICT;
+-- 每 unit 每表:
+CREATE TABLE IF NOT EXISTS "u__" (
+ key TEXT PRIMARY KEY, value TEXT NOT NULL) STRICT; -- value = 记录 JSON 文档
+```
+
+- Unit versions live in `units` rows; a descriptor mismatch → `version-mismatch`. Row granularity is document-per-row, preserving precise per-key durable updates (the path left open for high-frequency point-update tables like the session sidecar); when query needs appear, JSON1 reads the value column directly.
+- Write primitives are single statements and thus atomic; no cross-statement transactions needed (the domain layer has no cross-table transactions, see the out-of-scope list).
+
+### `dsh-domain`: the domain data form
+
+A single implementation, not abstracted; consumers depend on this layer only and never touch backends directly.
+
+```ts ignore-check
+export const Config = z.object({
+ backend: z.string().required(), // 默认后端名,必填
+ routes: z.dict(z.string()).default({}), // per-domain 覆盖:{ workspace: 'sqlite' }
+})
+
+export function apply(ctx: Context, config: Config) {
+ ctx.effect(() => ctx.storage.mount('domain', new DomainFacility(ctx, config)))
+}
+```
+
+(Facility unmount order: dispose each domain first (drain its write chain), then remove the name from the hub — in-flight writes still emit `domain/changed` during the drain, and the event-consistency invariant resolves domains back through the facility, so the name must stay resolvable at that point.)
+
+Domain declarations (the spec object is defined and exported by the package that owns the domain — the single source of type and runtime truth; schemas use zod with `z.infer` deriving the types without re-declaration — the record model projects into RPC wire schemas next phase and the wire boundary is all zod; schemastery still owns plugin Config only):
+
+```ts ignore-check
+export interface DomainGlobalSpec { readonly schema: ZodType; readonly initial: G }
+export interface DomainTableSpec { readonly valueSchema: ZodType }
+
+export interface DomainSpec {
+ readonly name: string // ^[a-z][a-z0-9_]*$
+ readonly version: number
+ readonly global?: DomainGlobalSpec
+ readonly tables: Record>
+}
+
+export function defineDomain(spec: S): S
+export function domainTable(schema: ZodType): DomainTableSpec
+```
+
+`DomainFacility.open(spec)` exact semantics (sequential; any failing step fails the whole open):
+
+1. A domain with this name already open → `DomainError('already-open')`.
+2. Backend name = `config.routes[spec.name] ?? config.backend`; `ctx.storage.backend.get(name)` (an unmounted name propagates `backend-not-found` — misconfiguration fails loud).
+3. Backend lacks the `kv` facet → `DomainError('facet-unsupported')`.
+4. `kv.open(descriptorOf(spec))` (the descriptor is a direct projection of the spec).
+5. `loadAll()`; every record passes `valueSchema.parse`, the global passes its schema (null takes `initial`, not persisted — first write materializes). A failure → `DomainError('invalid-record', { table, key })` (the durable boundary must validate; the write side does not re-validate).
+6. Construct the `Domain` and register `ctx.effect()`: the disposer drains the write chain → `unit.close()`.
+
+```ts ignore-check
+export interface Domain* 由 spec 推导 */> {
+ readonly name: string
+ readonly global: { get(): G; set(value: G): Promise } // 仅当 spec.global 声明
+ table(name: N): KvTable, ValueOf>
+}
+
+export interface KvTable {
+ get(key: K): V | undefined // 内存快照,同步
+ entries(): IterableIterator<[K, V]>
+ keys(): IterableIterator
+ readonly size: number
+ put(key: K, value: V): Promise
+ delete(key: K): Promise // false = 本就不存在
+ /** Atomic read-modify-write on the domain's single write chain; fn is sync-pure. */
+ update(key: K, fn: (current: V) => V): Promise // 缺 key → DomainError('missing-key')
+}
+```
+
+Rules:
+
+- **Single-level mapping**: key → record, no nested tables; hierarchical needs use composite keys or fields inside the value. The two backends stay isomorphic as a result (one JSON object level ↔ one SQLite row).
+- **Records are plain data**: immutable, directly JSON-serializable POJOs; values returned by `get`/`entries` must not be mutated in place (TypeScript readonly projection, no runtime freezing). Behavior-carrying domain objects belong to consumer packages.
+- **Serialized writes**: one promise chain per domain; `put`/`delete`/`update`/`global.set` all queue on it; `update`'s fn runs on the chain, so concurrency cannot interleave. No active-record (pulling out a mutable object that auto-persists — uncontrollable persist timing, in conflict with the whole-unit atomic-rewrite model).
+- **Version fails loud**: a stored version differing from the spec throws outright; no migration, no rebuild (the data is not regenerable; pre-release rejects old formats).
+- **Change events**: after each write's durability resolves, emit `domain/changed` (`@mode emit`), one per record, no old value (matching the repository's "new snapshot + operation discriminant" convention, template `goal/changed`); the payload `DomainChanged` is a put/deleted discriminated union — domain + table + key (both `''` for global changes) + operation, with the put branch carrying the new snapshot value and the deleted branch carrying none (`packages/storage/storage-domain/src/events.ts`). This is next phase's RPC push-frame event source. The error vocabulary is `DomainError`, codes: `already-open` / `facet-unsupported` / `invalid-record` (with `{ table, key }`) / `missing-key` / `closed`.
+
+### Future work: session-side deletion (design settled, not implemented this phase)
+
+This section is the settled construction spec; the implementation phase changes code only, not semantics. No session-persistence file is modified this phase.
+
+```ts ignore-check
+export abstract class SessionPersistence extends Service {
+ /**
+ * Permanently delete one session's stored log.
+ * Queued on the per-id write chain (serialized with in-flight appends).
+ * Unknown id → reject; un-materialized create intent → cancel it and resolve.
+ * After deletion the id behaves as unknown for every subsequent operation.
+ */
+ abstract delete(id: SessionId): Promise
+}
+```
+
+- JSONL backend: unlink the session's file (including the `.zstd` variant); neither file nor intent → reject.
+- SQLite backend: one transaction `DELETE FROM events…; DELETE FROM sessions…`; zero rows hit and no intent → reject.
+- After a successful delete, emit `'session-persistence/deleted'(id: SessionId)` (`@mode emit`; the session-persistence event surface, unrelated to `domain/changed`). Derived data (the session-query full-text index and the like) subscribes and cleans itself; the persistence layer never reaches into indexes, and the crash window is covered by derived indexes being droppable-and-rebuildable.
+
+Orchestration rules (implemented together with the cascade; the `session.delete` RPC and the workspace cascade reuse the same rules):
+
+| Check (in order) | On failure |
+| --- | --- |
+| No target (the whole subtree when recursive) is running in `ctx.sessions` | throw, delete nothing; callers cancel first then delete — the persistence layer never reaches back into the runtime |
+| Non-recursive: the target has no descendants (descendants = the `parentSessionId` transitive closure, derived from `list()` headers) | throw: by default only leaves are deletable; `recursive: true` opts into recursion |
+| Recursive order is bottom-up (leaves → root) | — a mid-way crash leaves only "half the subtree deleted, ancestors intact"; re-running the same delete converges, and no dangling parent exists at any moment |
+| Some id in the cascade is already gone from disk | skip (idempotent resumption); any other error aborts |
+
+### `dsh-workspace`
+
+The package owns the `WorkspaceId` brand and exposes `ctx.workspace`. The record key is a generated uuid — path is not the key: normalization rewrites it, and reference anchors must be stable.
+
+```ts ignore-check
+export type WorkspaceId = Branded<'WorkspaceId'>
+export function WorkspaceId(id: string): WorkspaceId
+
+const workspaceRecord = z.object({
+ path: z.string(), // realpath,见下
+ title: z.string(),
+ sessionIds: z.array(z.string().transform(SessionId)),
+ createdAt: z.string(), // ISO
+ updatedAt: z.string(),
+})
+export type WorkspaceRecord = z.infer
+
+export const workspaceDomainSpec = defineDomain({
+ name: 'workspace', version: 1,
+ tables: { workspaces: domainTable(workspaceRecord) },
+})
+
+declare module 'cordis' { interface Context { workspace: WorkspaceRegistry } }
+
+export interface Workspace {
+ readonly id: WorkspaceId
+ readonly path: string
+ readonly title: string
+ readonly sessionIds: readonly SessionId[] // 唯一真相且有序:数组序即展示序
+ setTitle(title: string): Promise
+ /** Record a session under this workspace (idempotent). Rejects when the session
+ * header's cwd (realpath) differs from this workspace's path. */
+ attachSession(sessionId: SessionId): Promise
+ detachSession(sessionId: SessionId): Promise
+ /** Live directory check, uncached. */
+ status(): Promise<'ok' | 'missing-dir'>
+}
+
+export class WorkspaceRegistry extends Service {
+ constructor(ctx: Context) // super(ctx, 'workspace')
+ // start(): this.domain = await ctx.storage.domain.open(workspaceDomainSpec)
+ // 实体缓存 Map 重建
+ create(path: string, title?: string): Promise // realpath 后撞已有 → reject
+ get(id: WorkspaceId): Workspace | undefined
+ list(): Workspace[]
+ resolveByPath(path: string): Promise // 同 realpath 口径,故 async
+ // delete:future work(与 session 级联删一起做,见下);本期不提供任何删除入口
+}
+```
+
+- **Path canon**: the stored value = `fs.realpath(input)` (trailing slashes, `..`, and symlinks all resolved); uniqueness = string equality after normalization (a symlink resolving to the same directory counts as a collision). A missing directory makes create reject outright (realpath fails — a workspace must point at an existing directory; "Create new = make the directory" is upper-layer interaction: mkdir first, then create). The session cwd in attach checks follows the same canon. Single-valued cwd + unique path ⇒ one session structurally belongs to at most one workspace; double bookkeeping is impossible on the write side.
+- **Title**: a display name, defaults to `basename(path)`, mutable, duplicates allowed. Ownership is never derived from cwd as a fallback — cwd cannot express ordering, and ownership is a workspace-side fact; sessions started headless belong to no workspace.
+- Consumers see only the `Workspace` interface; `WorkspaceEntity` stays inside the package (a single implementation does not pre-split a seam). Entities are unique per id (registry cache); the record snapshot is swapped in place after each write, and the outside sees getters only. Every write funnels through the entity's internal `mutate(fn)` → `table.update`, with `updatedAt` refreshed inside mutate. Domain objects never cross RPC; next phase the wire layer projects records into zod wire schemas.
+- **Workspace deletion is future work as a whole** (settled 2026-07-24): the registry ships no delete method this phase — the half-measure "delete the record, keep the sessions" is not exposed; deletion and the session cascade (`recursive` parameter, running checks, bottom-up order, crash-rerun convergence) land as one complete semantic together with the session delete primitive; the order then is delete sessions one by one → prune the ledger → delete the workspace record.
+
+Consistency doctrine (the ledger = the only ownership authority; the implementation and test baseline):
+
+| Situation | Behavior |
+| --- | --- |
+| A ledger id has no session on disk | filtered at `list()`/entity projection; pruned by the next mutate; no error (a normal product of deletion crash-consistency) |
+| A session's cwd matches a workspace but is not in the ledger | not owned: no merging, no adoption. The GUI may later build an "orphan sessions" area (orphans = the complement of all ledgers) |
+| One session in two ledgers | structurally blocked on the write side (attach check); detected at load → throw (externally hand-edited data, never masked) |
+| The workspace directory does not exist | record and ledger stay; `status()` = `'missing-dir'`; the storage layer never auto-deletes (the directory may only be temporarily moved) |
+
+### Reuse and the session-backend migration outlook
+
+**Long-term direction**: the pure medium operations inside session-persistence's JSONL/SQLite backends sink into `dsh-storage` backends (the session packages stay; the `SessionPersistence` seam and coordinator semantics do not move — only the file/db operation layer beneath them does). The motive for reuse: the medium layer is all filesystem operations, database calls, and cross-platform grit (Windows permission and atomic-publish variants, fsync semantics, exclusive file creation…), which should be written once; business semantics (how a session appends, when, and what) stay above — while "did this append complete correctly underneath" (durability/atomicity/platform correctness) is the lower layer's responsibility, and the responsibility boundary is the facet primitive contract. The backend interface is therefore designed as **medium owner + data-shape facets**: a session log is an append-only stream, a different shape from KV — forcing them into one set of primitives would deform both, so facets split them (`kv` this phase, `log` at migration) while sharing the medium and its lifecycle.
+
+The current reuse audit (an account already legible before the migration):
+
+| Existing session-persistence logic | Nature | Disposition |
+| --- | --- | --- |
+| JSONL: temp write + fsync + link/unlink atomic publish, 0o700/0o600 permissions, Windows variant (win32.ts) | pure medium | copied by `dsh-storage-json` this phase (whole-file atomic rewrite is the same protocol); becomes the shared implementation at migration |
+| JSONL: line-append, first-line header fast read, zstd per-frame compression | log shape | stays put; moves into the `log` facet at migration |
+| SQLite: openDatabase (mkdir/exclusive create/PRAGMA sequence/user_version check) | pure medium | copied by `dsh-storage-sqlite` this phase — the two openDatabase copies are already near line-identical and this group is the third user; copy now, extract at migration |
+| SQLite: events/sessions schema, same-transaction materialization | log shape | stays put; moves into the `log` facet at migration |
+| coordinator (per-id write chain, lazy materialization, crash repair, flush barrier) | session semantics | never sinks — event-log domain logic whose counterpart here is the domain layer's write chain; each owns its own |
+| encodeSegment (id-to-path escaping) | medium utility | unused on the domain side (keys never reach paths); sinks together with the `log` facet (one file per session) at migration |
+
+**This phase does not touch session-persistence's medium code** (only the delete primitive is added); the table above is the migration-phase work list and the design evidence that the backend interface must accommodate the log shape.
+
+### Test matrix
+
+| Suite | Coverage | Backends |
+| --- | --- | --- |
+| backend contract (shared suite, written once, run on both) | the seven contract clauses + version rejection + close idempotence | json, sqlite (`:memory:` + temp dirs) |
+| registry/mount | duplicate registration, unmounted access, disposer removal | — |
+| domain layer | the six open steps, schema rejection, update serialization (concurrent interleaving stress), `domain/changed` per record, global initial-value lazy materialization, routing and `facet-unsupported` | either (json) |
+| workspace | create/uniqueness/realpath, attach checks (including rejection when sessionPersistence is absent), the four consistency-doctrine cases | mock domain or json |
+| session delete contract (future work, joins runPersistenceContract at implementation) | unknown id, deleted-id reuse, un-materialized intent, serialization with in-flight appends, the deleted event | jsonl, sqlite |
+
+Snapshots: no model-visible or assembly surface this phase, none added; next phase's RPC wiring brings them with the `workspace.*` domain.
+
+### Out-of-scope list
+
+| Not doing | Trigger | Rework point | Groundwork |
+| --- | --- | --- | --- |
+| The full deletion suite (`SessionPersistence.delete`, the deleted event, `registry.delete` cascade, recursive delete, running checks) | future work starts (before the GUI needs delete interactions) | implement per the future-work section above: the session primitive + `registry.delete(id, { recursive? })` land as one | orchestration rules and rejection table settled in this note; no deletion entry exists this phase, so no half-semantics to stay compatible with |
+| The `log` facet and the session-backend migration | any phase after this one | sink the medium operations (the reuse audit table is the work list) | the facet structure is in place; both backends' medium code is organized in sinkable shape already |
+| Multi-process write protection | two host processes writing one medium | JSON backend file locks; SQLite WAL is natively multi-process | all writes already funnel through the domain's single point; locking touches backends only |
+| Cross-process change observation | GUI reconnect awareness | the revision pattern (copy session-persistence) | `domain/changed` already exists in-process |
+| Data migration | model changes after the first tagged release | version-driven per-domain migration | versions are on the medium from day one |
+| Large-table performance | a thousand-record domain routed to json | point `routes` at sqlite, migrate the data by hand once | routing is configuration; consumers unchanged |
+| Multi-segment keys | a real two-segment consumer appears (per-workspace per-session dimension data) | key generics become tuples, SQLite composite primary keys, JSON nested levels | single-level tables are the one-segment special case; no arbitrary-depth nesting; no string-concatenated keys |
+| The scope dimension | a "one per workspace" domain appears and composite keys cannot express it | DomainSpec gains a scope declaration + a scope segment in file names (encodeSegment) | the name character set is already restricted; file names cannot collide |
+| Cross-table atomic transactions | one business operation touching two tables of one domain atomically | `domain.transact(fn)`; JSON whole-unit rewrite is naturally atomic, SQLite wraps a transaction | — |
+| Secondary indexes / conditional queries | in-memory filtering stops scaling (tens of thousands of records) | SQLite JSON1 over the value column, a read-only query facet on the seam | the JSON backend does not follow |
+| Moving a session across workspaces | a product need appears | relax the attach check into a "detach first, then attach" orchestration | — |
+| RPC/GUI/boot | next phase | `workspace.*` + `session.delete` endpoints, wire schemas, boot mounting, sidebar on real data | this phase's model and semantics are the direct source of the wire projection |
+
+## Alternatives considered
+
+- **Reusing session-persistence's coordinator/backends**: event-log semantics (append-only, turn crash repair, lazy materialization) do not match KV overwrite semantics; only the layering idea is borrowed (a coordination layer owns write ordering, backends implement minimal primitives).
+- **A workspace-specific storage package, seam extracted later**: the second consumer (the session sidecar) is already foreseeable; generalizing later means touching the interface twice.
+- **Merging domain and storage into one layer**: backends would be forced to touch schema validation, change events, and write serialization — domain concerns; split apart, storage backends implement only opaque primitives (the smallest replaceable surface) while the single domain implementation concentrates all domain logic (zod/events/serialization written once, not doubled per backend).
+- **JSON backend as jsonl append + tombstones + compaction**: temp+fsync+rename crash safety is equivalent to append; rewriting keeps the file the net current state, human-readable, with no folding/compaction/torn-line tolerance; at domain scale a full rewrite costs the same as appending a line.
+- **JSON one file per table**: under whole-file rewrites the file granularity does not affect write cost; merging per domain means fewer files and gives the global singleton a home.
+- **SQLite storing a whole domain as one blob row**: any single-record change rewrites the whole domain, forfeiting per-key precise updates — SQLite's only edge over JSON reduced to zero.
+- **SQLite generating typed columns from the schema**: a DDL generator is over-engineering; document-per-row suffices, revisit when real query needs appear.
+- **One sqlite db file per domain**: contrary to the repository's one-database-many-tables convention.
+- **A single whole-store backend choice (the session-persistence single-slot pattern)**: the initial design; changed to coexisting backends + configured routing because the hub will carry multiple data forms whose backend preferences (human-readable vs high-frequency point updates) are bound to diverge — a single slot forces the coarse "swap everything + hand-migrate data" move. The cost is one extra name lookup, backed by fail-loud.
+- **path as the workspace key**: normalization/symlink resolution rewrites the path; reference anchors must be stable.
+- **Ownership derived from cwd (or merged with the ledger)**: two sources of truth; cwd cannot express ordering; ownership is a workspace-side fact to begin with.
+- **Change events carrying the old value**: the repository's change-event convention is "new snapshot + operation discriminant" (the sole exception, fs's before/after, is a method return value rather than an event, because the old value is unrecoverable afterwards and has a diff consumer); consumers needing diffs hold their own previous snapshot.
+- **Delete auto-cancelling a running session**: the persistence/orchestration layer reaching back into the runtime dirties the layering; cancel already exists, callers compose it.
+
+## Acceptance criteria
+
+- This phase's four test suites all green: the shared backend contract suite on both json/sqlite, registry/mount disposer semantics, the domain layer (including the six open steps and fail-loud routing), and full workspace semantics (create/attach checks/consistency doctrine).
+- `ctx.workspace` completes the create → attach → list lifecycle under a test assembly (deletion is future work).
+- Zero diff in the session-persistence packages (the acceptance line for not touching the session side this phase).
+- No new snapshots this phase (no model-visible or assembly surface); added next phase with the RPC wiring.
+
+## Risks
+
+- **The repository's first push-mode change event on a persistence surface** (session-persistence polls revisions): the shape has the `goal/changed` template, but "the storage layer emits events" is a new precedent, validated only when next phase's RPC consumes it.
+- **The JSON backend's whole-unit rewrite scale premise**: if the second consumer (the session sidecar) lands on the JSON backend at thousand-record scale before being routed to SQLite, the rewrite cost surfaces earlier than expected; the mitigation is exactly `routes` pointing at sqlite.
+- **The deletion orchestration's weak dependency on `ctx.sessions`**: a headless assembly without the runtime registry treats it as "no hot sessions", leaving a window (an external process running the session); multi-process is already out of scope, accepted.
+- **Facet generalization designed against the future `log` facet without implementing it this phase**: a "reserved shape does not fit" risk; mitigated by organizing both backends' medium code in the sinkable shape from the reuse audit, so when the `log` facet lands only the facet layer moves.
diff --git a/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.zh.md b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.zh.md
new file mode 100644
index 0000000000..81adf1eb6b
--- /dev/null
+++ b/.agents/notes/proposed/architecture/2026-07-24-domain-kv-storage-and-workspace.zh.md
@@ -0,0 +1,329 @@
+# Agent Note: Domain KV storage capability seam and the workspace entity
+
+Status: proposed
+
+[English](2026-07-24-domain-kv-storage-and-workspace.md) | 中文
+
+## Problem
+
+host 侧唯一的持久化面是 session 事件日志(`packages/session-persistence`:append-only、一 session 一文件)。凡是"不属于某个 session"的信息就没有落盘处,眼下有两个真实需求:
+
+- **workspace 实体**。GUI 要把 workspace 做成真实对象:路径、标题、关联 session 清单。归属关系由 workspace 持有——"哪些 session 属于这个 workspace"不是任何单个 session 自己的事实,塞进 session log 语义不成立。此前 workspace 只是 sidebar 上按 cwd 分组的视觉概念,没有实体(该结论已被推翻)。
+- **session 动态元信息**(可预见的第二个消费者)。冷会话列表只读日志首行 header(创建时的不可变快照),title、结束状态这类随会话推进变化的信息拿不到;补齐方向是 sidecar 元数据表——正是一张按 key 高频点更新的 KV 表。
+
+另外,workspace 删除最终需要删除其关联 session,而 `SessionPersistence` 没有删除原语,host 也没有 `session.delete` 端点——该空白的设计随本 Note 定案,但实施标记为 future work:本期不动 session 侧任何代码。
+
+## Proposal
+
+新建 `packages/storage/` 组——`ctx.storage` 存储枢纽(后端注册面 + 数据形式挂载面)、两个后端、domain 领域数据形式——及 workspace 消费者包;给 `SessionPersistence` 扩删除原语。
+
+| 包 | 路径 | ctx 面 | 本期 |
+| --- | --- | --- | --- |
+| `@deepseek-ai/dsh-storage` | `packages/storage/storage/` | `ctx.storage`(枢纽) | ✓ |
+| `@deepseek-ai/dsh-storage-json` | `packages/storage/storage-json/` | 注册 backend `json` | ✓ |
+| `@deepseek-ai/dsh-storage-sqlite` | `packages/storage/storage-sqlite/` | 注册 backend `sqlite` | ✓ |
+| `@deepseek-ai/dsh-storage-domain` | `packages/storage/storage-domain/` | 挂载 `ctx.storage.domain` | ✓ |
+| `@deepseek-ai/dsh-workspace` | `packages/workspace/workspace/` | `ctx.workspace` | ✓ |
+| `SessionPersistence.delete` 扩面 + 级联删编排 | `packages/session-persistence/*` | 既有 seam 新方法 | ✗ future work(本期不动 session 侧) |
+| `workspace.*` / `session.delete` RPC、GUI 接线、boot 组装 | — | — | ✗ 下期 |
+
+(workspace 放独立组不放 `packages/host/`:host 组命名规则要求 `dsh-host-*` 前缀,而包名定为 `dsh-workspace`;且 workspace 实体是领域概念,不绑定 host 装配层。与既有 `workspace-context` 包无关——那是 AGENTS.md 指令加载器。)
+
+依赖方向:`dsh-workspace` → `dsh-domain` → `dsh-storage` ← 两后端。`dsh-workspace` 另依赖 `ctx.sessionPersistence` 的只读面(attach 的 cwd 校验读 session header;服务缺席时 attach 直接拒绝——无法校验即不写账)。session 删除相关的 `ctx.sessions` 运行中检查随级联删一并归入 future work。
+
+### `dsh-storage`:存储枢纽
+
+纯注册枢纽,自身不做 IO,无 Config。`Storage` service 挂 `ctx.storage`,两个面:`backend`(`BackendRegistry`:`register(name, backend)` 返回 disposer、重名 throw;`get(name)` 未知名 throw `backend-not-found`)与数据形式挂载(`mount(form, facility)` 配 merge-extensible 的 `StorageForms` map,`dsh-domain` merge 进 `domain` 键;未挂载访问 throw `form-not-mounted`)。签名正文见 `packages/storage/storage/src/index.ts` 与 `src/registry.ts`。
+
+**多后端同时挂载**;域→后端的选择是 `dsh-domain` 的配置(见下),不是全局二选一。disposer 语义 = 从表中摘名;后端自身的 close 由后端包的 effect 闭包负责,顺序先摘名后 close。
+
+一个后端是一个**介质 owner**(一棵文件树 root / 一个 db 文件),通过**数据形状 facet** 暴露原语——本期只有 `kv`;session 迁移期加 `log`(见迁移节)。facet 是可选成员,缺席即该后端不支持该形状,解析时 fail loud。`kv` facet 的原语面:`open(descriptor)`(descriptor = 名字/版本/表名清单/有无 global,名字与表名限 `^[a-z][a-z0-9_]*$` 兼作文件名与 SQL 表名段)返回 unit,unit 提供 `loadAll` / `putRecord` / `deleteRecord`(缺 key 为 no-op)/ `setGlobal` / `close`(幂等);值对后端是不透明 JSON。规范正文(含逐方法 JSDoc)在 `packages/storage/storage/src/backend.ts`。
+
+backend 契约(共享契约测试逐条断言,两后端同套件):
+
+1. `open` 对不存在的介质创建(懒物化允许:可延迟到首写,但 `loadAll` 立即可用返回空表);对已存在介质载入。
+2. 介质上版本 ≠ descriptor.version → `StorageError('version-mismatch')`,不迁移不重建。
+3. 持久性:写原语 resolve 后进程崩溃再 open,`loadAll` 必须反映该写入。
+4. 后端不承诺 unit 内写并发序——**调用方负责串行**;后端只保证单次调用原子(JSON 整文件替换 / SQLite 单语句)。
+5. `deleteRecord` 幂等;`putRecord` 覆写。
+6. 任意字符串 key / 任意 JSON 值安全(key 不进文件路径,结构性质)。
+7. `close` 幂等;close 后任何操作 → `StorageError('closed')`。
+
+错误词汇是带 code 判别的 `StorageError`,码表:`backend-not-found` / `form-not-mounted` / `duplicate-backend` / `duplicate-mount` / `version-mismatch` / `malformed-medium` / `closed`(`packages/storage/storage/src/error.ts`)。
+
+### `dsh-storage-json`
+
+Config 仅 `root`(必填无默认,schemastery);apply 在 `ctx.effect()` 里注册后端 `json`,disposer 先摘名再 `backend.close()`。
+
+- 布局 `/.json`,一 unit 一文件;目录 0o700、文件 0o600。
+- 文件格式(版本戳在头,文件即当前净值,`JSON.stringify(…, null, 2)` 肉眼可读——这是该后端的存在理由):
+
+```json
+{
+ "unit": { "name": "workspace", "version": 1 },
+ "global": null,
+ "tables": { "workspaces": { "": {} } }
+}
+```
+
+- 写入:任何一次写原语 = 内存态全量序列化 → temp 写 + fsync → rename 原子发布(Windows 变体照抄 session-persistence-jsonl 的 win32 路径)。内存态是权威,盘是投影。
+- `loadAll`:open 时整文件 parse;缺 `unit` 头、tables 非对象等 → `malformed-medium`。文件不存在 = 空单元,首写才落盘。
+
+### `dsh-storage-sqlite`
+
+Config 为 `path`(必填,`':memory:'` 允许)+ `journalMode`(枚举,默认 `wal`);apply 同 json,注册后端 `sqlite`。
+
+- `node:sqlite` `DatabaseSync`;打开序列照抄 session-persistence-sqlite:mkdir 0o700 → 不存在则 `open(path,'wx',0o600)` 独占建文件 → `PRAGMA foreign_keys=ON` → journal_mode → 版本检查 → 建表。
+- 物理布局版本 `STORAGE_SQLITE_SCHEMA_VERSION = 1` 存 `PRAGMA user_version`:0 → 盖章;≠ → `version-mismatch`。
+- DDL(全 STRICT;表名由受限字符集拼接加 `u_` 前缀,杜绝外部输入进 DDL):
+
+```sql
+CREATE TABLE IF NOT EXISTS units (name TEXT PRIMARY KEY, version INTEGER NOT NULL) STRICT;
+CREATE TABLE IF NOT EXISTS unit_globals (
+ unit TEXT PRIMARY KEY REFERENCES units(name), value TEXT NOT NULL) STRICT;
+-- 每 unit 每表:
+CREATE TABLE IF NOT EXISTS "u__" (
+ key TEXT PRIMARY KEY, value TEXT NOT NULL) STRICT; -- value = 记录 JSON 文档
+```
+
+- unit 版本存 `units` 行,descriptor 不符 → `version-mismatch`。行粒度 document-per-row,保住按 key 精确落盘更新(为 session sidecar 这类高频点更新大表留路);查询需求出现时 JSON1 直查 value 列。
+- 写原语单语句即原子,无跨语句事务需求(domain 层无跨表事务,见不做清单)。
+
+### `dsh-domain`:领域数据形式
+
+单实现不抽象;消费者只依赖这层,不直接触后端。
+
+```ts ignore-check
+export const Config = z.object({
+ backend: z.string().required(), // 默认后端名,必填
+ routes: z.dict(z.string()).default({}), // per-domain 覆盖:{ workspace: 'sqlite' }
+})
+
+export function apply(ctx: Context, config: Config) {
+ ctx.effect(() => ctx.storage.mount('domain', new DomainFacility(ctx, config)))
+}
+```
+
+(facility 卸载顺序:先 dispose 各域(排空写链)再从枢纽摘名——排空期间在途写仍发 `domain/changed`,事件一致性 invariant 经 facility 反查域,要求此时域名仍可解析。)
+
+域声明(spec 对象由拥有该域的包定义导出,是类型与运行时的单一来源;schema 用 zod,`z.infer` 推导类型不重复声明——记录模型下期要投影成 RPC wire schema,wire 边界全是 zod;schemastery 仍只管插件 Config):
+
+```ts ignore-check
+export interface DomainGlobalSpec { readonly schema: ZodType; readonly initial: G }
+export interface DomainTableSpec { readonly valueSchema: ZodType }
+
+export interface DomainSpec {
+ readonly name: string // ^[a-z][a-z0-9_]*$
+ readonly version: number
+ readonly global?: DomainGlobalSpec
+ readonly tables: Record>
+}
+
+export function defineDomain(spec: S): S
+export function domainTable(schema: ZodType): DomainTableSpec
+```
+
+`DomainFacility.open(spec)` 精确语义(顺序执行,任一步失败即整体失败):
+
+1. 同名域已打开 → `DomainError('already-open')`。
+2. 后端名 = `config.routes[spec.name] ?? config.backend`;`ctx.storage.backend.get(name)`(未挂载穿透 `backend-not-found`——misconfiguration fails loud)。
+3. 后端无 `kv` facet → `DomainError('facet-unsupported')`。
+4. `kv.open(descriptorOf(spec))`(descriptor 由 spec 直接投影)。
+5. `loadAll()`;每条记录 `valueSchema.parse`,global 过 schema(null 取 `initial`,不落盘,首写才落盘)。失败 → `DomainError('invalid-record', { table, key })`(durable 边界必须校验;写侧不重复校验)。
+6. 构造 `Domain` 并注册 `ctx.effect()`:disposer 排空写链 → `unit.close()`。
+
+```ts ignore-check
+export interface Domain* 由 spec 推导 */> {
+ readonly name: string
+ readonly global: { get(): G; set(value: G): Promise } // 仅当 spec.global 声明
+ table(name: N): KvTable, ValueOf>
+}
+
+export interface KvTable {
+ get(key: K): V | undefined // 内存快照,同步
+ entries(): IterableIterator<[K, V]>
+ keys(): IterableIterator
+ readonly size: number
+ put(key: K, value: V): Promise
+ delete(key: K): Promise // false = 本就不存在
+ /** Atomic read-modify-write on the domain's single write chain; fn is sync-pure. */
+ update(key: K, fn: (current: V) => V): Promise // 缺 key → DomainError('missing-key')
+}
+```
+
+规则:
+
+- **一级 mapping**:key → 记录,不做嵌套表;层级需求用复合 key 或值内字段。两后端因此同构(JSON object 一层 ↔ SQLite 一行)。
+- **记录是纯数据**:可直接 JSON 序列化的不可变 POJO;`get`/`entries` 返回值不得原地改(TypeScript readonly 投影,不做运行时冻结)。带行为的领域对象属于消费者包。
+- **写串行**:域内一条 promise 链,`put`/`delete`/`update`/`global.set` 全排队;`update` 的 fn 在链上执行,并发不交错。不做 active-record(取出可变对象自动落盘——落盘时机不可控,与整域原子覆写冲突)。
+- **版本 fail loud**:盘上版本与 spec 不符直接报错,不迁移不重建(数据不可再生,pre-release 拒绝旧格式)。
+- **变更事件**:每次写落盘 resolve 后 emit `domain/changed`(`@mode emit`),逐条发、不带旧值(对齐仓库"新快照 + 操作判别"惯例,范本 `goal/changed`);payload `DomainChanged` 是 put/deleted 判别联合——域名 + 表名 + key(global 变更两者为 `''`)+ operation,put 支带新快照 value、deleted 支无 value(`packages/storage/storage-domain/src/events.ts`)。此为下期 RPC 推帧的事件源。错误词汇 `DomainError`,码表:`already-open` / `facet-unsupported` / `invalid-record`(带 `{ table, key }`)/ `missing-key` / `closed`。
+
+### Future work:session 侧删除(设计定案,本期不实施)
+
+本节是定案的施工规范,实施期不动语义只动代码;本期 session-persistence 的任何文件都不修改。
+
+```ts ignore-check
+export abstract class SessionPersistence extends Service {
+ /**
+ * Permanently delete one session's stored log.
+ * Queued on the per-id write chain (serialized with in-flight appends).
+ * Unknown id → reject; un-materialized create intent → cancel it and resolve.
+ * After deletion the id behaves as unknown for every subsequent operation.
+ */
+ abstract delete(id: SessionId): Promise
+}
+```
+
+- JSONL 后端:unlink 该 session 文件(含 `.zstd` 变体);文件与 intent 均无 → reject。
+- SQLite 后端:单事务 `DELETE FROM events…; DELETE FROM sessions…`;0 行命中且无 intent → reject。
+- 删除成功后 emit `'session-persistence/deleted'(id: SessionId)`(`@mode emit`;session-persistence 层事件面,与 `domain/changed` 无关)。派生数据(session-query 全文索引等)订阅自清;持久层不直连索引,崩溃窗口靠派生索引可丢弃重建兜底。
+
+编排层规则(随级联删一起实施;`session.delete` RPC 与 workspace 级联复用同一规则):
+
+| 检查(按序) | 不满足时 |
+| --- | --- |
+| 目标(递归时含整棵子树)无一在 `ctx.sessions` 运行 | throw,什么都不删;调用方先 cancel 再删,持久层不反向牵动运行时 |
+| 非递归时目标无后代(后代 = `parentSessionId` 传递闭包,由 `list()` header 求得) | throw:默认只能删叶子,`recursive: true` 显式递归 |
+| 递归序自底向上(叶→根) | ——中途崩溃只留"子树删一半、祖先在",重跑收敛,任何时刻无悬空 parent |
+| 级联中某 id 已不在盘上 | 跳过(幂等续删);其余错误中止 |
+
+### `dsh-workspace`
+
+包拥有 `WorkspaceId` brand,暴露 `ctx.workspace`。记录 key 为生成的 uuid——path 不做 key:规范化会改写它,引用锚点必须稳定。
+
+```ts ignore-check
+export type WorkspaceId = Branded<'WorkspaceId'>
+export function WorkspaceId(id: string): WorkspaceId
+
+const workspaceRecord = z.object({
+ path: z.string(), // realpath,见下
+ title: z.string(),
+ sessionIds: z.array(z.string().transform(SessionId)),
+ createdAt: z.string(), // ISO
+ updatedAt: z.string(),
+})
+export type WorkspaceRecord = z.infer
+
+export const workspaceDomainSpec = defineDomain({
+ name: 'workspace', version: 1,
+ tables: { workspaces: domainTable(workspaceRecord) },
+})
+
+declare module 'cordis' { interface Context { workspace: WorkspaceRegistry } }
+
+export interface Workspace {
+ readonly id: WorkspaceId
+ readonly path: string
+ readonly title: string
+ readonly sessionIds: readonly SessionId[] // 唯一真相且有序:数组序即展示序
+ setTitle(title: string): Promise
+ /** Record a session under this workspace (idempotent). Rejects when the session
+ * header's cwd (realpath) differs from this workspace's path. */
+ attachSession(sessionId: SessionId): Promise
+ detachSession(sessionId: SessionId): Promise
+ /** Live directory check, uncached. */
+ status(): Promise<'ok' | 'missing-dir'>
+}
+
+export class WorkspaceRegistry extends Service {
+ constructor(ctx: Context) // super(ctx, 'workspace')
+ // start(): this.domain = await ctx.storage.domain.open(workspaceDomainSpec)
+ // 实体缓存 Map 重建
+ create(path: string, title?: string): Promise // realpath 后撞已有 → reject
+ get(id: WorkspaceId): Workspace | undefined
+ list(): Workspace[]
+ resolveByPath(path: string): Promise // 同 realpath 口径,故 async
+ // delete:future work(与 session 级联删一起做,见下);本期不提供任何删除入口
+}
+```
+
+- **path 规范**:落盘值 = `fs.realpath(输入)`(尾斜杠、`..`、符号链接全解析);唯一性 = 规范化后字符串相等(符号链接指向同一目录算撞)。目录不存在时 create 直接 reject(realpath 失败——workspace 必须指向存在目录;"Create new = 建目录"是上层交互,先 mkdir 再 create)。attach 校验的 session cwd 同口径。cwd 单值 + path 唯一 ⇒ 一个 session 结构上最多归属一个 workspace,双重记账写侧不可能。
+- **title**:显示名,默认 `basename(path)`,可改,允许重复。归属不用 cwd 派生兜底——cwd 表达不了排序,归属是 workspace 侧事实;headless 直开的 session 不属于任何 workspace。
+- 消费者只见 `Workspace` 接口,`WorkspaceEntity` 不出包(单实现不预拆 seam);实体按 id 唯一(registry 缓存),记录快照写后原地换新,外部只见 getter;所有写收敛到实体内 `mutate(fn)` → `table.update`,`updatedAt` 在 mutate 内统一刷。领域对象不过 RPC,下期 wire 层把记录投影成 zod wire schema。
+- **workspace 删除整体为 future work**(2026-07-24 拍板):本期 registry 不提供 delete 方法——半截的"只删记录留 session"语义不对外暴露,删除与 session 级联(`recursive` 参数、运行中检查、自底向上、崩溃重跑收敛)作为一个完整语义随 session 删除原语一起落地;届时顺序为逐个删 session → 摘账 → 删记录。
+
+一致性口径(账 = 归属唯一依据;实现与测试基准):
+
+| 情形 | 行为 |
+| --- | --- |
+| 账中 id 盘上无 session | `list()`/实体投影时过滤;下次任何 mutate 顺手摘除;不报错(删除崩溃一致性的正常产物) |
+| session cwd 匹配某 workspace 但未上账 | 不属于:不合并不收编。GUI 将来可做"游离 session"专区(游离 = 全部账的补集) |
+| 同一 session 上两本账 | 写侧结构性堵死(attach 校验);load 检出 → throw(外部手改数据,不掩盖) |
+| workspace 目录不存在 | 记录与账保留,`status()` = `'missing-dir'`;存储层不自动删(目录可能只是暂时挪走) |
+
+### 复用与 session 后端迁移展望
+
+**长期方向**:session-persistence 的 JSONL/SQLite 后端里"纯介质操作"下沉到 `dsh-storage` 后端(session 包不删,`SessionPersistence` seam 与 coordinator 语义不动;动的只是它们脚下的文件/db 操作层)。复用的动机:介质层全是文件系统操作、数据库调用与跨平台兼容的脏活(Windows 权限与原子发布变体、fsync 语义、独占建文件……),这些只应写一遍;业务语义(session 怎么 append、何时 append、append 什么)留在上层——而"底下这次 append 是否正常完成"(持久性/原子性/平台正确性)是底层的责任,责任界面就是 facet 原语的契约。为此后端接口按**介质 owner + 数据形状 facet** 设计:session 日志是 append-only 流,与 KV 形状不同——强行统一进 KV 原语会两头变形,所以按 facet 分开(`kv` 本期、`log` 迁移期),介质与生命周期共享。
+
+现状复用审计(迁移前就能看清的账):
+
+| session-persistence 现有逻辑 | 归属 | 处置 |
+| --- | --- | --- |
+| JSONL:temp 写 + fsync + link/unlink 原子发布、0o700/0o600 权限、Windows 变体(win32.ts) | 纯介质 | 本期 `dsh-storage-json` 直接抄用(整文件原子覆写正是同一套);迁移期成为共享实现 |
+| JSONL:逐行 append、首行 header 快读、zstd 逐帧压缩 | log 形状 | 留在原地;迁移期进 `log` facet |
+| SQLite:openDatabase(mkdir/独占建文件/PRAGMA 序列/user_version 检查) | 纯介质 | 本期 `dsh-storage-sqlite` 抄用——两处 openDatabase 已几乎逐行同构,本组是第三个使用者;先抄后提,提取放迁移期 |
+| SQLite:events/sessions 表结构、同事务物化 | log 形状 | 留在原地;迁移期进 `log` facet |
+| coordinator(per-id 写链、懒物化、崩溃修复、flush 屏障) | session 语义 | 永不下沉——事件日志的领域逻辑,对应物在 domain 层(写串行链),各归各 |
+| encodeSegment(id 进路径转义) | 介质工具 | domain 侧 key 不进路径用不到;`log` facet(一 session 一文件)迁移时随之下沉 |
+
+**本期不改 session-persistence 的介质代码**(只加 delete 原语);上表是迁移期的施工清单,也是后端接口"必须装得下 log 形状"的设计依据。
+
+### 测试矩阵
+
+| 套件 | 覆盖 | 后端 |
+| --- | --- | --- |
+| backend 契约(共享套件,一次编写两端跑) | 七条契约 + 版本拒绝 + close 幂等 | json、sqlite(`:memory:` + 临时目录) |
+| registry/mount | 重复注册、未挂载访问、disposer 摘除 | — |
+| domain 层 | open 六步语义、schema 拒绝、update 串行(并发交错压测)、`domain/changed` 逐条、global 初值懒物化、路由与 `facet-unsupported` | 任一(json) |
+| workspace | create/唯一性/realpath、attach 校验(含 sessionPersistence 缺席拒绝)、一致性口径四情形 | mock domain 或 json |
+| session delete 契约(future work,随实施并入 runPersistenceContract) | 未知 id、已删 id 复用、未物化 intent、与在途 append 串行、deleted 事件 | jsonl、sqlite |
+
+快照:本期无模型可见面与组装面,不新增;下期 RPC 接线时随 `workspace.*` 域补。
+
+### 不做清单
+
+| 不做 | 触发条件 | 返工点 | 预埋 |
+| --- | --- | --- | --- |
+| 删除全套(`SessionPersistence.delete`、deleted 事件、`registry.delete` 级联、递归删、运行中检查) | future work 启动(GUI 需要删除交互前) | 按上文 future work 节实施:session 原语 + `registry.delete(id, { recursive? })` 一体落地 | 编排规则/拒绝清单已定案在本 Note;本期无任何删除入口,无半截语义要兼容 |
+| `log` facet 与 session 后端迁移 | 本期后任意期启动 | 介质操作下沉(复用审计表即施工清单) | facet 结构已留位;两后端介质代码本期即按可下沉形状组织 |
+| 多进程并发写保护 | 两 host 进程同写一介质 | JSON 后端文件锁;SQLite WAL 天然多进程 | 写全经 domain 单点串行,加锁只动后端 |
+| 跨进程变更观测 | GUI 断线重连感知 | revision 模式(抄 session-persistence) | 进程内已有 `domain/changed` |
+| 数据迁移 | 首个 tagged release 后模型再变 | 版本号驱动逐域迁移 | 版本号自第一天入介质 |
+| 大表性能 | 千级记录域挂 json | `routes` 改指 sqlite,数据手工导一次 | 路由即配置,消费者零改动 |
+| 多段 key | 两段 key 消费者出现(每 workspace 每 session 维度数据) | key 泛型换 tuple、SQLite 复合主键、JSON 嵌套层 | 一级表 = 段数 1 特例;不做任意深度嵌套;不拼字符串 key |
+| scope 维度 | "每 workspace 一份"的域出现且复合 key 表达不动 | DomainSpec 加 scope + 文件名 scope 段(encodeSegment) | 名字字符集已收紧,文件名不冲突 |
+| 跨表原子事务 | 同域两表一次原子操作需求 | `domain.transact(fn)`;JSON 天然原子,SQLite 包事务 | — |
+| 二级索引/条件查询 | 内存过滤不动(万级记录) | SQLite JSON1 查 value 列,加只读 query 面 | JSON 后端不陪跑 |
+| session 跨 workspace 移动 | 产品需求出现 | attach 校验放宽为"先 detach 后 attach"编排 | — |
+| RPC/GUI/boot | 下期 | `workspace.*` + `session.delete` 端点、wire schema、boot 挂载、sidebar 接真数据 | 本期模型与语义即 wire 投影的直接来源 |
+
+## Alternatives considered
+
+- **复用 session-persistence 的 coordinator/后端**:事件日志语义(append-only、turn 崩溃修复、懒物化)与 KV 覆写语义不匹配;只借其分层思想(协调层持写序、后端只实现最小原语)。
+- **workspace 专用存储包,后续再抽 seam**:第二个消费者(session sidecar)已可预见,届时泛化要再动一次接口。
+- **domain 与 storage 合为一层**:后端会被迫接触 schema 校验、变更事件、写串行等领域关切;拆开后 storage 后端只做不透明原语(可替换面最小),domain 单实现收敛全部领域逻辑(zod/事件/串行化只写一遍,不随后端翻倍)。
+- **整库单后端二选一(学 session-persistence 单坑位模式)**:曾是初版方案;改为多后端并存 + 配置路由,因为存储枢纽要承载多种数据形式,不同形式/域对后端的偏好(肉眼可读 vs 高频点更新)注定分化,单坑位会逼出"整体换挂 + 手工导数据"的粗粒度动作。代价是按名查找多一步,fail-loud 兜底。
+- **JSON 后端 jsonl 追加 + 墓碑 + 压实**:temp+fsync+rename 的崩溃安全与 append 等价;覆写让文件永远是净值、肉眼可读,免掉折叠/压实/断行容错。域规模下整写与追加一行同量级。
+- **JSON 一表一文件**:覆写下文件粒度不影响写成本,按域合并文件更少,global 单例有落点。
+- **SQLite 整域存单行 blob**:任何一条记录变更都重写整域,失去按 key 精确更新——SQLite 相对 JSON 的唯一优势归零。
+- **SQLite 按 schema 生成 typed columns**:DDL 生成器过度建设;document-per-row 足够,查询需求出现再议。
+- **每域独立 sqlite db 文件**:与仓库一库多表惯例相反。
+- **path 作为 workspace key**:规范化/符号链接解析会改写 path;引用锚点必须稳定。
+- **归属用 cwd 派生(或与账合并)**:双真相源;cwd 表达不了排序;归属本就是 workspace 侧事实。
+- **变更事件带旧值**:仓库变更事件惯例是"新快照 + 操作判别"(唯一例外 fs 的 before/after 是方法返回值而非事件,因旧值事后不可重建且有 diff 消费者);需要 diff 的消费者自己持有上次快照。
+- **删除自动 cancel 运行中 session**:持久层/编排层反向牵动运行时,层次变脏;cancel 机制已存在,调用方组合即可。
+
+## Acceptance criteria
+
+- 测试矩阵本期四套件全绿:backend 契约共享套件在 json/sqlite 双端、registry/mount disposer 语义、domain 层(含 open 六步与路由 fail-loud)、workspace 全语义(create/attach 校验/一致性口径)。
+- `ctx.workspace` 可在测试组装下完成 create → attach → list 生命周期(删除为 future work)。
+- session-persistence 包零 diff(本期不动 session 侧的验收线)。
+- 本期无新快照(无模型可见面与组装面);下期 RPC 接线时补。
+
+## Risks
+
+- **仓库持久化面第一个推式变更事件**(session-persistence 靠 revision 轮询):形态虽有 `goal/changed` 范本,但"存储层发事件"是新先例,下期 RPC 消费时才能验证形态是否合适。
+- **JSON 后端整域覆写的规模前提**:若第二个消费者(session sidecar)在路由到 SQLite 前就以千级记录落在 JSON 后端,整写成本会先于预期显现;缓解即 `routes` 改指 sqlite。
+- **删除语义的编排层检查依赖 `ctx.sessions` 弱依赖**:headless 组装拿不到运行时注册表时按"无热 session"处理,存在窗口(外部进程正在跑该 session);多进程本就在不做清单内,接受。
+- **facet 泛化以未来的 `log` facet 为设计依据但本期不实现它**:存在"预留形状不合身"的风险;缓解是本期后端介质代码按复用审计表的下沉形状组织,`log` facet 真正落地时只动 facet 层。
diff --git a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.i18n.yaml b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.i18n.yaml
index ba5d8cd8c9..23961c33c7 100644
--- a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.i18n.yaml
+++ b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.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
-2026-06-30-pre-tool-input-rewrite.md: 5de605e7edf63046b14a125d53072aab9bb5d37e
-2026-06-30-pre-tool-input-rewrite.zh.md: eab80764f5dfbd0448dabf0dc3c893ba2ec94928
+2026-06-30-pre-tool-input-rewrite.md: f35e6af465ce8cec5685911c43e12b8dd66f2e6a
+2026-06-30-pre-tool-input-rewrite.zh.md: c94c647bb6867199b72528bc84c58a08ae93e27e
diff --git a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.md b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.md
index 5de605e7ed..f35e6af465 100644
--- a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.md
+++ b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.md
@@ -14,7 +14,7 @@ In the loop, a tool call's arguments are committed to the log and read by live c
1. **`assistant/message`** is appended before tool dispatch — it is the model-history source `deriveMessages()` replays, so it carries the tool-call arguments the model itself emitted.
2. **`tool/call`** is the durable AUDIT record, appended before `ctx.tools.execute()`.
-3. **Live presentation reads `tool/call.arguments`**: the ACP bridge remembers them and passes them to `presentResult`; `dsh-tool-bash` derives the card title, the rawInput, the cwd, and the terminal-vs-background treatment from them.
+3. **Human-facing presentation reads `tool/call.arguments`**: UI renderers pass them to `presentResult`; `dsh-tool-bash` derives the card title, the rawInput, the cwd, and the terminal-vs-background treatment from them.
An execution-only rewrite would make the UI show one command while another ran and render the result against the wrong arguments. The registry prevents that failure mode today: it structured-clones and deep-freezes `arguments`, makes the execution identity properties non-writable, and exposes no test shim or listener path that can replace them. The rewrite design must preserve that protected-identity boundary rather than weaken it.
diff --git a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.zh.md b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.zh.md
index eab80764f5..c94c647bb6 100644
--- a/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.zh.md
+++ b/.agents/notes/proposed/feature/2026-06-30-pre-tool-input-rewrite.zh.md
@@ -14,7 +14,7 @@ Status: proposed
1. **`assistant/message`** 在工具分发之前追加——它是 `deriveMessages()` 回放时的模型历史来源,因此携带模型自身输出的工具调用参数。
2. **`tool/call`** 是持久化的审计记录,在 `ctx.tools.execute()` 之前追加。
-3. **展示层实时读取 `tool/call.arguments`**:ACP(Agent Client Protocol)桥接记住这些参数并传给 `presentResult`;`dsh-tool-bash` 从中派生卡片标题、rawInput、cwd 以及终端/后台处理方式。
+3. **面向人类的展示读取 `tool/call.arguments`**:UI 渲染器将这些参数传给 `presentResult`;`dsh-tool-bash` 从中派生卡片标题、rawInput、cwd 以及终端/后台处理方式。
如果只做执行层面的重写,UI 会显示一条命令而实际运行的是另一条,并且结果会对着错误的参数渲染。注册表目前通过以下方式防止这种失败模式:对 `arguments` 做 structured-clone 并深度冻结,将执行身份属性设为不可写,且不暴露任何可替换它们的测试 shim 或监听路径。重写设计必须维护这一受保护的身份边界,而非削弱它。
diff --git a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.i18n.yaml b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.i18n.yaml
index f64160a5a0..b2f1ade046 100644
--- a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.i18n.yaml
+++ b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.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
-2026-07-14-sdk-developer-projects.md: aa5cf64d7dd33dea229d74c2ae45a9244ee70e3c
-2026-07-14-sdk-developer-projects.zh.md: 8f7d1de5b16f38019c802f07eda701cee72deb4f
+2026-07-14-sdk-developer-projects.md: 65d2bf66232993222832eb0f2f4f56cfcf7afd16
+2026-07-14-sdk-developer-projects.zh.md: 8435a07d9b2545a8f41a1f96743c9c7e6d4daf3d
diff --git a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.md b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.md
index aa5cf64d7d..65d2bf6623 100644
--- a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.md
+++ b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.md
@@ -59,7 +59,7 @@ The table is the developer-visible support set for this phase. A `required` feat
| `hooks` | optional | `claude` (default) / `codex`, multiple | Each feature option creates a separate editable configuration file |
| `guard` | optional | `repeat-tool` | Provides repeated-tool-call reminders |
| `timeout-policy` | optional | `default` | Applies a uniform policy to tools that declare timeout budgets |
-| `ask-user` | optional | `default` | Provides the `ask_user_question` tool; only `acp` and `tui` can select it because those two feature options provide the injected user-interaction service |
+| `ask-user` | optional | `default` | Provides the `ask_user_question` tool; only `tui` can select it because ACP is an automation transport and embed provides no human-interaction service |
Both `bash` feature options apply to ACP, TUI, and embed and are not selected by the run interface. The sandbox feature option writes no active config key and therefore keeps `dsh-bash-sandbox`'s `read-only` default. Generated `cordis.yml` includes a commented example that developers can change explicitly to `workspace-write`:
@@ -107,7 +107,7 @@ Generated `package.json` provides the following scripts. `dev`, `build`, `start`
`dsh-sdk start` and `dsh-sdk dev` accept a module target and forward arguments after `--` unchanged to the project entrypoint. Generic argument parsing uses Node `parseArgs()` with zero schema: valued flags use `--key=value`, bare flags become `true`, and `--no-*` becomes `false`.
- TUI projects pass the selected model through `--model=` and create or resume an agent according to optional `--resume=`;
-- ACP uses protocol `session/load`
+- ACP clients create fresh sessions through protocol `session/new`;
- Embed uses the model written into the generated code.
Each feature-owned Cordis config entry keeps its developer-editable Cordis plugin config and explanatory comments in `cordis.yml`. When `dsh-sdk config` changes other features, it preserves unknown fields, formatting on untouched nodes, and comments. HMR is an ordinary leaf config entry: when the feature is selected, dev and start load the same watcher, and the command does not change the plugin tree implicitly.
diff --git a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.zh.md b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.zh.md
index 8f7d1de5b1..8435a07d9b 100644
--- a/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.zh.md
+++ b/.agents/notes/proposed/feature/2026-07-14-sdk-developer-projects.zh.md
@@ -59,7 +59,7 @@ create 还提供一次 `none / plugin / tool` 选择。`plugin` 固定生成 `pl
| `hooks` | optional | `claude`(默认)/ `codex`,可多选 | 各功能选项生成独立的可编辑配置文件 |
| `guard` | optional | `repeat-tool` | 提供重复工具调用提醒 |
| `timeout-policy` | optional | `default` | 对声明超时预算的工具执行统一策略 |
-| `ask-user` | optional | `default` | 提供 `ask_user_question` 工具;注入的 user-interaction 服务由 acp/tui 两个功能选项提供,因此仅这两个接口可选 |
+| `ask-user` | optional | `default` | 提供 `ask_user_question` 工具;只有 `tui` 可选,因为 ACP 是自动化传输,而 embed 不提供人类交互服务 |
`bash` 的两个功能选项都适用于 ACP、TUI 和 embed,不由运行接口决定。sandbox 功能选项不写任何生效的配置键,因而沿用 `dsh-bash-sandbox` 的 `read-only` 默认值;生成的 `cordis.yml` 保留注释示例,开发者可以显式改为 `workspace-write`:
@@ -107,7 +107,7 @@ my-agent/
`dsh-sdk start` 与 `dsh-sdk dev` 可以接收模块 target,并把 `--` 后的参数原样转发给工程入口。通用参数解析使用 Node `parseArgs()` 的零 schema 模式:带值 flag 采用 `--key=value`,bare flag 转换为 `true`,`--no-*` 转换为 `false`。
- TUI 工程通过 `--model=` 传入所选 model,并根据可选的 `--resume=` 创建或恢复 agent;
-- acp 使用协议 `session/load`
+- ACP 客户端通过协议 `session/new` 创建全新会话;
- embed 使用生成代码中的 model。
每个功能拥有的 Cordis 配置项在 `cordis.yml` 中保留自己的可编辑 Cordis 插件配置和说明注释;`dsh-sdk config` 修改其他功能时必须保留未知字段、未修改节点的格式和注释。HMR(热模块替换)是普通叶子配置项:选择该功能后,dev 和 start 加载同一个 watcher,命令不隐式改变插件树。
diff --git a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.i18n.yaml b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.i18n.yaml
index 024adc68d9..19915e8383 100644
--- a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.i18n.yaml
+++ b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.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
-2026-07-04-prune-dead-core-spine-surface.md: 432aaa848540fe7d1db3039399345a02afed5bd3
-2026-07-04-prune-dead-core-spine-surface.zh.md: 2c7cc4c0dc234afc107bdfc715cf717e5ad8c55c
+2026-07-04-prune-dead-core-spine-surface.md: a6c608617415f3af07de5c95fd20b0bde40bdef3
+2026-07-04-prune-dead-core-spine-surface.zh.md: 83603d8a8432b99d8f42222442b38005e196ac4e
diff --git a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.md b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.md
index 432aaa8485..a6c6086174 100644
--- a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.md
+++ b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.md
@@ -17,7 +17,7 @@ The production corpus is `packages/*/*/src`, example sources/config, and runtime
| `ReactLoopAgent` root export | Outside-package named imports are tests; production programs against `Agent` and creates/resumes through `ctx.agents`. | Return/interface-type `Agent` and make the concrete loop class package-internal; keep the deliberate synchronous config-only `AgentLoop.create()` path. |
| `workflow-workerthread` protocol/runtime/session re-exports and named `WorkerWorkflowEngine` | Every package-name consumer uses the default engine; the workflow Agent Note already defines the worker wire protocol as private. | Keep the default plugin class/config contract; drop the duplicate named class export and keep protocol modules source-private. |
| `code-runtime-worker` protocol/bootstrap re-exports | Outside-package production/e2e consumers use `WorkerCodeRuntime` and config, not `BootstrapPort`, `PatchableStream`, or worker message/boot types. | Keep the runtime class/config contract and make its wire/bootstrap vocabulary source-private. |
-| ACP translation/presenter root exports | `agentOptions`, `streamSessionEventUpdate`, `todosToPlan`, `ToolPresenter`, `nullToolPresenter`, and `TerminalRendering` have only same-file or ACP-test consumers; the sole outside-package production consumer mounts the plugin namespace. | Keep `name`, `inject`, `Config`, `AcpConfig`, and `apply`; make translation/presentation helpers source-private and test them in-package. |
+| ACP `agentOptions` root export | The helper has only same-file and ACP-test consumers; the sole outside-package production consumer mounts the plugin namespace. | Keep `name`, `inject`, `Config`, `AcpConfig`, and `apply`; make `agentOptions` source-private and test it through bridge behavior. |
| `providerWording` and `completedTurnPrefix` root exports | Each has one same-package production caller; only the balanced-prefix helper has a same-package white-box test. | Make them source-private and test provider behavior. |
| `depthOf`, `SubagentDepthError`, `SENSITIVE_ENV_PATTERN`, `waitForExit`, and `exitsWithin` root exports | Production subagent backends consume the in-process runner and subprocess construction/disposal helpers, not these enforcement/test internals. | Keep depth/environment/exit behavior but make the helpers and error/regex source-private; test through spawn and disposal. |
| `PersistenceCoordinator.inits`, backend `inits` accessors, `seedCoversPrefix`, and `assertSerializable` | The accessors exist for white-box tests; `seedCoversPrefix` has no outside production importer; `assertSerializable` has no production caller and duplicates the coordinator append boundary's lossless snapshot. | Observe initialization through `session/flush`, make `seedCoversPrefix` source-private, and delete `assertSerializable`. Keep both backends, `SessionHeader`, and SQLite's version contract. |
diff --git a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.zh.md b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.zh.md
index 2c7cc4c0dc..83603d8a84 100644
--- a/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.zh.md
+++ b/.agents/notes/proposed/simplification/2026-07-04-prune-dead-core-spine-surface.zh.md
@@ -17,7 +17,7 @@ Status: proposed
| `ReactLoopAgent` 根导出 | 包外的命名导入都是测试;生产代码面向 `Agent` 编程,通过 `ctx.agents` 创建/恢复。 | 返回/接口类型为 `Agent`,将具体循环类改为包内部;保留有意设计的同步、仅配置的 `AgentLoop.create()` 路径。 |
| `workflow-workerthread` 的 protocol/runtime/session 再导出与命名的 `WorkerWorkflowEngine` | 每个包名消费方都使用默认引擎;工作流 Agent Note 已将 worker 协议格式(wire format)定义为私有。 | 保留默认插件类/配置契约;移除重复的命名类导出,将协议模块保持为源码私有。 |
| `code-runtime-worker` 的 protocol/bootstrap 再导出 | 包外的生产/e2e 消费方使用 `WorkerCodeRuntime` 和配置,而非 `BootstrapPort`、`PatchableStream` 或 worker 消息/启动类型。 | 保留运行时类/配置契约,将其协议格式/bootstrap 词汇改为源码私有。 |
-| ACP 的 translation/presenter 根导出 | `agentOptions`、`streamSessionEventUpdate`、`todosToPlan`、`ToolPresenter`、`nullToolPresenter` 和 `TerminalRendering` 只有同文件或 ACP 测试消费方;唯一的包外生产消费方挂载的是插件命名空间。 | 保留 `name`、`inject`、`Config`、`AcpConfig` 和 `apply`;将 translation/presentation 辅助函数改为源码私有,在包内测试。 |
+| ACP 的 `agentOptions` 根导出 | 该辅助函数只有同文件和 ACP 测试消费方;唯一的包外生产消费方挂载的是插件命名空间。 | 保留 `name`、`inject`、`Config`、`AcpConfig` 和 `apply`;将 `agentOptions` 改为源码私有,通过桥接层行为测试。 |
| `providerWording` 与 `completedTurnPrefix` 根导出 | 各有一个同包生产调用者;只有 balanced-prefix 辅助函数有一个同包白盒测试。 | 改为源码私有,测试提供方行为。 |
| `depthOf`、`SubagentDepthError`、`SENSITIVE_ENV_PATTERN`、`waitForExit` 与 `exitsWithin` 根导出 | 生产 subagent 后端消费的是进程内 runner 和子进程构造/dispose(资源释放)辅助函数,而非这些强制/测试内部实现。 | 保留深度/环境/退出行为,但将辅助函数和 error/regex 改为源码私有;通过 spawn 和 dispose 测试。 |
| `PersistenceCoordinator.inits`、后端 `inits` 访问器、`seedCoversPrefix` 与 `assertSerializable` | 访问器为白盒测试而存在;`seedCoversPrefix` 没有包外生产导入者;`assertSerializable` 没有生产调用者,且与 coordinator append 边界的无损快照重复。 | 通过 `session/flush` 观察初始化,将 `seedCoversPrefix` 改为源码私有,删除 `assertSerializable`。保留两个后端、`SessionHeader` 和 SQLite 的版本契约。 |
diff --git a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.i18n.yaml b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.i18n.yaml
similarity index 58%
rename from .agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.i18n.yaml
rename to .agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.i18n.yaml
index 5c7c7e9cc1..0ee4dd2632 100644
--- a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.i18n.yaml
+++ b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.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
-2026-07-13-stream-workflow-progress-through-tool-calls.md: 668d38b38cab90f6ee9b613d1101d4d44302aa8a
-2026-07-13-stream-workflow-progress-through-tool-calls.zh.md: b0daba0c62c9b06f33f5e3c6319f90670941e762
+2026-07-13-stream-workflow-progress-through-tool-calls.md: 1b299ec323d32745cc504a90948b63a4dcaae64f
+2026-07-13-stream-workflow-progress-through-tool-calls.zh.md: f84ff3bb22a19ed7ad2f9fc262a6702e254972ad
diff --git a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md
similarity index 98%
rename from .agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md
rename to .agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md
index 668d38b38c..1b299ec323 100644
--- a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md
+++ b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.md
@@ -1,6 +1,6 @@
# Agent Note: Stream workflow progress through tool calls
-Status: proposed
+Status: rejected — ACP is automation-only; live workflow presentation needs a human-interface owner and a fresh design.
English | [中文](2026-07-13-stream-workflow-progress-through-tool-calls.zh.md)
diff --git a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md
similarity index 98%
rename from .agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md
rename to .agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md
index b0daba0c62..f84ff3bb22 100644
--- a/.agents/notes/proposed/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md
+++ b/.agents/notes/rejected/feature/2026-07-13-stream-workflow-progress-through-tool-calls.zh.md
@@ -1,6 +1,6 @@
# Agent Note: 通过工具调用流式传输工作流进度
-Status: proposed
+Status: rejected — ACP 仅面向自动化;实时工作流展示需要一个面向人类界面的归属方和全新设计。
[English](2026-07-13-stream-workflow-progress-through-tool-calls.md) | 中文
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.i18n.yaml b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.i18n.yaml
index d06734c3ea..e4e6a6613b 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.i18n.yaml
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.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
-2026-06-20-drop-acp-session-load.md: 2b6edf173506929f2d64e910b47aa70fb4f1d854
-2026-06-20-drop-acp-session-load.zh.md: 8e03acad17cf2831a15d5992a0419a73f275719d
+2026-06-20-drop-acp-session-load.md: bae71ba2968bbb10503619e764694a6712572efd
+2026-06-20-drop-acp-session-load.zh.md: cdf49039e889f8528f488b53ad01cc13eab2b9d6
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.md b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.md
index 2b6edf1735..bae71ba296 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.md
@@ -20,7 +20,7 @@ For now, ACP starts fresh sessions only. `initialize` advertises `loadSession: f
- `initialize` does not advertise load support.
- The `session/load` handler, loading-id tracking, cwd preflight for loaded sessions, and load replay tests are removed.
- Snapshot fixtures no longer rely on load replay presentation.
-- [ACP docs](../../../../packages/ui/acp/README.md) describe fresh-session support only.
+- [ACP docs](../../../../packages/acp/acp/README.md) describe fresh-session support only.
## What we give up
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.zh.md b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.zh.md
index 8e03acad17..cdf49039e8 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.zh.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-session-load.zh.md
@@ -20,7 +20,7 @@ ACP 声明 `loadSession: true` 并实现 `session/load`:向 bridge 注入持
- `initialize` 不再声明 load 支持。
- `session/load` handler、loading-id 追踪、已加载会话的 cwd 预检以及 load 回放测试均被移除。
- 快照 fixture(测试前置数据)不再依赖 load 回放展示。
-- [ACP 文档](../../../../packages/ui/acp/README.md)仅描述全新会话的支持。
+- [ACP 文档](../../../../packages/acp/acp/README.md)仅描述全新会话的支持。
## 放弃的能力
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.i18n.yaml b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.i18n.yaml
index 45949d3fcf..b7b5632e38 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.i18n.yaml
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.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
-2026-06-20-drop-acp-terminal-meta.md: 79da387ac1a7a0e6767e3bf24baa6039e39ef90d
-2026-06-20-drop-acp-terminal-meta.zh.md: d29fd54618611f56fd071a0ee4a63bc207895d89
+2026-06-20-drop-acp-terminal-meta.md: 84b9028392f967e72f7b5585d013d669735631de
+2026-06-20-drop-acp-terminal-meta.zh.md: 6ac7ba46bce20bcf2593ef422057e0562f8a1557
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.md b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.md
index 79da387ac1..84b9028392 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.md
@@ -1,14 +1,14 @@
# Agent Note: Drop ACP terminal `_meta` rendering
-Status: rejected — Zed is the current target client, and the terminal `_meta` convention is intentional Zed UX with a plain ACP fallback for other clients.
+Status: rejected — removing only Zed terminal metadata was rejected while ACP remained an editor bridge; automation-only ACP later removed the whole editor projection.
English | [中文](2026-06-20-drop-acp-terminal-meta.zh.md)
## Problem
-The ACP bridge implements a Zed-specific terminal-card convention through `_meta.terminal_info`, `_meta.terminal_output`, and `_meta.terminal_exit`. The implemented [rich ACP bash rendering Agent Note](../../implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md) deliberately avoided ACP's client-side `terminal/create` because bash execution belongs in the harness, but still adopted the reference agents' display-only `_meta` convention. That gives a nicer Zed card at the cost of bridge state, capability negotiation, terminal ids, special update mapping, text fallback tests, and exit-pill parsing in `dsh-tool-bash`.
+The former ACP editor bridge implemented a Zed-specific terminal-card convention through `_meta.terminal_info`, `_meta.terminal_output`, and `_meta.terminal_exit`. The current [render-intent decision](../../implemented/architecture/2026-07-02-tool-render-intent-union.md) preserves the underlying rule that bash execution belongs in the harness and terminal cards are display-only. The later [automation-only ACP decision](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md) removes the `_meta` projection, bridge state, capability negotiation, terminal ids, special update mapping, text fallback tests, and exit-pill parsing from ACP. TUI and the Web host/client runtime retain the tagged presentation contract, while ACP no longer renders editor cards.
-The fallback path already exists: render the tool call and completed output as normal ACP content blocks. Non-Zed clients rely on that path anyway, but the Zed terminal card is a current target-client feature rather than speculative decoration.
+At proposal time, the fallback path already existed: render the tool call and completed output as normal ACP content blocks. Non-Zed clients relied on that path, but the Zed terminal card was a target-client feature rather than speculative decoration.
## Proposal
@@ -22,10 +22,10 @@ This proposal is narrower than [collapsing tool-owned UI presentation](2026-06-2
- `TerminalRendering`, terminal ids, terminal cwd resolution, and `_meta.terminal_*` update mapping disappear from `@deepseek-ai/dsh-acp`.
- `ToolTerminal` disappears from `@deepseek-ai/dsh-tools`, or is unused and deleted with the presentation cleanup.
- Bash result presentation no longer parses exit status for terminal pills.
-- The implemented [rich ACP bash rendering Agent Note](../../implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md) stays in `implemented/` as shipped history and is cross-linked from this proposal if superseded.
+- The [automation-only ACP decision](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md) later removes ACP terminal cards and absorbs their execution-ownership rationale.
## What we give up
-Zed users lose the dedicated terminal card: no cwd header, terminal display, or exit pill. They still see the command and output as plain content. That is a reasonable simplification while the ACP bridge is still unreleased and the `_meta` keys are a convention rather than a standard.
+Under this proposal, Zed users would lose the dedicated terminal card: no cwd header, terminal display, or exit pill. They would still see the command and output as plain content. That was a reasonable simplification to consider while the ACP bridge was unreleased and the `_meta` keys were a convention rather than a standard.
diff --git a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.zh.md b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.zh.md
index d29fd54618..6ac7ba46bc 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.zh.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-drop-acp-terminal-meta.zh.md
@@ -1,14 +1,14 @@
# Agent Note: 移除 ACP(Agent Client Protocol)终端 `_meta` 渲染
-Status: rejected — Zed 是当前目标客户端,terminal `_meta` 约定是有意设计的 Zed UX,同时为其他客户端保留普通 ACP 回退。
+Status: rejected — 在 ACP 仍是编辑器桥接层时,仅移除 Zed 终端元数据的方案被否决;后续仅面向自动化的 ACP 则移除了整个编辑器投影。
[English](2026-06-20-drop-acp-terminal-meta.md) | 中文
## 问题
-ACP 桥接层通过 `_meta.terminal_info`、`_meta.terminal_output` 和 `_meta.terminal_exit` 实现了一套 Zed 特有的终端卡片约定。已实现的[富 ACP bash 渲染 Agent Note(agent 决策记录)](../../implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md)刻意回避了 ACP 客户端侧的 `terminal/create`(因为 bash 执行属于 harness 职责),但仍采用了参考 agent(智能体)的纯展示 `_meta` 约定。这在 Zed 中带来了更好的卡片效果,代价是桥接状态、能力协商、终端 id、特殊的 update 映射、文本回退测试,以及 `dsh-tool-bash` 中的 exit-pill 解析。
+原 ACP 编辑器桥接层通过 `_meta.terminal_info`、`_meta.terminal_output` 和 `_meta.terminal_exit` 实现了一套 Zed 特有的终端卡片约定。当前的 [render-intent 决策](../../implemented/architecture/2026-07-02-tool-render-intent-union.md)保留了底层规则:bash 执行属于 harness,terminal 卡片只用于展示。后续的[仅面向自动化 ACP 决策](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md)从 ACP 中移除了 `_meta` 投影、桥接状态、能力协商、终端 id、特殊 update 映射、文本回退测试和 exit-pill 解析。TUI 与 Web 宿主/客户端运行时保留带标签的展示契约,而 ACP 不再渲染编辑器卡片。
-回退路径已经存在:将工具调用和完成输出渲染为普通 ACP 内容块。非 Zed 客户端本来就依赖这条路径,但 Zed 终端卡片是当前目标客户端的功能特性,而非推测性装饰。
+本提案提出时,回退路径已经存在:将工具调用和完成输出渲染为普通 ACP 内容块。当时,非 Zed 客户端依赖这条路径,但 Zed 终端卡片是目标客户端的功能特性,而非推测性装饰。
## 提案
@@ -22,10 +22,10 @@ ACP 桥接层通过 `_meta.terminal_info`、`_meta.terminal_output` 和 `_meta.t
- `TerminalRendering`、终端 id、终端 cwd 解析与 `_meta.terminal_*` update 映射从 `@deepseek-ai/dsh-acp` 中消失。
- `ToolTerminal` 从 `@deepseek-ai/dsh-tools` 中消失,或在展示清理中因未使用而删除。
- Bash 结果展示不再为终端 pill 解析退出状态。
-- 已实现的[富 ACP bash 渲染 Agent Note](../../implemented/feature/2026-06-18-acp-terminal-and-tool-rendering.md) 作为已交付历史保留在 `implemented/` 中;如被本提案取代,则加上交叉链接。
+- [仅面向自动化 ACP 决策](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md)后来移除了 ACP 终端卡片,并吸收了其中有关执行归属的决策依据。
## 放弃的内容
-Zed 用户将失去专用终端卡片:没有 cwd 头部、终端展示或 exit pill。他们仍能以纯内容形式看到命令和输出。在 ACP 桥接层尚未发布、`_meta` 键只是约定而非标准的阶段,这是合理的简化。
+如果采用本提案,Zed 用户会失去专用终端卡片:没有 cwd 头部、终端展示或 exit pill。但他们仍会以纯内容形式看到命令和输出。当时 ACP 桥接层尚未发布,且 `_meta` 键只是约定而非标准;在这种情况下,考虑这项简化是合理的。
diff --git a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.i18n.yaml b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.i18n.yaml
index da93063670..14c7033185 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.i18n.yaml
+++ b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.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
-2026-06-20-generic-tool-rendering.md: 6fc610546da04e7d1e16fc17ada87483a142aa3c
-2026-06-20-generic-tool-rendering.zh.md: 11386b87d845129950a8473eb1cf4ea6ce697ac8
+2026-06-20-generic-tool-rendering.md: a9ceb7a0e016b57295e3226e98a7fce51e49c21f
+2026-06-20-generic-tool-rendering.zh.md: 553c1caa23ed34eea5f113372d12a7381dc2488a
diff --git a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.md b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.md
index 6fc610546d..a9ceb7a0e0 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.md
@@ -1,14 +1,16 @@
# Agent Note: Collapse tool-owned UI presentation
-Status: rejected — tool-owned presentation should wait for more real tools before being generalized or deleted. Bash and ACP currently need the existing richer presentation path.
+Status: rejected — TUI and the Web host/client runtime consume the tagged render-intent union, so tool-owned presentation remains current even though ACP no longer projects it.
English | [中文](2026-06-20-generic-tool-rendering.zh.md)
## Problem
-Tools can define `presentCall()` and `presentResult()` callbacks that return `ToolCallPresentation`, `ToolResultPresentation`, and optional `ToolTerminal` fields. The code itself flags the design as muddy: title, kind, raw input, content, terminal cwd, terminal output, exit code, and signal grew incrementally into a bag of optional fields. ACP then maintains pending call state to pair a result with the original args, creates replay-only presenters on `session/load`, and maps terminal subfields into Zed-specific `_meta`. `dsh-tool-bash` even parses exit status back out of rendered text because the pure replay-safe presenter no longer has the structured `BashRunResult`.
+The optional-field bag and ACP editor mapping below were the proposal-time context for this rejection. The current contracts live in [the tagged render-intent union](../../implemented/architecture/2026-07-02-tool-render-intent-union.md) and [automation-only ACP](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md).
-The real first-party use is bash presentation for ACP. That is too little evidence to freeze a cross-package UI presentation API.
+Tools could define `presentCall()` and `presentResult()` callbacks that returned `ToolCallPresentation`, `ToolResultPresentation`, and optional `ToolTerminal` fields. The code itself flagged the design as muddy: title, kind, raw input, content, terminal cwd, terminal output, exit code, and signal had grown incrementally into a bag of optional fields. ACP then maintained pending call state to pair a result with the original args, created replay-only presenters on `session/load`, and mapped terminal subfields into Zed-specific `_meta`. `dsh-tool-bash` even parsed exit status back out of rendered text because the pure replay-safe presenter no longer had the structured `BashRunResult`.
+
+The real first-party use was bash presentation for ACP. That was too little evidence to freeze a cross-package UI presentation API.
## Proposal
@@ -28,8 +30,8 @@ As a smaller alternative, replace the current optional-field bag with one explic
## What we give up
-Bash loses its custom terminal-looking card and model-written description placement. The fallback remains reasonable: the command appears as tool input, and the output appears as text. Rich rendering should be designed when the product has enough UI/tool variety to justify a stable presentation contract.
+Under this proposal, Bash would lose its custom terminal-looking card and model-written description placement. The fallback would remain reasonable: the command would appear as tool input, and the output as text. Rich rendering would be designed when the product had enough UI/tool variety to justify a stable presentation contract.
## Related
-This is the broad version of [dropping ACP terminal metadata](2026-06-20-drop-acp-terminal-meta.md). If this Agent Note is accepted, that narrower Agent Note becomes unnecessary.
+The later [tagged render-intent union](../../implemented/architecture/2026-07-02-tool-render-intent-union.md) implements the smaller alternative once multiple producer and consumer families provide enough evidence for the vocabulary. [Automation-only ACP](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md) removes ACP's editor projection without removing tool-owned presentation from TUI or the Web host/client runtime.
diff --git a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.zh.md b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.zh.md
index 11386b87d8..553c1caa23 100644
--- a/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.zh.md
+++ b/.agents/notes/rejected/simplification/2026-06-20-generic-tool-rendering.zh.md
@@ -1,14 +1,16 @@
# Agent Note: 收拢工具自有的 UI 展示逻辑
-Status: rejected — 工具拥有的呈现机制应等到出现更多真实工具后再进行泛化或删除。Bash 与 ACP(Agent Client Protocol)目前仍需要现有的丰富呈现路径。
+Status: rejected — 尽管 ACP(Agent Client Protocol)已不再投影这套契约,TUI 与 Web 宿主/客户端运行时仍消费带标签 render-intent 联合类型,因此工具自有的展示仍然有效。
[English](2026-06-20-generic-tool-rendering.md) | 中文
## 问题
-工具可以定义 `presentCall()` 和 `presentResult()` 回调,返回 `ToolCallPresentation`、`ToolResultPresentation` 以及可选的 `ToolTerminal` 字段。代码本身就标记了这个设计的混乱:title、kind、raw input、content、terminal cwd、terminal output、exit code 和 signal 逐步增长为一堆可选字段。ACP 随后维护 pending call 状态以将 result 与原始 args 配对,在 `session/load` 时创建仅用于回放的 presenter,并将 terminal 子字段映射为 Zed 特有的 `_meta`。`dsh-tool-bash` 甚至从渲染后的文本中反向解析退出状态,因为纯回放安全的 presenter 已经拿不到结构化的 `BashRunResult`。
+下文所述的可选字段集合与 ACP 的编辑器映射,是本提案遭否决时的背景。当前契约分别由[带标签 render-intent 联合类型](../../implemented/architecture/2026-07-02-tool-render-intent-union.md)与[ACP 作为仅面向自动化的协议](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md)承载。
-真正的第一方用途是为 ACP 提供 bash 展示。这不足以作为冻结一个跨包(package)UI 展示 API 的依据。
+当时,工具可以定义 `presentCall()` 和 `presentResult()` 回调,返回 `ToolCallPresentation`、`ToolResultPresentation` 以及可选的 `ToolTerminal` 字段。代码本身就标记了这个设计的混乱:title、kind、raw input、content、terminal cwd、terminal output、exit code 和 signal 已经逐步增长为一堆可选字段。ACP 随后维护 pending call 状态以将 result 与原始 args 配对,在 `session/load` 时创建仅用于回放的 presenter,并将 terminal 子字段映射为 Zed 特有的 `_meta`。`dsh-tool-bash` 甚至从渲染后的文本中反向解析退出状态,因为纯回放安全的 presenter 已经拿不到结构化的 `BashRunResult`。
+
+当时,真正的第一方用途是为 ACP 提供 bash 展示。这点证据不足以作为冻结一个跨包(package)UI 展示 API 的依据。
## 提案
@@ -28,8 +30,8 @@ Status: rejected — 工具拥有的呈现机制应等到出现更多真实工
## 放弃了什么
-Bash 失去其自定义的终端风格卡片和模型生成描述的放置位置。回退方案仍然合理:命令作为工具输入展示,输出作为文本展示。富展示应当在产品拥有足够的 UI/工具多样性、足以支撑一份稳定的展示契约时再行设计。
+如果采用本提案,Bash 会失去其自定义的终端风格卡片和模型生成描述的放置位置。届时,回退方案仍然合理:命令会作为工具输入展示,输出会作为文本展示。只有当产品拥有足够的 UI/工具多样性、足以支撑一份稳定的展示契约时,才会设计富展示。
## 相关
-这是[移除 ACP terminal 元数据](2026-06-20-drop-acp-terminal-meta.md)的宽泛版本。如果本 Agent Note(agent 决策记录)被接受,那个更窄的 Agent Note 就不再必要。
+后续的[带标签 render-intent 联合类型](../../implemented/architecture/2026-07-02-tool-render-intent-union.md)在多类生产者与消费方为这套词汇提供充分依据后,实现了较小的替代方案。[ACP 作为仅面向自动化的协议](../../implemented/simplification/2026-07-23-acp-automation-only-protocol.md)移除了 ACP 的编辑器投影,但没有从 TUI 或 Web 宿主/客户端运行时中移除工具自有的展示。
diff --git a/.agents/skills/dsh-find-simplifications/SKILL.md b/.agents/skills/dsh-find-simplifications/SKILL.md
index 2e1ccdc3c8..1c3b07e5a1 100644
--- a/.agents/skills/dsh-find-simplifications/SKILL.md
+++ b/.agents/skills/dsh-find-simplifications/SKILL.md
@@ -1,6 +1,6 @@
---
name: dsh-find-simplifications
-description: 'Use when working in the deepseek-harness repo to find non-obvious simplification candidates and write proposed Agent Notes or inline TODO/FIXME/XXX notes for dead, duplicated, speculative, or over-built code surfaces; especially for requests like "find simplification Agent Notes", "look for unnecessary complexity", "audit for removal-style cleanups", or "fold worthwhile simplification ideas from another PR".'
+description: 'Use when working in the deepseek-harness repo to find non-obvious simplification candidates, write proposed Agent Notes or inline TODO/FIXME/XXX notes, audit or coalesce superseded Agent Notes, or fold worthwhile simplification ideas from another PR; especially for dead, duplicated, speculative, over-built, or added-then-removed surfaces.'
---
# Finding DeepSeek Harness Simplifications
@@ -34,7 +34,7 @@ Thin candidates are usually not enough for an Agent Note: deleting one typo, run
Use parallel subagents when the user asks for breadth or many candidates. Give each agent a domain and require evidence, not guesses. Useful domains:
- Agent loop and session log: turn/step boundaries, steering, abort/cancel, durable events, replay, load/resume.
-- ACP and UI surfaces: `session/*` methods, terminal `_meta`, transcript rendering, single vs multi-session state.
+- ACP automation and human UI surfaces: prompt settlement and teardown on the protocol side; transcript rendering and interaction state on the UI side.
- LLM/tools/system prompt: stream/generate surfaces, assemblers, registries, tool schema defaults, presentation hooks.
- Bash and tool execution: foreground/background split, task ownership, output spill files, executor methods.
- Packages/examples/scripts/tests: package boundaries, static inventories, redundant snapshot expected outputs, support packages.
@@ -66,6 +66,22 @@ Reject or downgrade a candidate when:
- The removal would force unrelated churn without actually making the contract smaller.
- The idea is correct but tiny. Add a targeted TODO/FIXME/XXX instead, using the urgency semantics in [docs/development.md](../../../docs/development.md).
+## Coalesce Superseded Agent Notes
+
+Audit the Agent Note tree when the user asks to reduce or coalesce it, or when the simplification being implemented makes an owning note obsolete. Do not expand every code-simplification survey into a repository-wide note audit.
+
+Follow the deletion rule in the [Agent Note contract](../../notes/README.md#when-to-write-one); do not duplicate or weaken it here. For each candidate chain:
+
+1. Identify the current owner from shipped code, configuration, generated catalogs, package docs, newer Agent Notes, and inbound links; dates and titles are discovery hints, not proof.
+2. Classify the old note as fully or partially superseded. Any surviving behavior, current contract, durable format, compatibility obligation, or independently current rejected alternative makes it partial. Rationale that can be transferred to the current owner does not by itself make supersession partial.
+3. For full supersession, move every unique rationale, alternative, consequence, shipped verification contract, and named coverage gap into the current owner. An inventory that only describes deleted implementation mechanics is not one of those decision facts.
+4. Repair every inbound link, then delete the English note, Chinese counterpart, consistency record, and required-pair manifest entry together.
+5. Search exact filenames, symbols, config keys, event names, and wire strings after the edit. Keep partial supersessions cross-linked and current.
+
+An added-then-removed feature is a common full-supersession case. Let the removal note own the history only when the feature is absent from production code, configuration, schemas, durable or wire formats, migration, and compatibility behavior; no current documentation presents it as available; and no test exercises it as supported behavior. Removal rationale and tests that enforce absence may remain. Preserve why the feature originally existed, why that motivation no longer justified it, alternatives to full removal, the capability given up, conditions for reintroduction, and evidence that removal is complete. Old tests and implementation mechanics that verified only the deleted behavior are not current verification contracts.
+
+Reject consolidation when the removal is only one transport, default, implementation, or presentation of a feature; when persisted data or compatibility handling survives; or when the removal note does not yet carry enough rationale to prevent accidental reintroduction. A current negative design decision may legitimately need its own note even though the removed implementation is gone.
+
## Write The Agent Note
Create one file per durable proposal under `.agents/notes///yyyy-mm-dd-topic.md`, following the lifecycle/classification contract in `.agents/notes/README.md`. Keep prose paragraphs on one physical line and use relative Markdown links.
@@ -106,9 +122,11 @@ For docs-only Agent Note work, run at least `pnpm run doc-sync`, `pnpm run lint`
When opening or updating a PR, summarize:
-- How many Agent Notes and inline notes were added.
+- How many Agent Notes and inline notes were added, consolidated, retained as partial supersessions, or deleted.
- The main areas surveyed.
- What was intentionally excluded.
- Which checks passed.
+For each consolidation group, name the old and current owners, state the evidence for full supersession, and explain why deletion is safe. If an added-then-removed scan finds no qualifying note, report that result and the representative partial cases retained.
+
Use a draft PR while the survey is still expanding; mark ready only when the candidate set, review responses, and validation are settled.
diff --git a/.agents/skills/dsh-prose-standard/SKILL.md b/.agents/skills/dsh-prose-standard/SKILL.md
index faf234ae7c..37b070e466 100644
--- a/.agents/skills/dsh-prose-standard/SKILL.md
+++ b/.agents/skills/dsh-prose-standard/SKILL.md
@@ -7,6 +7,8 @@ description: Use when writing, reviewing, restoring, trimming, or auditing prose
Write enough to preserve the contract, then remove reasoning transcripts, repetition, and decoration. This skill owns editorial judgment and required prose coverage; use [dsh-doc-standards](../dsh-doc-standards/SKILL.md) for placement, budgets, bilingual pairs, and documentation gates. It is guidance, not a script.
+Comments describe non-obvious contracts or rationale that code cannot express; they do not restate what code already implies.
+
## Inputs and exclusions
Require an explicit `scope`. If it is missing, report the required input and stop; do not infer a repository-wide scope or begin an interview.
diff --git a/.gitignore b/.gitignore
index ae9b4b5ddd..d6b400aeb3 100644
--- a/.gitignore
+++ b/.gitignore
@@ -8,6 +8,7 @@ pnpm-debug.log
.cache/
examples/*/*.jsonl
.sessions/
+.storages/
examples/*/.sessions/
coverage/
.doc-typecheck-*/
diff --git a/AGENTS.md b/AGENTS.md
index 8007f32b90..1ce8e5e0c8 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -30,7 +30,8 @@ packages/ @deepseek-ai/dsh- workspaces at packages///
cordis/ self-referential toolset: the agent inspects/mounts plugins in its own runtime
hooks/ Claude Code / Codex hook bridges + shared wire-protocol library
session-persistence/ persistence seam + JSONL/SQLite backends
- ui/ ACP/TUI/JSON-RPC bridges; boot, approval, interaction plugins
+ acp/ automation-only Agent Client Protocol server
+ ui/ TUI/JSON-RPC bridges; boot, approval, interaction plugins
examples/ demo bundles (agent-spine + TUI/CLI/ACP/JSON-RPC bins) leaves load
support/ dev/test infrastructure packages
util/ zero-dependency utilities
@@ -49,6 +50,7 @@ Package groups: [packages/README.md](packages/README.md).
```sh
pnpm install # pnpm workspaces, node ^22.19 || >=24
+pnpm run clean # remove build outputs and safe residue from deleted packages
pnpm run test # vitest unit tests
pnpm run test:coverage # CI coverage gate: per-file 100% on packages/*/*/src
pnpm run test:e2e # real-API tests; self-skip without DEEPSEEK_API_KEY
@@ -64,7 +66,7 @@ pnpm run website:build # VitePress build (doubles as the site's dead-link check
pnpm run demo:headless "task" # one-shot agent (needs DEEPSEEK_API_KEY)
pnpm run demo:tui # full-screen TUI coding agent (needs DEEPSEEK_API_KEY)
pnpm run demo:cordis # self-referential demo: the agent modifies its own runtime (needs key)
-pnpm run demo:acp # ACP server agent (needs DEEPSEEK_API_KEY)
+pnpm run demo:acp # ACP automation server (needs DEEPSEEK_API_KEY)
```
### Host sandbox failures
@@ -107,7 +109,7 @@ Real-API tests and demos read `DEEPSEEK_API_KEY`, optional `DEEPSEEK_BASE_URL`,
- **Tests describe behavior, not correctness.** Change obsolete behavior with its tests; explain why in the PR.
- **Every non-trivial change MUST include at least one Agent Note in the same PR.** Update the owning note or add one, validate its premises against code, and exempt only mechanical/local edits ([scope](.agents/notes/README.md#when-to-write-one)).
- **Testing policy** — [docs/testing.md](docs/testing.md). Every non-trivial model- or product-user-visible behavior change adds or updates a keyless snapshot through a real runnable example in the same PR; package tests, e2e-only assertions, and mock-only fixtures do not substitute for the assembled application transcript. Fixtures must replay on macOS/Linux; fix fixtures, not normalizers.
-- **A tool's ACP render intent is part of its design**, decided up front (`generic`/`terminal`/`diff`, `locations`); presentation methods are pure functions of `args` ([cookbook](docs/cookbook/adding-a-tool.md)).
+- **A tool's UI render intent is part of its design**, decided up front (`generic`/`terminal`/`diff`, `locations`); presentation methods are pure functions of `args` ([cookbook](docs/cookbook/adding-a-tool.md)).
- **Plan unit, e2e, and snapshot coverage** for new seams, lifecycle shapes, and transcript surfaces; missing snapshot-harness support is part of the implementation, not deferred follow-up.
- **Keep PRs coherent and merge with merge commits.** Split an independently meaningful feature or design decision into a separate or stacked PR when combining it obscures ownership, intent, or verification. Never squash/rebase or rewrite pushed branches; put a review fix on its introducing PR, then merge down the stack ([guide](docs/cookbook/responding-to-pr-review-on-a-stack.md)).
- TODO markers: `FIXME`/`TODO`/`XXX` by urgency ([semantics](docs/development.md)).
diff --git a/apps/cli/README.md b/apps/cli/README.md
index e6a33247ca..4ff9034dd3 100644
--- a/apps/cli/README.md
+++ b/apps/cli/README.md
@@ -1,16 +1,18 @@
# `@deepseek-ai/dsh`
-The `dsh` command-line entry follows the `apps/` assembly tier: `apps/*` are product assemblies over `packages/*` libraries. Plain `dsh [config.yml]` boots the interactive TUI coding agent, `dsh -p "task"` runs one headless turn, and `dsh web` serves the browser UI.
+The `dsh` command-line entry follows the `apps/` assembly tier: `apps/*` are product assemblies over `packages/*` libraries. Plain `dsh` boots the interactive TUI coding agent, `dsh -p "task"` runs one headless turn, and `dsh web` serves the browser UI.
+
+Argv is parsed once through a [Commander](https://github.com/tj/commander.js) adapter ([`src/args.ts`](src/args.ts)): one program whose default (no subcommand) is the TUI/headless surface (`--config`, `-p`/`--prompt`, `--resume`) and whose `web` subcommand is the browser UI. `src/bin.ts` switches on the resolved mode and dynamic-imports only that mode's module. `dsh --help` lists every mode and `dsh web --help` renders the web usage, `dsh --version` prints this app's version, and an unknown option or a mistyped `--resume` fails loud (stderr, exit 1) instead of misrouting. `dsh web`'s `--host`/`--port` are unvalidated pass-through overrides: the `dsh-host-webserver` schema is the single source of both the default (the shipped `cordis.yml` value when a flag is absent) and validity, and rejects a bad value at boot.
The TUI surface:
-- boots the shipped default config (`examples/tui-agent/cordis.yml`) or an explicit config argument, through [`dsh-app-boot`](../../packages/ui/app-boot/README.md);
-- resumes a persisted session with `dsh --resume ` and, when the Node host exposes `process.execve`, supplies the TUI's in-place handoff host: after selector preflight and current-session flush, the host disposes the app and replaces the process with a normalized resume flag; runtimes without process replacement keep the displayed command fallback, the flag still sets `RESUME_SESSION_ID` before boot, and a missing or unreadable id fails loud instead of creating a fresh session;
+- boots the shipped default config (`examples/tui-agent/cordis.yml`), or the tree named by `--config ` (the demo/test escape for booting an alternate example tree), through [`dsh-app-boot`](../../packages/ui/app-boot/README.md);
+- resumes a persisted session with `dsh --resume ` and, when the Node host exposes `process.execve`, supplies the TUI's in-place handoff host: after selector preflight and current-session flush, the host disposes the app and replaces the process with a normalized `dsh --resume `; runtimes without process replacement keep the displayed command fallback. The flag provides the id on the boot context under `RESUME_SESSION_ID_KEY` (no environment variable), which the shipped config reads through `!!js`, and a missing or unreadable id fails loud instead of creating a fresh session;
- treats the **invoking directory** as the workspace — sessions, relative paths, and workspace instructions resolve from the cwd;
- tells the agent where its own source lives: after boot it adds a prompt section naming this harness checkout, resolved from the launcher's real path so it holds under a PATH symlink and an arbitrary cwd, so the self-referential `cordis` toolset can read and modify it;
- applies the personal overlay from `~/.dsh` (see [app-boot's Personal config](../../packages/ui/app-boot/README.md#personal-config)): `.env` fills environment gaps (ambient > project `.env` > personal `.env`), `config.yaml` patches the booted tree.
-The Web surface treats its invoking directory as the default project, loads applicable `AGENTS.md`/`CLAUDE.md` instructions into each agent-loop request prefix with a 65,536-byte render budget, and opts into first-message model titles. The headless surface retains deterministic fallback titles without making the auxiliary title-model request.
+The Web and headless surfaces boot one shared composition (`cordis.yml`): both treat the invoking directory as the default project and Workspace root, create named Workspaces beneath that root unless `--workspace-root ` overrides it, load applicable `AGENTS.md`/`CLAUDE.md` instructions into each agent-loop request prefix with a 65,536-byte render budget, and opt into first-message model titles. Headless differs only in listening on an OS-assigned port (parallel `dsh -p` runs never collide; the stderr-printed URL opens the live session in a browser). Both need the frontend dist and client bundles built (`pnpm run build && pnpm run build:web`).
## Install (developer machine)
@@ -20,4 +22,4 @@ Symlink the source-running launcher onto your PATH; it resolves the checkout thr
ln -sf "$(pwd)/bin/dsh" ~/.local/bin/dsh
```
-`pnpm run demo:tui` runs the same entry from the repo root. The built form (`lib/bin.js`, via `pnpm run build`) boots the same config under plain Node.
+`pnpm run dsh` runs the same entry from the repo root and forwards arguments directly, for example `pnpm run dsh -p "task"`. The built form (`lib/bin.js`, via `pnpm run build`) boots the same config under plain Node.
diff --git a/apps/cli/cordis.yml b/apps/cli/cordis.yml
new file mode 100644
index 0000000000..b89df77c46
--- /dev/null
+++ b/apps/cli/cordis.yml
@@ -0,0 +1,243 @@
+# dsh web — the full web-shape composition: host runtime (layer 1), the
+# transport/service layer (layer 2), and the browser plugin roster (dshClient
+# rows the modules node half scans into window.__DSH_BOOT__). Row order
+# carries no load semantics (activation is service-availability driven); the
+# grouping below is for readers. `--dev` appends the dsh-client-hmr row in
+# code (AppCLIEntry) — prod and dev differ by exactly that one row.
+# AppCLIEntry patches this tree before boot: profile json + CLI flags +
+# distIndex land as config patches over the rows below (yaml = engineering
+# defaults, json = user config, user wins per field).
+
+# ── layer 1: runtime ────────────────────────────────────────────────────────
+
+- id: timer
+ name: '@cordisjs/plugin-timer'
+
+- id: llm
+ name: '@deepseek-ai/dsh-llm'
+
+- id: session
+ name: '@deepseek-ai/dsh-session'
+
+- id: session-title
+ name: '@deepseek-ai/dsh-session-title'
+ config:
+ fallbackMaxWords: 5
+ fallbackMaxBytes: 40
+ maxTitleBytes: 80
+
+# Model-made titles on the first-message cadence (the web sidebar renders
+# session/title). Same values as the TUI composition.
+- id: session-title-llm
+ name: '@deepseek-ai/dsh-session-title-first-message-llm'
+ config:
+ targetWords: 5
+ targetCjkCharacters: 10
+ maxInputBytes: 4096
+ maxOutputTokens: 64
+ timeoutMs: 60000
+
+- id: system-prompt
+ name: '@deepseek-ai/dsh-system-prompt'
+ config:
+ persona: ''
+
+- id: tools
+ name: '@deepseek-ai/dsh-tools'
+
+- id: user-interaction
+ name: '@deepseek-ai/dsh-user-interaction'
+
+- id: agent
+ name: '@deepseek-ai/dsh-agent'
+
+- id: tasks
+ name: '@deepseek-ai/dsh-tasks'
+
+- id: agent-loop
+ name: '@deepseek-ai/dsh-agent-loop'
+ config:
+ agents: []
+
+# The native DeepSeek adapter; reads the key/base-url the boot's layered
+# .env loading (cwd then $DSH_HOME) left in the environment.
+- id: llm-deepseek
+ name: '@deepseek-ai/dsh-llm-deepseek'
+ config:
+ apiKey: !!js process.env.DEEPSEEK_API_KEY
+ baseURL: !!js process.env.DEEPSEEK_BASE_URL
+
+- id: session-persistence-jsonl
+ name: '@deepseek-ai/dsh-session-persistence-jsonl'
+ config:
+ root: './.sessions'
+
+- id: storage
+ name: '@deepseek-ai/dsh-storage'
+
+- id: storage-json
+ name: '@deepseek-ai/dsh-storage-json'
+ config:
+ root: './.storages'
+
+- id: storage-domain
+ name: '@deepseek-ai/dsh-storage-domain'
+ config:
+ backend: json
+
+- id: workspace
+ name: '@deepseek-ai/dsh-workspace'
+
+- id: bash-local
+ name: '@deepseek-ai/dsh-bash-local'
+
+- id: tool-bash
+ name: '@deepseek-ai/dsh-tool-bash'
+
+- id: tool-todo
+ name: '@deepseek-ai/dsh-tool-todo'
+
+- id: tool-tasks
+ name: '@deepseek-ai/dsh-tool-tasks'
+
+# fs cwd stays the package default (process.cwd()) — the same value the
+# gateway injects into session.cwd, so paths and sessions agree.
+- id: fs-local
+ name: '@deepseek-ai/dsh-fs-local'
+
+- id: fs-policy
+ name: '@deepseek-ai/dsh-fs-policy'
+
+- id: tool-fs
+ name: '@deepseek-ai/dsh-tool-fs'
+
+- id: tool-fs-search
+ name: '@deepseek-ai/dsh-tool-fs-search'
+
+- id: workspace-context
+ name: '@deepseek-ai/dsh-workspace-context'
+ config:
+ maxBytes: 65536
+
+- id: skill
+ name: '@deepseek-ai/dsh-skill'
+
+- id: skill-local
+ name: '@deepseek-ai/dsh-skill-local'
+
+- id: tool-skill
+ name: '@deepseek-ai/dsh-tool-skill'
+
+# token-meter rejects unknown config keys — keep this row bare.
+- id: token-meter
+ name: '@deepseek-ai/dsh-token-meter'
+
+- id: compact-basic
+ name: '@deepseek-ai/dsh-compact-basic'
+
+- id: subagent
+ name: '@deepseek-ai/dsh-subagent'
+
+- id: subagent-spawn
+ name: '@deepseek-ai/dsh-subagent-spawn'
+ config:
+ providerName: spawn
+
+- id: subagent-fork
+ name: '@deepseek-ai/dsh-subagent-fork'
+ config:
+ providerName: fork
+
+- id: tool-subagent
+ name: '@deepseek-ai/dsh-tool-subagent'
+ config:
+ provider: spawn
+ toolName: subagent
+
+- id: tool-subagent-fork
+ name: '@deepseek-ai/dsh-tool-subagent'
+ config:
+ provider: fork
+ toolName: subagent_fork
+
+- id: workflow-workerthread
+ name: '@deepseek-ai/dsh-workflow-workerthread'
+ config:
+ provider: spawn
+
+- id: tool-workflow
+ name: '@deepseek-ai/dsh-tool-workflow'
+
+- id: timeout-policy
+ name: '@deepseek-ai/dsh-timeout-policy'
+
+- id: spill-local
+ name: '@deepseek-ai/dsh-spill-local'
+
+# Omitting maxInlineBytes makes the whole policy a silent no-op — always
+# state it explicitly.
+- id: spill-policy
+ name: '@deepseek-ai/dsh-spill-policy'
+ config:
+ maxInlineBytes: 50000
+
+# The API gateway: the transport-agnostic dispatch face every client shape
+# shares. provider/model are the host default routing — the profile json's
+# mapping target (user config overrides these engineering defaults).
+- id: api-gateway
+ name: '@deepseek-ai/dsh-host-apiproxy'
+ config:
+ provider: deepseek
+ model: deepseek-v4-flash
+
+# ── layer 2: transport/service ──────────────────────────────────────────────
+
+# Plain route-registration carrier. distIndex is an assembly fact, not user
+# config — AppCLIEntry resolves the frontend dist and patches it in; host and
+# port arrive as CLI-flag patches over these defaults.
+- id: webserver
+ name: '@deepseek-ai/dsh-host-webserver'
+ config:
+ host: 127.0.0.1
+ port: 3080
+
+# ── browser plugin roster (dshClient rows; node halves are layer-2 hosts) ──
+
+# Dual-face: node half scans this very tree for dshClient rows, composes
+# window.__DSH_BOOT__, serves /plugins//client.js; browser half is the
+# module table the shell kernel constructs before cordis exists (§4.7 —
+# adopted as a plugin entry by the kernel, never fetched).
+- id: modules
+ name: '@deepseek-ai/dsh-client-modules'
+
+# Owns both ends of the web transport: node half binds the gateway to the
+# webserver under /api; browser half is the fetch/SSE client.
+- id: connection
+ name: '@deepseek-ai/dsh-client-connection'
+
+- id: client-runtime
+ name: '@deepseek-ai/dsh-client-runtime'
+
+- id: ui-theme
+ name: '@deepseek-ai/dsh-client-ui-theme'
+
+- id: i18n
+ name: '@deepseek-ai/dsh-client-i18n'
+
+- id: ui-layout
+ name: '@deepseek-ai/dsh-client-ui-layout'
+
+- id: ui-sidebar
+ name: '@deepseek-ai/dsh-client-ui-sidebar'
+
+- id: ui-conversation
+ name: '@deepseek-ai/dsh-client-ui-conversation'
+
+- id: ui-workspace
+ name: '@deepseek-ai/dsh-client-ui-workspace'
+
+- id: ui-question
+ name: '@deepseek-ai/dsh-client-ui-question'
+
+- id: ui-trajectory
+ name: '@deepseek-ai/dsh-client-ui-trajectory'
diff --git a/apps/cli/package.json b/apps/cli/package.json
index 8557965d34..8799669e8d 100644
--- a/apps/cli/package.json
+++ b/apps/cli/package.json
@@ -9,14 +9,22 @@
},
"files": [
"lib/bin.js",
+ "cordis.yml",
"src"
],
"license": "BSD-3-Clause",
"dependencies": {
+ "@cordisjs/plugin-include": "workspace:*",
+ "@cordisjs/plugin-loader": "workspace:*",
+ "@cordisjs/plugin-timer": "workspace:*",
+ "@deepseek-ai/dsh-agent": "workspace:^",
+ "@deepseek-ai/dsh-agent-loop": "workspace:^",
"@deepseek-ai/dsh-app-boot": "workspace:^",
+ "@deepseek-ai/dsh-bash-local": "workspace:^",
"@deepseek-ai/dsh-client-connection": "workspace:^",
"@deepseek-ai/dsh-client-hmr": "workspace:^",
"@deepseek-ai/dsh-client-i18n": "workspace:^",
+ "@deepseek-ai/dsh-client-modules": "workspace:^",
"@deepseek-ai/dsh-client-runtime": "workspace:^",
"@deepseek-ai/dsh-client-ui-conversation": "workspace:^",
"@deepseek-ai/dsh-client-ui-layout": "workspace:^",
@@ -24,13 +32,53 @@
"@deepseek-ai/dsh-client-ui-sidebar": "workspace:^",
"@deepseek-ai/dsh-client-ui-theme": "workspace:^",
"@deepseek-ai/dsh-client-ui-trajectory": "workspace:^",
+ "@deepseek-ai/dsh-client-ui-workspace": "workspace:^",
+ "@deepseek-ai/dsh-compact-basic": "workspace:^",
"@deepseek-ai/dsh-frontend": "workspace:^",
+ "@deepseek-ai/dsh-fs-local": "workspace:^",
+ "@deepseek-ai/dsh-fs-policy": "workspace:^",
"@deepseek-ai/dsh-host-apiproxy": "workspace:^",
- "@deepseek-ai/dsh-host-runtime": "workspace:^",
"@deepseek-ai/dsh-host-webserver": "workspace:^",
+ "@deepseek-ai/dsh-llm": "workspace:^",
+ "@deepseek-ai/dsh-llm-deepseek": "workspace:^",
"@deepseek-ai/dsh-paths": "workspace:^",
"@deepseek-ai/dsh-session": "workspace:^",
+ "@deepseek-ai/dsh-session-persistence-jsonl": "workspace:^",
+ "@deepseek-ai/dsh-session-title": "workspace:^",
+ "@deepseek-ai/dsh-session-title-first-message-llm": "workspace:^",
+ "@deepseek-ai/dsh-skill": "workspace:^",
+ "@deepseek-ai/dsh-skill-local": "workspace:^",
+ "@deepseek-ai/dsh-spill-local": "workspace:^",
+ "@deepseek-ai/dsh-spill-policy": "workspace:^",
+ "@deepseek-ai/dsh-storage": "workspace:^",
+ "@deepseek-ai/dsh-storage-domain": "workspace:^",
+ "@deepseek-ai/dsh-storage-json": "workspace:^",
+ "@deepseek-ai/dsh-subagent": "workspace:^",
+ "@deepseek-ai/dsh-subagent-fork": "workspace:^",
+ "@deepseek-ai/dsh-subagent-spawn": "workspace:^",
+ "@deepseek-ai/dsh-system-prompt": "workspace:^",
+ "@deepseek-ai/dsh-tasks": "workspace:^",
+ "@deepseek-ai/dsh-timeout-policy": "workspace:^",
+ "@deepseek-ai/dsh-token-meter": "workspace:^",
+ "@deepseek-ai/dsh-tool-bash": "workspace:^",
+ "@deepseek-ai/dsh-tool-fs": "workspace:^",
+ "@deepseek-ai/dsh-tool-fs-search": "workspace:^",
+ "@deepseek-ai/dsh-tool-skill": "workspace:^",
+ "@deepseek-ai/dsh-tool-subagent": "workspace:^",
+ "@deepseek-ai/dsh-tool-tasks": "workspace:^",
+ "@deepseek-ai/dsh-tool-todo": "workspace:^",
+ "@deepseek-ai/dsh-tool-workflow": "workspace:^",
+ "@deepseek-ai/dsh-tools": "workspace:^",
"@deepseek-ai/dsh-tui": "workspace:^",
- "cordis": "^4.0.0-rc.7"
+ "@deepseek-ai/dsh-user-interaction": "workspace:^",
+ "@deepseek-ai/dsh-workflow-workerthread": "workspace:^",
+ "@deepseek-ai/dsh-workspace": "workspace:^",
+ "@deepseek-ai/dsh-workspace-context": "workspace:^",
+ "commander": "^15.0.0",
+ "cordis": "^4.0.0-rc.7",
+ "js-yaml": "^4.2.0"
+ },
+ "devDependencies": {
+ "@types/js-yaml": "^4.0.9"
}
}
diff --git a/apps/cli/src/app-cli-entry.ts b/apps/cli/src/app-cli-entry.ts
new file mode 100644
index 0000000000..29e20b87b8
--- /dev/null
+++ b/apps/cli/src/app-cli-entry.ts
@@ -0,0 +1,240 @@
+/**
+ * AppCLIEntry — the pre-cordis boot glue the config-tree dsh surfaces share
+ * (`dsh web` and `dsh -p` boot the one composition; TUI migrates later).
+ * Everything here is what must exist before the Loader runs: layered env,
+ * the patch composition over the shipped cordis.yml (profile json + CLI
+ * flags + the resolved frontend dist), and the fail-loud triple after the
+ * tree settles.
+ */
+
+import { readFileSync } from 'node:fs'
+import { createRequire } from 'node:module'
+import { join, resolve } from 'node:path'
+import { pathToFileURL } from 'node:url'
+import { Context } from 'cordis'
+import type { FiberState } from 'cordis'
+import Loader from '@cordisjs/plugin-loader'
+import Include, { type PatchOptions } from '@cordisjs/plugin-include'
+import yaml from 'js-yaml'
+import { assertEntriesLoaded, installFailLoud, loadEnv } from '@deepseek-ai/dsh-app-boot'
+import { resolveDshHome } from '@deepseek-ai/dsh-paths'
+// Empty type import carries the httpServer Context merge for the port read below.
+import type {} from '@deepseek-ai/dsh-host-webserver'
+
+/** Profile file under the invoking directory (read-only this round; never created — see the design's profile ruling). */
+const PROFILE_DIR = '.dsh-tmp-profile'
+const PROFILE_FILE = 'config.json'
+
+/** One profile-json key mapped onto a yml row's config field. */
+interface ProfileMapping {
+ jsonPath: string
+ entryId: string
+ configKey: string
+}
+
+/**
+ * The static profile→row mapping table. json is user config and wins over the
+ * yml engineering default per field; a json key absent from this table fails
+ * loud (a typo silently ignored would read as "setting has no effect").
+ * Developers extend deployments by adding rows here.
+ */
+const PROFILE_MAPPINGS: ProfileMapping[] = [
+ { jsonPath: 'provider', entryId: 'api-gateway', configKey: 'provider' },
+ { jsonPath: 'model', entryId: 'api-gateway', configKey: 'model' },
+ { jsonPath: 'persistenceRoot', entryId: 'session-persistence-jsonl', configKey: 'root' },
+]
+
+// The include's YAML dialect: `!!js` scalars become expression nodes the
+// Loader evaluates at entry activation. The bypass parse below must accept
+// them (and passing one through a patch unchanged is legal).
+const jsExprType = new yaml.Type('tag:yaml.org,2002:js', {
+ kind: 'scalar',
+ resolve: data => typeof data === 'string',
+ construct: data => ({ __jsExpr: String(data) }),
+})
+const includeYamlSchema = yaml.JSON_SCHEMA.extend(jsExprType)
+
+/**
+ * Value mirror of cordis's `FiberState` const enum members the sweep needs
+ * (a const enum has no runtime object to import; same rationale as the
+ * client-side mirror in dsh-client-web).
+ */
+const FIBER_ACTIVE = 2 as FiberState.ACTIVE
+const FIBER_PENDING = 0 as FiberState.PENDING
+
+/** Constructor facts for one dsh invocation over the shared composition (argv already parsed by the surface bin). */
+export interface AppCLIEntryOptions {
+ /** Absolute path of the shipped cordis.yml. */
+ configPath: string
+ /** Whether to append the HMR row (the whole prod/dev difference; web surface only). */
+ dev: boolean
+ /** --host when explicitly passed; undefined keeps the yml engineering default. */
+ host?: string
+ /**
+ * Listen port override onto the webserver row. Web passes the --port flag
+ * value; headless passes 0 (an OS-assigned port, so parallel `dsh -p` runs
+ * never collide — and the printed URL still opens the live session in a
+ * browser).
+ */
+ port?: number
+ /** Parent directory for name-created Workspaces; undefined uses the gateway's cwd fallback. */
+ workspaceRoot?: string
+}
+
+/**
+ * Boot driver for the config-tree dsh surfaces (web and headless share the
+ * one composition; the surfaces differ only in constructor facts): holds only
+ * what exists independently of (and prior to) cordis — argv facts, the
+ * composed patch set, and finally the root ctx.
+ */
+export class AppCLIEntry {
+ /** The root context, set by {@link run}. */
+ ctx!: Context
+
+ private patches: PatchOptions[] = []
+
+ constructor(private readonly options: AppCLIEntryOptions) {}
+
+ /**
+ * Run the boot chain: layered env → patch composition → Loader include
+ * boot (dev row before await) → fail-loud triple.
+ * @returns the settled root context and the listening port.
+ */
+ async run(): Promise<{ ctx: Context; port: number }> {
+ this.loadEnvLayers()
+ this.composePatches()
+ await this.bootTree()
+ this.assertBoot()
+ const port = this.ctx.get('httpServer')?.port
+ /* v8 ignore next -- the sweep above guarantees an ACTIVE webserver row */
+ if (port === undefined) throw new Error('dsh: httpServer service missing after settled boot')
+ return { ctx: this.ctx, port }
+ }
+
+ /** Layered .env: ambient > cwd (bin already loaded) > $DSH_HOME (loadEnvFile never overrides). */
+ private loadEnvLayers(): void {
+ loadEnv('dsh', resolveDshHome())
+ }
+
+ /**
+ * Compose the patch set from the three non-yml config sources: profile
+ * json (user config), CLI flags, and the resolved frontend dist. Patches
+ * replace a row's config wholesale, so each patched row's yml static
+ * values are re-read here (bypass parse) and merged under the overrides.
+ */
+ private composePatches(): void {
+ const rows = this.parseYmlRows()
+ const overrides = new Map>()
+ const put = (entryId: string, key: string, value: unknown): void => {
+ const bag = overrides.get(entryId) ?? {}
+ bag[key] = value
+ overrides.set(entryId, bag)
+ }
+
+ // Source 1: profile json (missing file = empty; unmapped key = loud).
+ for (const [key, value] of Object.entries(this.readProfile())) {
+ const mapping = PROFILE_MAPPINGS.find(m => m.jsonPath === key)
+ if (mapping === undefined) {
+ throw new Error(`dsh: profile key "${key}" has no mapping (known: ${PROFILE_MAPPINGS.map(m => m.jsonPath).join(', ')})`)
+ }
+ put(mapping.entryId, mapping.configKey, value)
+ }
+
+ // Source 2: CLI flags (field set disjoint from the json mappings).
+ if (this.options.host !== undefined) put('webserver', 'host', this.options.host)
+ if (this.options.port !== undefined) put('webserver', 'port', this.options.port)
+ if (this.options.workspaceRoot !== undefined) put('api-gateway', 'workspaceRoot', this.options.workspaceRoot)
+
+ // Source 3: the frontend dist — an assembly fact of this app, never yml
+ // user config. Workspace knowledge stays here.
+ put('webserver', 'distIndex', this.resolveDistIndex())
+
+ this.patches = [...overrides.entries()].map(([id, bag]) => {
+ const yml = rows.get(id)
+ if (yml === undefined) throw new Error(`dsh: patch target row "${id}" not found in ${this.options.configPath}`)
+ return { id, config: { ...(yml.config ?? {}) as Record, ...bag } }
+ })
+ }
+
+ /** Loader include boot; the dev HMR row mounts before await so the fail-loud triple covers it. */
+ private async bootTree(): Promise {
+ const ctx = new Context()
+ ctx.baseUrl = pathToFileURL(join(resolve(this.options.configPath), '..')).href + '/'
+ await ctx.plugin(Loader)
+ ctx.loader.builtins.include = Include
+ await ctx.loader.create({
+ name: 'cordis:include',
+ config: {
+ path: pathToFileURL(resolve(this.options.configPath)).href,
+ ...this.patches.length > 0 ? { patches: this.patches } : {},
+ },
+ })
+ if (this.options.dev) {
+ await ctx.loader.create({ name: '@deepseek-ai/dsh-client-hmr' })
+ }
+ this.ctx = ctx
+ await ctx.loader.await()
+ }
+
+ /**
+ * Fail-loud triple: assertEntriesLoaded catches import failures,
+ * installFailLoud catches late apply rejections, and the all-ACTIVE sweep
+ * below catches PENDING fibers (cordis inject waiting has no timeout).
+ */
+ private assertBoot(): void {
+ installFailLoud('dsh')
+ assertEntriesLoaded(this.ctx, 'dsh')
+ const failures: string[] = []
+ for (const entry of this.ctx.loader.entries()) {
+ if (entry.fiber === undefined || entry.disabled) continue
+ const state = entry.fiber.state
+ if (state === FIBER_ACTIVE) continue
+ if (state === FIBER_PENDING) {
+ const missing = Object.keys(entry.fiber.inject).filter(service => this.ctx.get(service) === undefined)
+ failures.push(`${entry.options.name}: pending (waiting for service${missing.length === 1 ? '' : 's'}: ${missing.join(', ') || 'unknown'})`)
+ } else {
+ failures.push(`${entry.options.name}: fiber state ${String(state)}`)
+ }
+ }
+ if (failures.length > 0) {
+ throw new Error(`dsh: ${String(failures.length)} entr${failures.length === 1 ? 'y' : 'ies'} did not activate\n${failures.join('\n')}`)
+ }
+ }
+
+ /** Bypass parse of the shipped yml (id → row) for patch-merge inputs; Loader still reads the file itself. */
+ private parseYmlRows(): Map {
+ const doc = yaml.load(readFileSync(this.options.configPath, 'utf8'), { schema: includeYamlSchema })
+ if (!Array.isArray(doc)) throw new Error(`dsh: ${this.options.configPath} is not a top-level entry list`)
+ const rows = new Map()
+ for (const row of doc as { id?: string; config?: unknown }[]) {
+ if (typeof row.id === 'string') rows.set(row.id, row)
+ }
+ return rows
+ }
+
+ /** Profile json under cwd; read-only — never created here, absent = no user config. */
+ private readProfile(): Record {
+ let raw: string
+ try {
+ raw = readFileSync(join(process.cwd(), PROFILE_DIR, PROFILE_FILE), 'utf8')
+ } catch (error) {
+ if ((error as NodeJS.ErrnoException).code === 'ENOENT') return {}
+ throw error
+ }
+ const parsed: unknown = JSON.parse(raw)
+ if (parsed === null || typeof parsed !== 'object' || Array.isArray(parsed)) {
+ throw new Error(`dsh: ${PROFILE_DIR}/${PROFILE_FILE} must hold a JSON object`)
+ }
+ return parsed as Record
+ }
+
+ /** Dist location is workspace knowledge of this app: resolved through the frontend package exports, not configured. */
+ private resolveDistIndex(): string {
+ const require = createRequire(import.meta.url)
+ try {
+ return require.resolve('@deepseek-ai/dsh-frontend/dist/index.html')
+ } catch {
+ throw new Error('dsh: frontend dist not built; run pnpm --filter @deepseek-ai/dsh-frontend build first')
+ }
+ }
+}
diff --git a/apps/cli/src/args.ts b/apps/cli/src/args.ts
new file mode 100644
index 0000000000..9fd0f4d9bf
--- /dev/null
+++ b/apps/cli/src/args.ts
@@ -0,0 +1,143 @@
+/**
+ * Commander adapter for the `dsh` command-line entry: the one place argv is
+ * parsed and routed to a mode. `bin.ts` switches on the returned discriminant
+ * and dynamic-imports that mode's module. One program: the default (no
+ * subcommand) is the TUI/headless surface with option-only flags; `web` is a
+ * real subcommand. Commander owns `--help`/`--version` and parse errors — it
+ * prints and exits at the point of failure (a domain failure routes through
+ * `command.error`), so this returns only a resolved mode.
+ * @module @deepseek-ai/dsh/args
+ */
+
+import { Command, CommanderError } from 'commander'
+
+/** Interactive TUI: the default mode. `--config` swaps the tree; `--resume ` rehydrates a session. */
+interface TuiInvocation {
+ mode: 'tui'
+ config?: string
+ resume?: string
+}
+
+/** Headless one-shot: `dsh -p "task"`. */
+interface HeadlessInvocation {
+ mode: 'headless'
+ prompt: string
+}
+
+/**
+ * Browser UI: `dsh web`. `host`/`port` are present only when the flag was
+ * passed — pass-through overrides with no CLI default and no CLI validation:
+ * the `dsh-host-webserver` schema (`host` a loopback/all-interfaces literal,
+ * `port` a natural ≤ 65535) is the single source of both the default (the
+ * shipped `cordis.yml` value stands when a flag is absent) and validity (a bad
+ * value fails loud at boot). `port` is `Number`-coerced only because the schema
+ * wants a number, not a string. `dev` mounts the client HMR driver;
+ * `workspaceRoot` is the parent directory for name-created workspaces.
+ */
+interface WebInvocation {
+ mode: 'web'
+ host?: string
+ port?: number
+ dev: boolean
+ workspaceRoot?: string
+}
+
+/** The resolved `dsh` invocation: exactly one mode. `--help`/`--version`/errors exit inside {@link parseDshArgs}. */
+export type DshInvocation = TuiInvocation | HeadlessInvocation | WebInvocation
+
+/** Raw web-subcommand options straight from Commander. */
+interface WebOptions {
+ host?: string
+ port?: string
+ dev?: boolean
+ workspaceRoot?: string
+}
+
+/**
+ * Narrow the raw `web` options into a {@link WebInvocation}. No host/port
+ * validation: both flow to the webserver schema, which is the sole gate. `port`
+ * is coerced to a number (the schema rejects a string) but not range-checked
+ * here — `NaN`/out-of-range fail loud at the schema on boot.
+ */
+function resolveWeb(options: WebOptions): WebInvocation {
+ return {
+ mode: 'web',
+ ...options.host !== undefined && { host: options.host },
+ ...options.port !== undefined && { port: Number(options.port) },
+ dev: options.dev === true,
+ ...options.workspaceRoot !== undefined && { workspaceRoot: options.workspaceRoot },
+ }
+}
+
+/**
+ * Resolve the raw argv into a {@link DshInvocation}, or print and exit for
+ * `--help`/`--version`/a parse error. The default (no subcommand) is the
+ * TUI/headless surface; `web` is a subcommand.
+ * @param argv - the arguments after the node binary and script (`process.argv.slice(2)`).
+ * @param version - the version string `--version` prints; read from this app's package.json.
+ * @returns the resolved invocation (only reached on a valid, non-help invocation).
+ */
+export function parseDshArgs(argv: readonly string[], version: string): DshInvocation {
+ let resolved: DshInvocation | undefined
+ const program = new Command()
+ .name('dsh')
+ .version(version, '-V, --version', 'output the version number')
+ .description('dsh: interactive TUI (default), headless task, and browser UI')
+ .exitOverride()
+ // Default surface: option-only (no positional), so `web` can be a real
+ // subcommand without a positional collision.
+ .option('--config ', 'boot an alternate cordis.yml instead of the shipped tree (TUI mode)')
+ .option('-p, --prompt ', 'run one headless turn for this task, print the result, and exit')
+ .option('--resume ', 'resume the persisted session with this id (TUI mode)')
+ .action((options: { config?: string; prompt?: string; resume?: string }) => {
+ if (options.prompt !== undefined) {
+ // A headless prompt owns the invocation; an empty task has nothing to
+ // run, and --config/--resume are TUI inputs that must not silently
+ // vanish from a headless run.
+ if (options.prompt === '') program.error('error: --prompt needs a task')
+ if (options.config !== undefined || options.resume !== undefined) {
+ program.error('error: --prompt takes no --config or --resume')
+ }
+ resolved = { mode: 'headless', prompt: options.prompt }
+ return
+ }
+ // An empty --resume= id would silently start a fresh session downstream
+ // (agent-loop treats '' as no-resume), so a mistyped resume must fail loud.
+ if (options.resume === '') program.error('error: --resume needs a session id')
+ resolved = {
+ mode: 'tui',
+ ...options.config !== undefined && { config: options.config },
+ ...options.resume !== undefined && { resume: options.resume },
+ }
+ })
+
+ const web = program.command('web').description('serve the browser UI (host/port default to the shipped config)')
+ web
+ .option('--host