Merge remote-tracking branch 'origin/master' into worktree/web-multimodal-image-input
# Conflicts: # .agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.i18n.yaml # .agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md # .agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.zh.md # .agents/notes/implemented/architecture/2026-07-25-web-input-machine-and-slash-pipeline.i18n.yaml # .agents/notes/implemented/architecture/2026-07-25-web-input-machine-and-slash-pipeline.md # .agents/notes/implemented/architecture/2026-07-25-web-input-machine-and-slash-pipeline.zh.md # THIRD_PARTY_NOTICES.md # apps/cli/composition.md # apps/cli/config/base.cordis.yml # apps/cli/package.json # apps/cli/src/app-cli-entry.ts # apps/cli/src/bin.ts # apps/cli/tests/args.spec.ts # apps/web/tests/built-boot.snapshot.ts # apps/web/tests/navigation-panes.e2e.ts # docs/architecture.i18n.yaml # docs/architecture.md # docs/architecture.zh.md # docs/config-catalog.md # docs/cordis-catalog/services.md # docs/core-data-structures/core.i18n.yaml # docs/core-data-structures/llm-streaming.i18n.yaml # docs/event-producer-consumer.md # docs/module-graph.md # examples/acp-agent/tests/snapshots/cordis-inspect-jsdoc/session.jsonl # packages/README.i18n.yaml # packages/bundle/README.i18n.yaml # packages/client/connection/README.i18n.yaml # packages/client/connection/README.md # packages/client/connection/README.zh.md # packages/client/connection/src/client/fixture.ts # packages/client/connection/src/http-bridge.ts # packages/client/connection/src/index.ts # packages/client/connection/tests/fixture.spec.ts # packages/client/connection/tests/node-half.spec.ts # packages/client/runtime/README.i18n.yaml # packages/client/runtime/README.md # packages/client/runtime/README.zh.md # packages/client/runtime/src/client/contract/session.ts # packages/client/runtime/src/client/sessions/session.ts # packages/client/ui-conversation/README.i18n.yaml # packages/client/ui-conversation/README.md # packages/client/ui-conversation/README.zh.md # packages/client/ui-conversation/src/client/apply.ts # packages/client/ui-conversation/src/client/chat/AssistantMarkdown.tsx # packages/client/ui-conversation/src/client/chat/ChatView.tsx # packages/client/ui-conversation/src/client/chat/MessageItem.module.css # packages/client/ui-conversation/src/client/chat/MessageItem.tsx # packages/client/ui-conversation/src/client/contract/slots.ts # packages/client/ui-conversation/src/client/index.ts # packages/client/ui-conversation/src/client/input/contract.ts # packages/client/ui-conversation/src/client/input/facade.ts # packages/client/ui-conversation/src/client/input/hub.ts # packages/client/ui-conversation/src/client/locales.ts # packages/client/ui-conversation/src/client/service.ts # packages/client/ui-conversation/src/client/skeleton/ConversationSession.tsx # packages/client/ui-conversation/src/client/skeleton/InputBar.tsx # packages/client/ui-conversation/tests/apply-inject.spec.tsx # packages/client/ui-conversation/tests/input-bar.spec.tsx # packages/client/ui-conversation/tests/input-matrix.spec.tsx # packages/client/ui-conversation/tests/input-scenarios.spec.tsx # packages/client/ui-conversation/tests/service-orchestration.spec.ts # packages/client/ui-conversation/tests/skeleton.spec.tsx # packages/client/ui-trajectory/tests/views.spec.tsx # packages/compact/compact-basic/README.i18n.yaml # packages/cordis/tool-cordis/src/api-catalog.ts # packages/host/apiproxy/README.i18n.yaml # packages/host/apiproxy/README.md # packages/host/apiproxy/README.zh.md # packages/host/apiproxy/src/api-proxy.ts # packages/host/apiproxy/src/api/rpc.ts # packages/host/apiproxy/src/api/sessions.ts # packages/host/apiproxy/src/index.ts # packages/host/apiproxy/tests/api-proxy-models.spec.ts # packages/host/apiproxy/tests/rpc-schemas.spec.ts # packages/llm/llm-pi-ai/README.i18n.yaml # packages/llm/llm-pi-ai/README.md # packages/llm/llm-pi-ai/README.zh.md # packages/llm/llm-pi-ai/src/adapter.ts # packages/llm/llm/README.i18n.yaml # packages/ui/tui/README.md # packages/ui/tui/README.zh.md # packages/ui/tui/src/components/content.ts # packages/ui/tui/src/components/transcript.ts # packages/ui/tui/tests/tui.spec.ts # pnpm-lock.yaml
This commit is contained in:
Yichen Jiang
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-19-drop-mutable-session-summary.md
|
||||
2026-06-19-drop-mutable-session-summary.md: f87378a1c3737950eb536be8e3f6776586eb8a01
|
||||
2026-06-19-drop-mutable-session-summary.zh.md: 05b6711b71ef87602b46706cd4340a10453e66ad
|
||||
2026-06-19-drop-mutable-session-summary.zh.md: 41dab3fa0b487b057d33335a21cee92cf211560f
|
||||
|
||||
@@ -6,14 +6,14 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
[会话持久化 seam](../architecture/2026-06-14-session-persistence.md)将会话的日志外元数据拆分为 `dsh-session` 拥有的两种类型：一个不可变的 `SessionHeader`（`version`、`id`、`createdAt`、`cwd?`、`parentSession?`），在创建时一次性写入；一个可变的 `SessionSummary`（`updatedAt`、`title?`、`firstPrompt?`），「可在不触碰仅追加日志的情况下更新」。二者的联合类型为 `SessionMeta = SessionHeader & SessionSummary`，抽象的 `SessionPersistence` 服务为此多出第七个方法 `update(id, summary)`，用于重写摘要。各后端各自实现可变存储：JSONL 在日志旁写一个独立的原子 `.summary.json` **伴随文件**（临时写入 + rename，尽力保证）；SQLite 在追加事务内更新 `updated_at`/`title`/`first_prompt` **列**。
|
||||
[会话持久化 seam](../architecture/2026-06-14-session-persistence.md)将会话的日志外元数据拆分为 `dsh-session` 拥有的两种类型：一个不可变的 `SessionHeader`（`version`、`id`、`createdAt`、`cwd?`、`parentSession?`），在创建时一次性写入；一个可变的 `SessionSummary`（`updatedAt`、`title?`、`firstPrompt?`），「可在不触碰仅追加日志的情况下更新」。二者合并为 `SessionMeta = SessionHeader & SessionSummary`，抽象的 `SessionPersistence` 服务为此多出第七个方法 `update(id, summary)`，用于重写摘要。各后端各自实现可变存储：JSONL 在日志旁通过临时写入后 rename，以尽力而为的方式原子发布一个独立的 `.summary.json` **伴随文件**；SQLite 则使用 `updated_at`/`title`/`first_prompt` **列**，并在追加事务内更新其中的时间列。
|
||||
|
||||
摘要是为未来的会话选择器设计的（通过 `updatedAt` 排序近期会话，用 `title`/`firstPrompt` 做预览）。该选择器从未实现。对整个仓库的审计表明，`SessionSummary` 的全部表面积都是**死状态**：
|
||||
摘要是为未来的会话选择器设计的（通过 `updatedAt` 排序近期会话，用 `title`/`firstPrompt` 做预览）。该选择器从未实现。对整个仓库的审计表明，`SessionSummary` 的整套相关接口都只是在维护**无用状态**：
|
||||
|
||||
- `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**。
|
||||
- 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。
|
||||
- 持久化列表的消费方使用不可变 header 中的标识、创建时间、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
|
||||
@@ -22,14 +22,14 @@ Status: implemented
|
||||
|
||||
摘要原本要提供的一切，在消费方真正需要时都**可从仅追加日志中派生**（`firstPrompt` = 第一条 `user/message`；近期度 = 最后一个事件的 `time` 或文件 mtime），或者已经存在于不可变 header 中（`createdAt`、`cwd`）。唯一*不可*派生的是用户*手动编辑*的标题，但它从未实现，纯属 YAGNI；如果未来真有功能需要，它可以作为独立的日志事件或 header 字段回归。
|
||||
|
||||
这被记录为一项决策，因为它具有**持久性**（它同时收窄两个后端的公共服务契约和磁盘格式）、**争议性**（summary 是有意为未来设计的结果，而非意外），也具有**意外性**（未来读者在原 Agent Note（agent 决策记录）描述 `SessionMeta` 的位置发现 `SessionHeader`，否则会追问 summary 为何消失）。它还为[共享持久化写入协调器](../architecture/2026-06-18-shared-persistence-write-coordinator.md)扫清障碍：不再有可变 summary 后，协调器的钩子接口不需要 `updateSummary` 钩子，JSONL sidecar 与 SQLite 列之间的持久性分歧也随之消失，使两个后端的写入路径趋于一致。
|
||||
这被记录为一项决策，因为它具有**持久性**（它同时收窄两个后端的公开服务契约和磁盘格式）、**争议性**（summary 是有意为未来设计的结果，而非意外），也具有**反直觉性**（未来读者在原 Agent Note 描述 `SessionMeta` 的位置发现 `SessionHeader` 时，若无此记录便会追问 summary 为何消失）。它还为[共享持久化写入协调器](../architecture/2026-06-18-shared-persistence-write-coordinator.md)扫清障碍：不再有可变 summary 后，协调器的钩子接口不需要 `updateSummary` 钩子，JSONL sidecar 与 SQLite 列之间的持久性分歧也随之消失，使两个后端的写入路径趋于一致。
|
||||
|
||||
## 无需迁移
|
||||
|
||||
这是未发布的软件（见[根 AGENTS.md](../../../../AGENTS.md)「Pre-release stance: foundation over blast radius」一节），因此没有需要保留的磁盘数据库或日志。SQLite 不迁移 v1 数据库：`openDatabase` 守卫现在拒绝任何非当前版本的磁盘 `user_version`（`onDisk !== 0 && onDisk !== SCHEMA_VERSION`），无论更旧*还是*更新，因此陈旧的 v1 数据库会被干净地拒绝，而非在新列集下被半读取。新建数据库写入当前版本号；这是唯一需要正常工作的路径。
|
||||
这是未发布的软件（见[根 AGENTS.md](../../../../AGENTS.md)「Pre-release stance: foundation over blast radius」一节），因此没有需要保留的磁盘数据库或日志。SQLite 不迁移 v1 数据库：`openDatabase` 守卫现在拒绝任何非当前版本的磁盘 `user_version`（`onDisk !== 0 && onDisk !== SCHEMA_VERSION`），无论版本更旧*还是*更高，因此陈旧的 v1 数据库会被干净地拒绝，而不会按新的列集合进行不完整读取。新建数据库写入当前版本号；这是唯一需要正常工作的路径。
|
||||
|
||||
## 后果
|
||||
|
||||
未来的会话选择器现在必须从日志派生预览/排序信息（或重新引入一个类型化字段），而不能直接读取现成的摘要行。这是正确的代价：为一个尚不存在的功能维护缓存，是每个后端都要付出维护成本、每个契约测试都要付出断言成本的死重。这一原则——**通过的测试固定的是当前行为，不一定是正确行为；行为可能是过去妥协的产物**——现已作为独立约定记录在[根 AGENTS.md](../../../../AGENTS.md) 中，本次变更即为其实例。
|
||||
未来的会话选择器现在必须从日志派生预览/排序信息（或重新引入一个类型化字段），而不能直接读取现成的摘要行。这是正确的代价：为一个尚不存在的功能维护缓存，是每个后端都要承担维护成本、每个契约测试都要承担断言成本的无谓负担。这一原则——**通过的测试固定的是当前行为，不一定是正确行为；行为可能是过去妥协的产物**——现已作为独立约定记录在[根 AGENTS.md](../../../../AGENTS.md) 中，本次变更即为其实例。
|
||||
|
||||
<!-- agent-note-format: alternatives-not-recorded (pre-format Agent Note) -->
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-20-collapse-trace-only-session-events.md
|
||||
2026-06-20-collapse-trace-only-session-events.md: c77062c3cd44286b43c175702c35b36f9cc31da6
|
||||
2026-06-20-collapse-trace-only-session-events.zh.md: b232b3fb60822e60b1f5767066db42b0228269a8
|
||||
2026-06-20-collapse-trace-only-session-events.zh.md: bc7e33f2d5dce370b846b1451fca6a652660dc92
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Agent Note: 将仅用于追踪的会话事实折叠进承载性事件
|
||||
# Agent Note: 将仅用于追踪的会话事实折叠进承载实际功能的事件
|
||||
|
||||
Status: implemented
|
||||
|
||||
@@ -8,11 +8,11 @@ Status: implemented
|
||||
|
||||
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。
|
||||
这些事件让规范的 transcript（文本记录）看起来比实际更适合作为遥测数据。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。
|
||||
|
||||
## 决策
|
||||
|
||||
仅在信息已被保留、无需并行记录的情况下，移除独立的追踪事件：
|
||||
仅在信息已被保留、无需并行记录的情况下，移除独立的、仅用于追踪的事件：
|
||||
|
||||
- 成功步骤的 usage 折叠进匹配的 `assistant/message`（`assistant/message { turn, step, content, usage? }`），使组装好的模型输出与其计费信息一同传递。
|
||||
- 失败或中止的步骤如果有 usage 但没有 assistant 内容，则将 usage 放在一个空内容的 `assistant/message` 上（下方实现说明给出了无信息丢失的证明）——不会有已持久化的 usage 分片无处安放。
|
||||
@@ -23,22 +23,22 @@ Status: implemented
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**保留独立行作为遥测**——这些事件让规范 transcript 看起来比实际更像遥测数据，代价是增加了事件变体、不变式、测试、快照和持久化用例，却没有任何消费方使用。如果分析需求真正出现，正确的形态是投影辅助工具或带有独立保留策略的专用遥测存储，而非对话日志中的重复追踪行。
|
||||
**保留独立行作为遥测**——这些事件让规范 transcript 看起来比实际更适合作为遥测数据，代价是增加了事件变体、不变式、测试、快照和持久化用例，却没有任何消费方使用。如果分析需求真正出现，正确的形态是投影辅助工具或带有独立保留策略的专用遥测存储，而非对话日志中的重复追踪行。
|
||||
|
||||
## 验证
|
||||
|
||||
`SessionEventMap` 不再包含独立的 `usage` 或 `error`；agent loop（智能体循环）不再追加独立的 usage 事件，持久性失败通过 `turn/end { kind: 'error', step, message, code? }` 记录；ACP 快照和持久化测试断言不存在仅追踪行；已录制的 fixture（测试前置数据）使用新事件形状，会话格式版本固定为 `0`（后端按预发布格式策略拒绝任何非 `0` 的存储日志）；文档说明了 token 用量和操作错误的观测位置。
|
||||
`SessionEventMap` 不再包含独立的 `usage` 或 `error`；agent loop（智能体循环）不再追加独立的 usage 事件，并通过 `turn/end { kind: 'error', step, message, code? }` 持久记录失败；ACP 快照和持久化测试断言不存在仅用于追踪的行；已录制的 fixture（测试前置数据）使用新事件形状，会话格式版本固定为 `0`（后端按预发布格式策略拒绝任何版本非 `0` 的已存储日志）；文档说明了 token 用量和运行时错误的观测位置。
|
||||
|
||||
## 后果
|
||||
|
||||
消费方不能再从规范日志中筛选独立的 `usage` 或步骤级 `error` 行，必须从承载它们的 assistant/failure 事件中读取这些事实。只有在实现 PR（Pull Request）证明相同事实仍然存在的前提下，这才是合理的简化；否则独立事件应予保留。
|
||||
消费方不能再从规范日志中筛选独立的 `usage` 或步骤级 `error` 行，而必须从承载这些信息的助手消息或失败事件中读取这些事实。只有在实现 PR（Pull Request）证明相同事实仍然存在的前提下，这才是合理的简化；否则独立事件应予保留。
|
||||
|
||||
## 实现说明
|
||||
|
||||
按提案落地，但有一处范围细化（遵循 AGENTS.md 所述“Agent Note（agent 决策记录）是提案，而非绝对真理”）：
|
||||
按提案落地，但有一处范围细化（遵循 AGENTS.md 所述「Agent Note 是提案，而非绝对真理」）：
|
||||
|
||||
- **空内容 `assistant/message` 承载 usage，无数据丢失。** 提案要求的证明（不会有已持久化的 usage 分片无处安放）落在 max-tokens 路径上：一个被截断的步骤有 usage 但内容为空（例如只有一个被丢弃的工具调用），以前会发出独立的 `usage`。现在它记录一个空内容的 `assistant/message { content: [], usage }`。为防止这向提供方 transcript 注入一个无内容的虚假 assistant 轮次，`deriveMessages()` 跳过空内容的 `assistant/message` 事件。回归测试断言 usage 仍被表示，且派生历史未被破坏。
|
||||
- **空内容 `assistant/message` 承载 usage，无数据丢失。** 提案要求的证明（不会有已持久化的 usage 分片无处安放）落在 max-tokens 路径上：一个被截断的步骤有 usage 但内容为空（例如只有一个被丢弃的工具调用），以前会发出独立的 `usage`。现在它记录一个空内容的 `assistant/message { content: [], usage }`。为防止这向提供方 transcript 注入一个多余的无内容 assistant 轮次，`deriveMessages()` 跳过空内容的 `assistant/message` 事件。回归测试断言 usage 仍有记录，且派生历史未被破坏。
|
||||
|
||||
**格式版本。** 此变更影响已持久化的事件，但预发布会话格式仍固定为 `0`，拒绝任何其他版本且不做迁移。`dsh-session` 拥有写入方和加载校验使用的常量。单调递增的格式版本从首次正式发布开始。
|
||||
|
||||
Usage 现在通过 `assistant/message.usage` 观测；操作错误的步骤编号通过 `turn/end.reason`（当 `kind: 'error'` 时）观测。`agent/error` 与日志用于实时诊断，保持不变。
|
||||
Usage 现在通过 `assistant/message.usage` 观测；运行时错误的步骤编号通过 `turn/end.reason`（当 `kind: 'error'` 时）观测。`agent/error` 与日志用于实时诊断，保持不变。
|
||||
|
||||
@@ -1,6 +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
|
||||
2026-06-20-public-agent-stop-surface.md: e22c4389df18f3c9ca96763fc097eabefcc5b761
|
||||
2026-06-20-public-agent-stop-surface.zh.md: e2647b498a8c906579b4fd2b50f94d1c326fe784
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-20-public-agent-stop-surface.md
|
||||
2026-06-20-public-agent-stop-surface.md: e9c713c7ba8c784d3d66fe8f391e1b197f7394d2
|
||||
2026-06-20-public-agent-stop-surface.zh.md: 5c1eb0ec8245c60613d6d4d7c1d105b5243bc938
|
||||
|
||||
@@ -8,15 +8,15 @@ English | [中文](2026-06-20-public-agent-stop-surface.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The public `Agent` handle exposed two overlapping ways to stop in-flight work: step-only `abort()` and queue-aware `cancel()`. The former preserved queued input while the latter clears queued and steering work and aborts the active turn. In production, ACP uses `cancel()` for `session/cancel`, while lifecycle owners tear down agents through `AgentHandle.dispose()`. No production caller needs a bare step-only abort.
|
||||
The public `Agent` handle exposed two overlapping ways to stop in-flight work: step-only `abort()` and queue-aware `cancel()`. The former preserved queued input while the latter originally only exposed its broad default, which clears queued and steering work while aborting the active turn. `cancel(cause, { keepInbox: true })` now covers the production Web stop policy without exposing the private turn holder; ACP retains broad cancellation, while lifecycle owners tear down agents through `AgentHandle.dispose()`. No production caller needs a bare step-only abort.
|
||||
|
||||
The behavioral distinction is real, but no shipping code needs the narrower operation. AgentLoop instead owns one private cancellation holder for the whole turn. `cancel(cause?)` carries a typed `user` or `parent` cause, defaults to `user`, and drops pending input; disposal remains a separate lifecycle interruption. The complete ownership and propagation contract lives in the [explicit turn cancellation RFC](../architecture/2026-07-16-explicit-turn-cancellation.md).
|
||||
The behavioral distinction is real, but no shipping code needs a separate narrower verb. AgentLoop owns one private cancellation holder for the whole turn. `cancel(cause, options?)` carries an explicit typed `user` or `parent` cause; its broad default drops pending input, while `keepInbox` preserves pending work for later turns. Disposal remains a separate lifecycle interruption. The complete ownership and propagation contract lives in the [explicit turn cancellation RFC](../architecture/2026-07-16-explicit-turn-cancellation.md).
|
||||
|
||||
The extra surface area made the loop carry a public verb that is mostly a teardown internal: `abort()` had to be documented as distinct from queue-aware cancellation even though a UI cancellation almost always wants the broader operation.
|
||||
The extra surface area made the loop carry a public verb that was mostly a teardown internal. An options-bearing `cancel()` expresses caller policy without exposing a second holder-shaped operation.
|
||||
|
||||
## Decision
|
||||
|
||||
`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.
|
||||
`cancel()` is the only public *stop* primitive on `Agent`. Lifecycle owners use `AgentHandle.dispose()` to stop and unregister an agent; non-owners use broad `cancel()` to abandon current and queued work or `keepInbox` to abort the active turn while retaining pending work. The implementation keeps a private turn cancellation holder, but it is not part of the plugin-facing `Agent` contract. The [Web stop decision](../bug-fix/2026-07-31-web-stop-preserves-queue.md) is the production `keepInbox` consumer.
|
||||
|
||||
`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.
|
||||
|
||||
@@ -28,12 +28,12 @@ Public `abort()` is absent, and the disposer remains async and waits for the loo
|
||||
|
||||
## Verification
|
||||
|
||||
`Agent` exposes no public `abort()` while `cancel()`, `whenIdle()`, and `steer()` remain; ACP cancellation calls `cancel()`; teardown awaits quiescence through handle disposal, with `whenIdle()` resolving on quiescence for non-owner observers; and the suites cover cancellation and disposal as the two supported stop paths.
|
||||
`Agent` exposes no public `abort()` while `cancel()`, `whenIdle()`, and `steer()` remain; ACP cancellation calls broad `cancel()`, Web stop calls `cancel(..., { keepInbox: true })`, and teardown awaits quiescence through handle disposal. `whenIdle()` resolves on quiescence for non-owner observers, and the suites cover cancellation and disposal as the two supported stop paths.
|
||||
|
||||
## Consequences
|
||||
|
||||
A future plugin cannot abort only the current model/tool step while preserving queued prompts through the public interface. If that use case becomes real, it should return with a named consumer and a narrower contract. Today it is latent generality that keeps a private loop mechanic public.
|
||||
A plugin can abort the active turn while preserving queued prompts through `keepInbox`, but it cannot abort only one model/tool step while leaving that turn running. A step-only use case would need a named consumer and a narrower contract; exposing the private loop mechanic remains unjustified.
|
||||
|
||||
## Related
|
||||
|
||||
This Agent Note only removes the redundant stop verb. Mid-turn steering remains an intentional message path; quiescence observation remains via `whenIdle()`. The resulting public surface is `send()`, `steer()`, `inject()`, `cancel()`, `whenIdle()`, status, options, session, and identity.
|
||||
This Agent Note only removes the redundant stop verb. Mid-turn steering remains an intentional message path; quiescence observation remains via `whenIdle()`. The resulting delivery surface is `followup()`, `steer()`, and `inject()`; stopping and observation remain with `cancel()` and `whenIdle()`.
|
||||
|
||||
@@ -8,15 +8,15 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
公共 `Agent` handle 暴露了两种相互重叠的在途工作停止方式：仅针对步骤的 `abort()` 和感知队列的 `cancel()`。前者保留已排队输入，后者则清除已排队和 steering（中途引导）工作，并中止活动轮次。在生产中，ACP（Agent Client Protocol）对 `session/cancel` 使用 `cancel()`，生命周期拥有者则通过 `AgentHandle.dispose()` 拆除 agent（智能体）。没有生产调用方需要一个裸的、仅针对步骤的 abort。
|
||||
公共 `Agent` handle 暴露了两种相互重叠的在途工作停止方式：仅针对步骤的 `abort()` 和感知队列的 `cancel()`。前者保留已排队输入，后者原本只暴露广义默认行为，该行为会清除已排队和 steering（中途引导）工作，同时中止活动轮次。`cancel(cause, { keepInbox: true })` 现在无需暴露私有轮次 holder 即可覆盖生产环境的 Web 停止策略；ACP 保留广义取消，生命周期拥有者则通过 `AgentHandle.dispose()` 拆除 agent（智能体）。没有生产调用方需要一个裸的、仅针对步骤的 abort。
|
||||
|
||||
行为差异确实存在，但已发布代码不需要较窄的操作。AgentLoop 改为为整个轮次拥有一个私有取消 holder。`cancel(cause?)` 携带类型化的 `user` 或 `parent` 原因，默认为 `user`，并丢弃待处理输入；释放仍是单独的生命周期中断。完整的归属与传播契约位于[显式轮次取消 Agent Note（agent 决策记录）](../architecture/2026-07-16-explicit-turn-cancellation.md)。
|
||||
行为差异确实存在，但已发布代码不需要独立的更窄动词。AgentLoop 为整个轮次拥有一个私有取消 holder。`cancel(cause, options?)` 携带显式且类型化的 `user` 或 `parent` 原因；其广义默认行为丢弃待处理输入，`keepInbox` 则为后续轮次保留待处理工作。资源释放仍是单独的生命周期中断。完整的归属与传播契约位于[显式轮次取消 Agent Note](../architecture/2026-07-16-explicit-turn-cancellation.md)。
|
||||
|
||||
多余的公开接口使得循环不得不承载一个本质上属于内部拆卸的公开动词：`abort()` 必须被文档描述为有别于队列感知的取消，尽管 UI 取消几乎总是需要更广泛的操作。
|
||||
多余的公开接口使循环承载了一个本质上属于内部拆卸的公开动词。带选项的 `cancel()` 可以表达调用方策略，而无需暴露第二个 holder 形态的操作。
|
||||
|
||||
## 决策
|
||||
|
||||
`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。
|
||||
`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；非拥有者使用广义 `cancel()` 放弃当前和已排队工作，或使用 `keepInbox` 中止活动轮次并保留待处理工作。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。[Web 停止决策](../bug-fix/2026-07-31-web-stop-preserves-queue.md)是生产环境中的 `keepInbox` 消费方。
|
||||
|
||||
`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。
|
||||
|
||||
@@ -28,12 +28,12 @@ Status: implemented
|
||||
|
||||
## 验证
|
||||
|
||||
`Agent` 不再暴露公开的 `abort()`，而 `cancel()`、`whenIdle()` 和 `steer()` 保留；ACP 取消调用 `cancel()`；拆卸通过 handle disposal 等待完全停稳，`whenIdle()` 在完全停稳时为非所有者观测者 resolve；测试套件覆盖取消和 disposal 作为两条受支持的停止路径。
|
||||
`Agent` 不再暴露公开的 `abort()`，而 `cancel()`、`whenIdle()` 和 `steer()` 保留；ACP 取消调用广义 `cancel()`，Web 停止调用 `cancel(..., { keepInbox: true })`，拆卸则通过 handle disposal 等待完全停稳。`whenIdle()` 在完全停稳时为非所有者观测者 resolve；测试套件覆盖取消和资源释放这两条受支持的停止路径。
|
||||
|
||||
## 后果
|
||||
|
||||
未来的插件无法通过公开接口仅中止当前模型/工具步骤而保留队列中的提示词。如果该用例变为现实需求，它应当带着一个具名消费方和更窄的契约回归。目前它是将私有循环机制保持公开的潜在泛化。
|
||||
插件可以通过 `keepInbox` 在保留已排队提示词的同时中止活动轮次，但不能只中止某一个模型／工具步骤而让该轮次继续运行。仅步骤用例需要具名消费方和更窄契约；暴露私有循环机制仍缺乏正当理由。
|
||||
|
||||
## 相关
|
||||
|
||||
本 Agent Note 只移除冗余的停止动词。轮次中途 steering 仍是一条有意保留的消息路径；完全停稳观察仍通过 `whenIdle()` 完成。最终公共表面包括 `send()`、`steer()`、`inject()`、`cancel()`、`whenIdle()`、status、options、会话和 identity。
|
||||
本 Agent Note 只移除冗余的停止动词。轮次中途 steering 仍是一条有意保留的消息路径；完全停稳观察仍通过 `whenIdle()` 完成。最终交付接口包括 `followup()`、`steer()` 和 `inject()`；停止与观察仍通过 `cancel()` 和 `whenIdle()` 完成。
|
||||
|
||||
@@ -3,4 +3,4 @@
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-20-remove-agent-boundary-mirror-events.md
|
||||
2026-06-20-remove-agent-boundary-mirror-events.md: 24ac1e32667eb325044f4df583cde509421af3de
|
||||
2026-06-20-remove-agent-boundary-mirror-events.zh.md: c7bf4fe9cc9ff50deb99527530ee0256fca10cb3
|
||||
2026-06-20-remove-agent-boundary-mirror-events.zh.md: 633c349a3e3a1efc4089fa9ae639aa457dd3d619
|
||||
|
||||
@@ -7,7 +7,7 @@ Status: implemented
|
||||
<!-- 以修订、收窄后的形式落地：
|
||||
移除了四个轮次/步骤边界镜像；此处保留了 `agent/steering` 和
|
||||
`agent/stream-chunk`（它们不是持久边界镜像——参见
|
||||
“范围：移除什么、不移除什么”）。原始提案将 `agent/steering` 与其他项一并
|
||||
「范围：移除什么、不移除什么」）。原始提案将 `agent/steering` 与其他项一并
|
||||
移除；把它排除在外，使本 Agent Note 的范围保持在边界上。后来每个保留事件
|
||||
都由各自的决策移除——参见
|
||||
[停止将 token 流镜像为 agent 事件](../../archived/simplification/2026-07-02-remove-stream-chunk-mirror.md)
|
||||
@@ -15,17 +15,17 @@ 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。
|
||||
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环境消费方；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。
|
||||
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事件的先后关系变得微妙：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。
|
||||
|
||||
## 决策
|
||||
|
||||
将 `session/event` 作为唯一的实时边界/transcript（文本记录）流。需要渲染轮次、工具调用、工具结果、助手消息和持久化边界的消费方统一订阅 `session/event`，从持久化层使用的同一套事件词汇中派生 UI。
|
||||
将 `session/event` 作为唯一的实时边界/transcript 流。需要渲染轮次、工具调用、工具结果、助手消息和持久化边界的消费方统一订阅 `session/event`，从持久化层使用的同一套事件词汇中派生 UI。
|
||||
|
||||
四个持久边界镜像——`agent/turn-start`、`agent/turn-end`、`agent/step-start`、`agent/step-end`——已从 agent（智能体）事件分类中移除。希望在边界处取得 agent handle 的 UI 会保留来自 `agent/created`/`agent/disposed` 的实时目标对象，并直接比较其会话；`dsh-ui-stdio` 据此为应用拥有的 agent 标记 `[main turn N]` 头部，其他会话则渲染其持久 id。规范记录仍是事件溯源会话日志。
|
||||
四个持久边界镜像——`agent/turn-start`、`agent/turn-end`、`agent/step-start`、`agent/step-end`——已从 agent（智能体）事件分类体系中移除。希望在边界处取得 agent handle 的 UI 会保留来自 `agent/created`/`agent/disposed` 的实时目标对象，并直接比较其会话；`dsh-ui-stdio` 据此为应用拥有的 agent 标记 `[main turn N]` 头部，其他会话则渲染其持久 id。规范记录仍是事件溯源会话日志。
|
||||
|
||||
步骤镜像（完全没有消费方）最先在[事件域语义 Agent Note（agent 决策记录）](../architecture/2026-06-30-event-domain-semantics.md) 中移除；该 Agent Note 当时以 stdio UI 需要在轮次边界取得 `Agent` handle 为由，保留了轮次镜像。本 Agent Note 完成余下工作：`dsh-ui-stdio` 是可随时丢弃的测试 REPL，其渲染可以自由变化，因此“ui-stdio 需要它”并不是保留镜像的理由——它读取 `session/event`，只保留自己的实时目标对象。
|
||||
步骤镜像（完全没有消费方）最先在[事件域语义 Agent Note](../architecture/2026-06-30-event-domain-semantics.md) 中移除；该 Agent Note 当时以 stdio UI 需要在轮次边界取得 `Agent` handle 为由，保留了轮次镜像。本 Agent Note 完成余下工作：`dsh-ui-stdio` 是可随时丢弃的测试 REPL，其渲染可以自由变化，因此「ui-stdio 需要它」并不是保留镜像的理由——它读取 `session/event`，只保留自己的实时目标对象。
|
||||
|
||||
## 范围：移除什么、不移除什么
|
||||
|
||||
@@ -35,7 +35,7 @@ Status: implemented
|
||||
|
||||
- `agent/steering`——不是边界，因此不在本决策范围内（原始提案将其一并移除；在此会造成范围蔓延）。它镜像持久的 `steering/message` 控制记录，而非边界，后来由自己的后续决策移除：[移除 `agent/steering` 镜像 emit](../../archived/simplification/2026-07-04-remove-agent-steering-mirror.md)。
|
||||
- `agent/stream-chunk`——实时 token 流。不在本决策范围内（它镜像持久的 `assistant/chunk`，而非边界），后来由自己的后续决策移除：[停止将 token 流镜像为 agent 事件](../../archived/simplification/2026-07-02-remove-stream-chunk-mirror.md)。
|
||||
- `agent/created`、`agent/disposed`、`agent/status`、`agent/error`、`agent/queued`——不属于 transcript 数据的生命周期/控制事件。尤其是 `agent/queued`，它是在任何持久事件存在之前触发的 inbox 确认（取消的排队工作可能永远不会进入日志），所以有意只保留为实时事件。
|
||||
- `agent/created`、`agent/disposed`、`agent/status`、`agent/error`、`agent/queued`——不属于 transcript 数据的生命周期/控制事件。尤其是 `agent/queued`，它是在任何持久事件存在之前触发的收件箱确认（取消的排队工作可能永远不会进入日志），所以有意只保留为实时事件。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
@@ -44,4 +44,4 @@ Status: implemented
|
||||
|
||||
## 后果
|
||||
|
||||
插件不能再从便捷的 `Agent` 优先事件观察轮次/步骤边界。它需要订阅 `session/event`；如果需要实时对象，则通过 `ctx.agents` 解析共享 id，或保留自己已经拥有的对象。这是可以接受的取舍：边界消费方不应依赖可能与持久日志发生漂移的第二条事件 feed。
|
||||
插件不能再通过便捷的、以 `Agent` 为首个参数的事件观察轮次/步骤边界。它需要订阅 `session/event`；如果需要实时对象，则通过 `ctx.agents` 查找共享 id 对应的对象，或保留自己已经拥有的对象。这是可以接受的取舍：边界消费方不应依赖可能与持久日志发生漂移的第二个事件源。
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-20-unify-agent-and-session-id.md
|
||||
2026-06-20-unify-agent-and-session-id.md: c55152f4f13fe0acb530503e84f465799007cff7
|
||||
2026-06-20-unify-agent-and-session-id.zh.md: 1fa2fe1fd64478bfe17c590e45abd0cf8281cbe4
|
||||
2026-06-20-unify-agent-and-session-id.zh.md: 92774a3a1b2ff90ccc6163967b4281552a253575
|
||||
|
||||
@@ -16,7 +16,7 @@ ACP（Agent Client Protocol）对两种 identity 使用相同值。Stdio 和钩
|
||||
|
||||
## 决策
|
||||
|
||||
agent 的注册表 id 等于其会话 id。`CreateAgentOptions` 接受一个 `sessionId`，同时用于两个最终注册表条目；恢复时以 `resumeSessionId` 注册 agent；进程内 subagent 创建使用子会话 id；`Session.id` 则派生自 `header.id`。远程 ACP 运行没有本地 agent/会话对：它保留一个由父项铸造的生命周期 id，而子服务器线协议内的会话 id 仅用于 ACP 调用。现有创建事务、最终条目冲突检查和精确条目分离语义保持不变；唯一职责是在本地 id 之间转换的 map 与字段已经消失。
|
||||
agent 的注册表 id 等于其会话 id。`CreateAgentOptions` 接受一个 `sessionId`，同时用于两个最终注册表条目；恢复时以 `resumeSessionId` 注册 agent；进程内 subagent 创建使用子会话 id；`Session.id` 则派生自 `header.id`。远程 ACP 运行没有本地 agent/会话对：它保留一个由父项铸造的生命周期 id，而子服务器仅在协议交互中使用的会话 id 对 ACP 调用保持私有。现有创建事务、最终条目冲突检查和精确条目分离语义保持不变；唯一职责是在本地 id 之间转换的 map 与字段已经消失。
|
||||
|
||||
配置驱动路径保留 `agents[].id` 作为稳定配置标签，而非实时路由 identity。普通的全新启动会铸造组合 id `${label}-session-${randomUUID()}`，使持久重启不会冲突。耦合应用可以预先铸造并传入精确的 `sessionId`：首次使用时创建它，而当持久化服务已经存在时，AgentLoop 重新挂载会在同一 identity 下恢复已物化历史。`resumeSessionId` 则要求已有的持久化 identity。两个精确 id 输入互斥。Stdio 使用“恢复或创建”形式，使配置创建的 agent 和 UI 在循环重载之间共享一个不透明 identity，而不是根据前缀猜测。日志可以使用稳定标签，而所有实时与持久查找都使用同一个 `SessionId`。
|
||||
|
||||
@@ -28,12 +28,12 @@ agent 的注册表 id 等于其会话 id。`CreateAgentOptions` 接受一个 `se
|
||||
|
||||
## 验证
|
||||
|
||||
- Agent 创建/恢复和 subagent 创建只携带一个 identity，`Session` 也只在一个位置存储它。
|
||||
- agent 创建/恢复和 subagent 创建只携带一个 identity，`Session` 也只在一个位置存储它。
|
||||
- 创建事务继续覆盖最终条目冲突、精确条目分离、回滚和完全停稳，无需 identity 特有的生命周期状态。
|
||||
- ACP、stdio、钩子、bash 归属、持久化和 lineage 直接使用共享 `SessionId`。ACP subagent 后端在父命名空间中铸造其生命周期 id，因为子服务器返回的会话 id 仅在服务器本地有效；ACP bridge 根据正向会话 map 验证精确的 `Agent` 归属；JSON-RPC 只转发生命周期事件中由服务快照保存的 `local` 标记为 true 的事件，从带范围的事件 carrier 取得委托父项，并且不保留子 identity 或 lineage cache。
|
||||
- 配置驱动的恢复还是创建策略是显式的，并在持久化重启场景下得到覆盖。
|
||||
- 配置驱动的恢复或创建策略是显式的，并在持久化重启场景下得到覆盖。
|
||||
- 生产监听器搜索确认保留 `agent/created`/`agent/disposed` 及其发布语义。
|
||||
- 类型检查、覆盖率、快照、doc-sync、module-graph 校验、构建与 hygiene 全部通过。
|
||||
- 类型检查、覆盖率、快照、doc-sync（文档同步门禁）、module-graph 校验、构建与 hygiene 全部通过。
|
||||
|
||||
## 后果
|
||||
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-06-26-fsspec-style-fs-seam.md
|
||||
2026-06-26-fsspec-style-fs-seam.md: b5c201fb192782f130d3609978d16f0fc6d4d55e
|
||||
2026-06-26-fsspec-style-fs-seam.zh.md: 3e4e6c439c85cc8e105766ee7f43c95640e26a43
|
||||
2026-06-26-fsspec-style-fs-seam.zh.md: 8346ba76d099a458236e200f304c4921553584d3
|
||||
|
||||
@@ -15,7 +15,7 @@ Status: implemented
|
||||
|
||||
这还造成了一个真实的用户体验死胡同：窗口化读取记录 `view: partial`，而 partial 视图无法授权 `edit`。一个模型读取了大文件的第 100-150 行，如果想编辑第 120 行，就必须先获取一次 `full` 读取，而对于超过读取上限的文件这可能做不到。字面编辑实际上只需要新鲜度：被匹配的字节仍然来自模型所读取的那个版本即可。
|
||||
|
||||
旧 Agent Note（agent 决策记录）已经推迟了独立的 `@deepseek-ai/dsh-fs-policy` 包。本 Agent Note 构建该层，使 `ctx.fs` 保持接近 fsspec 风格的存储原语（`info`/`cat`/`open`），但不把它变成完整的 fsspec。
|
||||
旧 Agent Note 已经推迟了独立的 `@deepseek-ai/dsh-fs-policy` 包。本 Agent Note 构建该层，使 `ctx.fs` 保持接近 fsspec 风格的存储原语（`info`/`cat`/`open`），但不把它变成完整的 fsspec。
|
||||
|
||||
## 决策
|
||||
|
||||
@@ -57,15 +57,15 @@ type FsWriteIntent =
|
||||
|
||||
`stat` 返回元数据而非内容。`version` 是新鲜度令牌；`type` 让执行器在读取前拒绝目录/特殊文件；`size` 让 `read` 工具无需通过失败探测即可选择 `readText` 还是 `streamText`。`undefined` 表示目标不存在。
|
||||
|
||||
`readText` 读取整个常规文本文件。`streamText` 以相同的文本语义流式读取大文件。两个提供方原语负责常规文件检查、UTF-8 解码、二进制/NUL 拒绝以及 `FS_NOT_TEXT`；策略层从不处理原始字节，也不重新实现跨分片解码。`readText` 是小文件/直接全文件原语，而面向模型的大文件读取使用 `streamText`。
|
||||
`readText` 读取整个普通文本文件。`streamText` 以相同的文本语义流式读取大文件。两个提供方原语负责普通文件检查、UTF-8 解码、二进制/NUL 拒绝以及 `FS_NOT_TEXT`；策略层从不处理原始字节，也不重新实现跨分片解码。`readText` 是小文件/直接全文件原语，而面向模型的大文件读取使用 `streamText`。
|
||||
|
||||
`writeText` 是原子的临时文件 + rename，带有显式的写入期望。`createIfAbsent` 创建不存在的目标，对已存在的目标以 `FS_NOT_OBSERVED` 拒绝；这是 owner 没有先前读取时使用的路径。`replaceIfVersion` 仅在目标以观测到的版本存在时替换；目标不存在或版本不匹配时抛出 `FS_STALE_VERSION`。
|
||||
`writeText` 通过临时文件 + rename 实现原子写入，并带有显式的写入期望。`createIfAbsent` 创建不存在的目标，对已存在的目标以 `FS_NOT_OBSERVED` 拒绝；这是 owner 没有先前读取时使用的路径。`replaceIfVersion` 仅在目标以观测到的版本存在时替换；目标不存在或版本不匹配时抛出 `FS_STALE_VERSION`。
|
||||
|
||||
`editText` 是提供方级别的受保护文本变更。启用守卫时，它首先验证目标仍以 `expected.version` 存在，然后读取当前文本、应用字面替换并原子写入。陈旧检查必须在字面匹配之前发生，这样基于旧读取的编辑会报告 `FS_STALE_VERSION`，而不是对更新内容进行匹配后报告 `FS_EDIT_NOT_FOUND` 或 `FS_AMBIGUOUS_EDIT`。将此原语保留在提供方 seam 上，保持了后端本地锁定的能力，也让未来的远程后端能够实现原生的 compare-and-edit，而无需策略层拉取整个文件。
|
||||
|
||||
这是一个*文本存储* seam，刻意比字节级 fsspec（`cat`/`open` 返回原始字节）高半个层次。UTF-8 解码、二进制/NUL 拒绝、受保护的全文件写入和受保护的字面文本编辑都在提供方内完成，因此策略层从不接触原始字节、不重新实现跨分片解码、也不将陈旧检查与变更临界区分离。面向模型的概念仍然不下沉到提供方：行窗口、带行号的行、渲染的页脚、观测状态存储都不会泄漏下去。
|
||||
|
||||
从 `dsh-fs` 删除：`readPage`、`FsExpectation`、`FsView`、`FsStateSource`、`FsReadRequest`、`FsTextLine`、行/窗口常量、`formatReadBody` 和 observed-state `WeakMap`。`applyEdit` 由更窄的提供方原语 `editText` 取代，其契约是带版本守卫的字面文本变更，而非策略层读取授权。`FS_PARTIAL_OBSERVATION` code 也从 `FsErrorCode` 分类中移除：新鲜度授权没有部分/完整之分，因此没有任何路径会抛出它。`FsTargetKey` 和 `FsVersion` 按现有[品牌化 id Agent Note](../architecture/2026-06-20-branded-ids.md) 成为品牌化不透明 id。
|
||||
从 `dsh-fs` 删除：`readPage`、`FsExpectation`、`FsView`、`FsStateSource`、`FsReadRequest`、`FsTextLine`、行/窗口常量、`formatReadBody` 和 observed-state `WeakMap`。`applyEdit` 由更窄的提供方原语 `editText` 取代，其契约是带版本守卫的字面文本变更，而非策略层读取授权。`FS_PARTIAL_OBSERVATION` 错误码也从 `FsErrorCode` 分类体系中移除：新鲜度授权没有部分/完整之分，因此没有任何路径会抛出它。`FsTargetKey` 和 `FsVersion` 按现有[品牌化 id Agent Note](../architecture/2026-06-20-branded-ids.md) 成为品牌化不透明 id。
|
||||
|
||||
## 策略契约
|
||||
|
||||
@@ -76,14 +76,14 @@ type FsWriteIntent =
|
||||
该插件决定三个 `fs/*` 事件：
|
||||
|
||||
- `fs/write-intent`——无先前观测 ⇒ `{ kind: 'createIfAbsent' }`（只有新文件可以盲创建）；有先前观测 ⇒ `{ kind: 'replaceIfVersion', version: vObserved }`（已有文件仅在自观测以来未变时才替换）。单槽决策；不调用 `next()`。
|
||||
- `fs/edit-intent`——要求 owner 有先前观测（否则 `FS_NOT_OBSERVED`）；返回 `{ version: vObserved }` 作为 CAS 基础。它不实现字面替换——它授权并提供版本，提供方的变更临界区负责应用守卫，因此基于同一观测版本的并发编辑仍然是一赢一陈旧。
|
||||
- `fs/edit-intent`——要求 owner 有先前观测（否则 `FS_NOT_OBSERVED`）；返回 `{ version: vObserved }` 作为 CAS 基础。它不实现字面替换——它授权并提供版本，提供方的变更临界区负责应用守卫，因此基于同一观测版本的并发编辑仍然是一个成功，另一个因版本陈旧而失败。
|
||||
- `fs/observed`——在成功的读取/写入/编辑后，为该 owner+target 记录 `{ version }`。同步、仅副作用的 `WeakMap.set`。
|
||||
|
||||
该插件不做任何文件系统 I/O：「你是否观测过此文件？」是一次 `WeakMap` 查找，而「你读取的版本是否仍然是当前版本？」在 `ctx.fs.editText`/`writeText` 内部、与执行变更相同的原子锁中决定——插件只提供 `vObserved` 作为基础。
|
||||
|
||||
## 工具契约
|
||||
|
||||
`dsh-tool-fs` 保持相同的 schema 和提示词表面。`read` 仍然暴露 `file_path`、`offset` 和 `limit`；`write` 和 `edit` 不变。它是执行器：验证模型参数，通过 `ctx.fs` 直接读取/写入/编辑，拥有行窗口化和结果渲染（`N: text`、页脚、`<path>/<content>` 信封），并分发 `fs/*` 事件。
|
||||
`dsh-tool-fs` 保持相同的 schema 和提示词表面。`read` 仍然暴露 `file_path`、`offset` 和 `limit`；`write` 和 `edit` 不变。它是执行器：验证模型参数，通过 `ctx.fs` 直接读取/写入/编辑，拥有行窗口化和结果渲染（`N: text`、页脚、`<path>/<content>` 封装），并分发 `fs/*` 事件。
|
||||
|
||||
每个变更操作先分发其 intent waterfall（瀑布式事件），带有 `undefined` 裸提供方默认值，然后调用 `ctx.fs`，再发出 `fs/observed`。例如 `write` 执行 `ctx.waterfall('fs/write-intent', target, exec, () => undefined)` → `ctx.fs.writeText(target, content, intent)` → `ctx.emit('fs/observed', …)`。`read` 先 stat 一次，然后读取/流式读取，构建窗口，最后发出 `fs/observed`。将 `exec` 作为 actor 传递，让 `dsh-fs-policy` 无需工具深入策略即可派生 owner。
|
||||
|
||||
@@ -109,7 +109,7 @@ type FsWriteIntent =
|
||||
|
||||
## 验证
|
||||
|
||||
`dsh-fs` 精确暴露 `resolve`/`stat`/`readText`/`streamText`/`writeText`/`editText`（`stat` 返回 `FsInfo | undefined`，`writeText` 接受 `FsWriteIntent`），已删除的类型/原语不再存在；`dsh-fs-local` 不包含行、视图或 `formatReadBody` 逻辑；面向模型的 schema 保持逐字节不变。测试固定了以下行为：窗口化读取授权对未变文件的后续编辑；基于陈旧读取的编辑在尝试字面匹配之前报告 `FS_STALE_VERSION`；版本 CAS 行为得以保留；观测契约成立（`read` 工具的读取记录观测状态；直接 `ctx.fs` 读取不记录）；`dsh-fs-policy` 具有 HMR（热模块替换）/dispose（资源释放）覆盖率。
|
||||
`dsh-fs` 精确暴露 `resolve`/`stat`/`readText`/`streamText`/`writeText`/`editText`（`stat` 返回 `FsInfo | undefined`，`writeText` 接受 `FsWriteIntent`），已删除的类型/原语不再存在；`dsh-fs-local` 不包含行、视图或 `formatReadBody` 逻辑；面向模型的 schema 保持逐字节不变。测试固定了以下行为：窗口化读取授权对未变文件的后续编辑；基于陈旧读取的编辑在尝试字面匹配之前报告 `FS_STALE_VERSION`；版本 CAS 行为得以保留；观测契约成立（`read` 工具的读取记录观测状态；直接 `ctx.fs` 读取不记录）；`dsh-fs-policy` 具有 HMR（热模块替换）/dispose（资源释放）测试覆盖。
|
||||
|
||||
## 后续扩展
|
||||
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-04-drop-image-content-block.md
|
||||
2026-07-04-drop-image-content-block.md: cdedf4bd5dfe60c72cea185d88b83d3b93928ff0
|
||||
2026-07-04-drop-image-content-block.zh.md: 683fd1cdb47e3fcd68ff601c0b0ce4b46f8b06d2
|
||||
2026-07-04-drop-image-content-block.zh.md: 4caf717a360c1a3d536415a84e57fd9fff32c710
|
||||
|
||||
@@ -6,24 +6,24 @@ 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；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。
|
||||
`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）估算器为其按固定常量计入 token 用量，并将其渲染为 `[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参考文档。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。
|
||||
当适配器和压缩支持 image 时，`ContentBlockMap` 可以重新引入 image 内容块。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。
|
||||
|
||||
## 验证
|
||||
|
||||
除 Agent Note（agent 决策记录）之外，没有任何地方构造 harness `ImageBlock`。ACP 独立的入站图像拒绝路径仍有测试；适配器、codec 和压缩的默认分支则使用插件定义的块类型覆盖。
|
||||
除 Agent Note 之外，没有任何地方构造 harness `ImageBlock`。ACP 独立的入站图像拒绝路径仍有测试；适配器、codec 和压缩的默认分支则使用插件定义的块类型覆盖。
|
||||
|
||||
## 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（适配器映射与压缩定价），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——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。
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-04-tighten-hook-protocol-contract.md
|
||||
2026-07-04-tighten-hook-protocol-contract.md: a67d0e8447e36516006e57051581c03877c1ba12
|
||||
2026-07-04-tighten-hook-protocol-contract.zh.md: c0f97cf39adb0fd17caa3ad2c518d26736bfc7a6
|
||||
2026-07-04-tighten-hook-protocol-contract.zh.md: 638455d89923d0636483c8e8f39fb0613daa1376
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Agent Note: 收紧 hook-protocol 契约——dialect、废弃字段、双重默认值与 lib 拥有的 `hook/result` 语义
|
||||
# Agent Note: 收紧 hook-protocol 契约——dialect、被丢弃的字段、双重默认值与 lib 拥有的 `hook/result` 语义
|
||||
|
||||
Status: implemented
|
||||
|
||||
@@ -6,16 +6,16 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
`dsh-hook-protocol`/bridge 契约中有四部分没有遵守 [subagent observe/enrich Agent Note（agent 决策记录）](../../archived/feature/2026-06-30-subagent-observe-enrich.md)记下的准则——后者因缺少消费方而删除 `agentType` 生命周期字段，以下各项没有通过同一检验：
|
||||
`dsh-hook-protocol`/bridge 契约中有四部分没有遵守 [subagent observe/enrich Agent Note](../../archived/feature/2026-06-30-subagent-observe-enrich.md)记下的准则——后者因缺少消费方而删除 `agentType` 生命周期字段，以下各项没有通过同一检验：
|
||||
|
||||
1. **`HookDialect` 的 `'native'` 变体**（`packages/hooks/hook-protocol/src/types.ts`）没有生产者——bridge 会标记 `'claude'` 和 `'codex'`；所有位置中唯一构造 `'native'` 的是该库自己的单元测试。字段自身的 JSDoc 将 `dialect` 定义为“运行它的 bridge”，而 native 不是 bridge：[拦截 seam Agent Note](../feature/2026-06-30-interception-seams.md) 记载 native 钩子不是一个包，并且“native 插件无需持久钩子日志即可使用类型化 Decision”；旗舰 native 插件实践示例恰好断言了这一点（完全没有 `hook/*` 事件）。
|
||||
2. **`HookOutput.suppressOutput`**（同一文件）被 codec 解析后在所有路径上均被丢弃：没有 bridge 分支处理它、没有合并 fold、没有 warn、没有 deferred-list 行——在所有「被解析但未兑现」的同类字段中它是唯一没有明确延期声明的（`updatedInput` → 一条 warn 日志加 [pre-tool-input-rewrite 提案](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)；`systemMessage` → 一条 warn 日志加 README deferred 行；`continue`/`stopReason` → 一个 `TODO(hook-continue-false)` 锚点加 `'stop'` decision 记录）。从结构上看根本无物可抑制：钩子 stdout 从不进入任何 transcript（文本记录）（上下文仅通过 `additionalContext` 流入；日志只记录 `decision`/`stderrSummary`），因此钩子作者设置 `suppressOutput: true` 得到的是无声的空操作，且无任何警告。
|
||||
3. **`defaultTimeoutMs` 在两个 bridge 配置中以浮动字面量双重默认**——schema 的 `.default(600_000)` 加上一个 `?? 600_000` 回退（`packages/hooks/hooks-claude/src/index.ts`、`packages/hooks/hooks-codex/src/index.ts`），一个协议级常量在每个 bridge 中有两个归属地，两个 bridge 可能在共享默认值上悄然分歧。*提案最初的补救措施是彻底删除该旋钮，但被 no-hardcoded-tunables 审计所取代：审计保留了该旋钮作为 bridge 拥有的显式配置（并在旁边新增了 `stderrSummaryMaxChars`）；剩下要修的是字面量的归属地。*
|
||||
2. **`HookOutput.suppressOutput`**（同一文件）被 codec 解析后在所有路径上均被丢弃：没有 bridge 分支处理它、没有合并 fold、没有 warn、没有 deferred-list 行——在所有「被解析但未兑现」的同类字段中它是唯一没有明确延期声明的（`updatedInput` → 一条 warn 日志加 [pre-tool-input-rewrite 提案](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)；`systemMessage` → 一条 warn 日志加 README deferred 行；`continue`/`stopReason` → 一个 `TODO(hook-continue-false)` 锚点加 `'stop'` decision 记录）。从结构上看根本无物可抑制：钩子 stdout 从不进入任何 transcript（文本记录）；上下文仅通过 `additionalContext` 流入，日志也只记录 `decision`/`stderrSummary`。因此，钩子作者设置 `suppressOutput: true` 得到的是无声的空操作，且无任何警告。
|
||||
3. **`defaultTimeoutMs` 在两个 bridge 配置中都以游离的字面量重复设置了默认值**——schema 的 `.default(600_000)` 加上一个 `?? 600_000` 回退（`packages/hooks/hooks-claude/src/index.ts`、`packages/hooks/hooks-codex/src/index.ts`），一个协议级常量在每个 bridge 中有两个归属地，两个 bridge 可能在共享默认值上悄然分歧。*提案最初的补救措施是彻底删除该旋钮，但被 no-hardcoded-tunables 审计所取代：审计保留了该旋钮作为 bridge 拥有的显式配置（并在旁边新增了 `stderrSummaryMaxChars`）；剩下要修的是字面量的归属地。*
|
||||
4. **`hook/result` 的语义存在于两个 bridge 中（各一份），而非拥有该事件的 lib。** `summarize()`——stderr 截断规则——在 `packages/hooks/hooks-claude/src/index.ts` 与 `packages/hooks/hooks-codex/src/index.ts` 中逐字节相同；decision 字符串规则 `output.decision ?? (output.continue === false ? 'stop' : 'pass')` 同样如此。然而 `dsh-hook-protocol` 声明了 `hook/result`、在文档中将 `stderrSummary` 描述为「已截断」却不拥有截断逻辑，记录了 decision 值却不拥有映射逻辑。如果某个 bridge 漂移（不同的上限、不同的回退），共享持久化事件的语义就会悄然分叉。
|
||||
|
||||
## 决策
|
||||
|
||||
`HookDialect` 是封闭的 bridge 集合：`'claude' | 'codex'`；`HookOutput` 移除了不受支持的 `suppressOutput`。`hook/result.durationMs` 保留为持久化的审计计时，仅在快照中做归一化。参考默认值各只存在一处：`DEFAULT_HOOK_TIMEOUT_MS` 与 `DEFAULT_STDERR_SUMMARY_MAX_CHARS`。`HookResultRecord` 与 `appendHookResult` 为两个 bridge 统一拥有 stderr 摘要化和 decision 推导逻辑。`BLOCKING_EXIT_CODE` 为 codec 内部常量。
|
||||
`HookDialect` 是封闭的 bridge 集合：`'claude' | 'codex'`；`HookOutput` 移除了不受支持的 `suppressOutput`。`hook/result.durationMs` 保留为持久化的审计计时，仅在快照中做归一化。参考默认值各只存在一处：`DEFAULT_HOOK_TIMEOUT_MS` 与 `DEFAULT_STDERR_SUMMARY_MAX_CHARS`。`HookResultRecord` 与 `appendHookResult` 共同负责两个 bridge 的 stderr 摘要化和 decision 推导逻辑。`BLOCKING_EXIT_CODE` 为 codec 内部常量。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
@@ -29,4 +29,4 @@ Status: implemented
|
||||
|
||||
## 后果
|
||||
|
||||
`dialect`、`suppressOutput`、可调参数和语义变更在线协议和预期输出中均不可见。代价是 `dsh-hook-protocol` 和两个 bridge 中的改动——在预发布立场下成本很低，也比让一项持久事件语义的两个副本各自老化更便宜。
|
||||
`dialect`、`suppressOutput`、可调参数和语义变更在协议格式（wire format）和预期输出中均不可见。代价是 `dsh-hook-protocol` 和两个 bridge 中的改动——在预发布立场下成本很低，也比让一项持久事件语义的两个副本各自老化更便宜。
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-12-simplify-session-log-representation.md
|
||||
2026-07-12-simplify-session-log-representation.md: a40f4013a97a9c940012dbb37d59beb2faf8fb22
|
||||
2026-07-12-simplify-session-log-representation.zh.md: a4ecd8c7340affda71d95505ab44910539e36bb9
|
||||
2026-07-12-simplify-session-log-representation.zh.md: 30808a4eff83b32bd0c47cc6e6cb211f2222b363
|
||||
|
||||
@@ -12,24 +12,24 @@ Status: implemented
|
||||
|
||||
请求头子系统实现了一套自定义的系统/工具增量编解码器和传输决策层，尽管其契约声明增量只是编码优化，而非可重建性要求。在每个 agent loop（智能体循环）实例边界保留初始/恢复的完整快照，然后在该实例的组装头发生变化时写入一条规范的完整 `request/header`，即可保留回放能力，同时删除 `SystemDelta`、`ToolsDelta`、往返回退逻辑以及持久化的 `request/header-delta` 变体。编解码器专属的词汇随编解码器一起消失，并非因为其各分支本身无效。
|
||||
|
||||
实现保留追加与替换 `sourceEventSeqs`、崩溃修复 provenance，以及所有 `SessionStartSource` 变体，因为这些字段承担审计/拦截职责，当前没有读取方并不能推翻这一点。
|
||||
实现保留追加与替换操作的 `sourceEventSeqs`、崩溃修复的溯源信息，以及所有 `SessionStartSource` 变体，因为这些字段承担审计/拦截职责，当前没有读取方并不能推翻这一点。
|
||||
|
||||
## 决策
|
||||
|
||||
`SurfaceManager.nodes` 是由事件序号组成的 `readonly number[]`；公共 `SurfaceNode` 形状、node 链接和 seq-to-node map 均已移除。内部替换 generation 信号保留。session-query 使用的完整 `foldSurface()` 读取会返回相同的数字数组表示和替换元数据，而无需让增量 manager 保留历史。工具配对 balance 和压缩使用事件序号与 surface 位置；由 compact 拥有的每切点 balance cache 不依赖 node 链接。
|
||||
`SurfaceManager.nodes` 是由事件序号组成的 `readonly number[]`；公共 `SurfaceNode` 形状、node 链接和 seq-to-node map 均已移除。内部替换 generation 信号保留。session-query 使用的完整 `foldSurface()` 读取会返回相同的数字数组表示和替换元数据，而无需让增量 manager 保留历史。工具配对 balance 和压缩（compaction）使用事件序号与 surface 位置；由 compact 拥有的每个切点的 balance cache 不依赖 node 链接。
|
||||
|
||||
请求头只使用规范的完整快照。初始与恢复锚点即使没有变化也仍是完整快照；实例内变化会追加另一个完整 `request/header`，reason 为 `change`。delta 事件、codec 类型、diff/apply 辅助函数，以及仅供 codec 使用的 `fallback` reason 均已移除。请求重建选择最新快照。
|
||||
|
||||
`SESSION_FORMAT_VERSION` 仍固定为 `0`，因此 seed、追加和持久化加载验证会显式拒绝旧 v0 `request/header-delta` 事件，以及携带已删除 `fallback` reason 的完整快照。不存在兼容性 fold 或迁移。JSONL 与 SQLite 测试固定了这一响亮失败边界；ACP（Agent Client Protocol）快照 harness 则把合法的会话中途变更表示为完整固定请求头和完整可读提示词。
|
||||
`SESSION_FORMAT_VERSION` 仍固定为 `0`，因此 seed、追加和持久化加载验证会显式拒绝旧 v0 `request/header-delta` 事件，以及携带已删除 `fallback` reason 的完整快照。不存在兼容性 fold 或迁移。JSONL 与 SQLite 测试固定了这一失败即报错的边界；ACP（Agent Client Protocol）快照 harness 则把合法的会话中途变更表示为完整固定请求头和完整可读提示词。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**保留链表节点和紧凑增量以备未来扩展。** 链接可能有助于未来的游标 API，增量在大型工具 schema 仅有少量变化时可以缩减日志。但没有已发布的游标使用这些链接，而完整快照以磁盘空间换取了显著更简单的正确性。如果头部体积确实成为问题，可以基于真实 trace 设计压缩方案或经过度量的规范增量方案。
|
||||
**保留链表节点和紧凑增量以备未来扩展。** 链接可能有助于未来的游标 API，增量在大型工具 schema 仅有少量变化时可以缩减日志。但没有已发布的游标使用这些链接，而完整快照以磁盘空间为代价，显著简化了正确性保障。如果头部体积确实成为问题，可以基于真实 trace 设计压缩方案或经过度量的规范增量方案。
|
||||
|
||||
## 验证
|
||||
|
||||
单元覆盖率固定有序 surface 的追加/替换行为、工具配对、压缩、完整请求头 fold/记录、请求重建和开发不变量。Seed 验证以及 JSONL、SQLite 加载测试会在重放前拒绝旧事件。无密钥 ACP 套件以新形状覆盖记录、刷新、重放、变化请求头固定，以及沙箱模式切换 fixture（测试前置数据）。
|
||||
单元测试覆盖并锁定有序 surface 的追加/替换行为、工具配对、压缩、完整请求头 fold/记录、请求重建和开发不变量。Seed 验证以及 JSONL、SQLite 加载测试会在回放前拒绝旧事件。无密钥 ACP 套件以新形状覆盖记录、刷新、回放、变更后请求头的固定，以及沙箱模式切换 fixture（测试前置数据）。
|
||||
|
||||
## 后果
|
||||
|
||||
完整请求头会增加日志体积，线性替换查找在极大 surface 上也可能较慢。由于先前实现调用 `indexOf`，替换原本就是线性的；benchmark 推迟到真实 trace 表明更简单的数组成为瓶颈时再进行。格式版本仍为 `0`，因此显式拒绝旧事件是预发布格式边界的永久组成部分。作为交换，surface 顺序和请求头状态现在各自只有一种表示，删除了链接维护、map、codec 分支、往返 fallback 和感知 delta 的快照规范化。
|
||||
完整请求头会增加日志体积，线性替换查找在极大 surface 上也可能较慢。由于先前实现调用 `indexOf`，替换原本就是线性的；benchmark 推迟到真实 trace 表明更简单的数组成为瓶颈时再进行。格式版本仍为 `0`，因此显式拒绝旧事件是预发布格式边界的永久组成部分。作为交换，surface 顺序和请求头状态现在各自只有一种表示，删除了链接维护、map、codec 分支、往返 fallback 和针对 delta 的快照规范化。
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-17-one-send-one-turn.md
|
||||
2026-07-17-one-send-one-turn.md: 3ae43f137206f25bdbc563875c17e24211f17d6b
|
||||
2026-07-17-one-send-one-turn.zh.md: 5ccdb2192048ecf795415bcd427f967df6a609fb
|
||||
2026-07-17-one-send-one-turn.md: 834eb2b3aace0a0aa5f86f483e7b7d7aabbe467e
|
||||
2026-07-17-one-send-one-turn.zh.md: d22893b912462cffffb0d6ecd81811e9cb5c523d
|
||||
|
||||
@@ -10,21 +10,21 @@ Suppose a caller submits message A and then message B with two `Agent.send()` ca
|
||||
|
||||
That grouping depends on timing rather than caller intent. Calls from one synchronous stack, neighboring microtasks, event listeners, and model callbacks could be grouped differently even though every caller used the same API.
|
||||
|
||||
This grouping changes behavior, not just the number of model calls. One ordinary turn owns prompt admission, `turn/start`, `turn/end`, and a durability checkpoint. If message B shares message A's turn, B can enter A's model request instead of first seeing A's closed result in the session log. Allowing one message while blocking another also requires a mixed state that no caller requested.
|
||||
This grouping changes behavior, not just the number of model calls. One ordinary turn owns one claimed follow-up, `turn/start`, `turn/end`, and a durability checkpoint. If message B shares message A's turn, B can enter A's model request instead of first seeing A's closed result in the session log. Entering one follow-up while rejecting another also requires a mixed state that no caller requested.
|
||||
|
||||
## Decision
|
||||
|
||||
The rule is simple: each successful `send()` creates one independent FIFO queue item. If that item runs, it is the only ordinary message in its turn. An item can be dropped before it starts, so the precise guarantee is at most one turn rather than exactly one; two sends are never silently combined.
|
||||
|
||||
Before enqueueing an item, `send()` checks the agent state and accepts an already identified, deeply frozen message. It mints an occurrence-local `InboxItemId` and publishes `agent/inbox/enqueue`; the pending occurrence remains addressable under the [addressable queue operations](../feature/2026-07-29-addressable-queue-operations.md) decision until the driver claims or discards it.
|
||||
Before inserting a message, `send()` checks the agent state and accepts an already identified, deeply frozen value. The durable splice and `agent/inbox/inserted { message }` retain its `MessageId`; the pending message remains addressable through `Inbox.replace()` and `Inbox.remove()` until the driver claims or discards it. The [claimed pre-step inbox decision](../architecture/2026-07-31-claimed-pre-step-inbox-lifecycle.md) owns the current lifecycle.
|
||||
|
||||
If messages A and B are both processed, B's turn starts only after A records `turn/end` and A's durability checkpoint settles. B's request therefore sees whatever closed result A left in the same session log. A checkpoint error is reported, but settlement only releases this ordering barrier; it does not make a failed write durable. Broad `cancel()`, disposal, or a failure before `turn/start` can instead discard an unstarted item without opening an empty turn.
|
||||
|
||||
Prompt admission decides one message at a time before a turn opens. An allowed prompt becomes that turn's `user/message`; a blocked prompt is discarded without opening a turn or writing session history. Mixed-batch and all-blocked-batch branches do not exist.
|
||||
At a turn boundary, the loop opens the turn and claims one follow-up after pending next-step input. `agent/pre-step` either rejects the proposal or returns the complete entering batch. A rejected follow-up remains removed and closes a blocked no-step turn without writing model-visible history. Mixed ordinary follow-up branches do not exist.
|
||||
|
||||
The no-batching rule applies only to ordinary `send()`. Running `steer()` puts input in the outbox. While a turn remains open, the loop records that input at the next step boundary and steering makes another step the default. A failure before that boundary leaves the steering staged without waking the agent; a request-error retry action or a later prompt takes it, while cancellation or disposal can discard it. When the agent is idle, `steer()` delegates to `send()`, so it creates an independent ordinary queue item.
|
||||
The no-batching rule applies only to ordinary follow-up input. `steer()` puts input in the next-step inbox and wakes the driver. During a turn, the loop can claim it at a later step boundary; while idle, the waking next-step batch starts a new turn. Input arriving after a batch was claimed waits for a later boundary, while cancellation or disposal can discard it.
|
||||
|
||||
`inject()` continues to add model-facing context without submitting an ordinary message. During a turn it waits in the outbox for a safe step boundary; while idle it appends a `user/message` directly, without opening a turn or running the model. Persistence owns the resulting eager drain. `cancel()` remains a whole-agent operation that can clear all unstarted ordinary and steering input and abort the current step. `status` and `whenIdle()` also describe the whole agent, not one message. Several one-message turns can share one `running` interval, so `running` does not prove that a turn is open.
|
||||
`inject()` continues to add model-facing context without submitting ordinary input or waking the driver. It always waits in the next-step inbox for a later pre-step, including while idle; AgentLoop records it as `user/message` only when an enter decision returns it inside a turn. `cancel()` remains a whole-agent operation that can clear all unstarted ordinary input, steering, and injection and abort the current step. `status` and `whenIdle()` also describe the whole agent, not one message.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
@@ -35,11 +35,11 @@ The no-batching rule applies only to ordinary `send()`. Running `steer()` puts i
|
||||
- Unit and property tests submit sends from the same stack, neighboring microtasks, different producers, and reentrant callbacks; every message gets its own FIFO-ordered turn.
|
||||
- A built-stdio test submits two lines and observes two model requests and two turn boundaries.
|
||||
- Delayed and rejected first-turn checkpoints keep the next turn waiting and prove that its request sees the preceding assistant result.
|
||||
- Failure-path tests cover prompt veto, listener failure, broad cancellation, disposal, and failure before `turn/start`; rejected admission creates no turn, recorded turns stay balanced, messages do not merge, and surviving queued work still drains.
|
||||
- Separate tests cover open-turn, failed-turn, and idle `steer()`, plus `inject()`, whole-agent status, and `whenIdle()`.
|
||||
- Failure-path tests cover pre-step rejection, listener failure, broad cancellation, disposal, and failure before `turn/start`; initial pre-step exits close balanced no-step turns, messages do not merge, and surviving later work still drains.
|
||||
- Separate tests cover open-turn, failed-turn, and idle `steer()`, pending `inject()`, whole-agent status, and `whenIdle()`.
|
||||
|
||||
## Consequences
|
||||
|
||||
Ordinary turn boundaries are predictable: messages A and B stay separate, and B runs only after A has closed and reached its checkpoint. Callers still do not receive a per-send completion handle; a pending occurrence can be removed through its live `InboxItemId`, broad cancellation can discard the entire unstarted tail, and status and quiescence remain agent-wide observations.
|
||||
Ordinary turn boundaries are predictable: messages A and B stay separate, and B runs only after A has closed and reached its checkpoint. Callers still do not receive a per-send completion handle; a pending message can be removed through its `MessageId`, broad cancellation can discard the entire unstarted tail, and status and quiescence remain agent-wide observations.
|
||||
|
||||
The trade-off is more model requests and more checkpoints. A busy queue can take longer to drain and can grow under sustained producers. Ordinary-send batching returns only through an explicit, measured contract.
|
||||
|
||||
@@ -10,21 +10,21 @@ Status: implemented
|
||||
|
||||
这种分组取决于运行时机，而不是调用方的意图。因此，即使所有调用方使用相同 API，来自同一个同步调用栈、相邻微任务、事件监听器和模型回调的调用也可能产生不同分组。
|
||||
|
||||
这种分组改变的不只是模型调用次数。一个普通轮次包含提示词准入、`turn/start`、`turn/end` 和持久性检查点。如果消息 B 与消息 A 共用轮次，B 可能直接进入 A 的模型请求，而不是先看到 A 在会话日志中已经关闭的结果。若系统允许一条消息、阻止另一条消息，还需要引入调用方没有请求的混合状态。
|
||||
这种分组改变的不只是模型调用次数。一个普通轮次包含一条已领取 follow-up、`turn/start`、`turn/end` 和持久性检查点。如果消息 B 与消息 A 共用轮次，B 可能直接进入 A 的模型请求，而不是先看到 A 在会话日志中已经关闭的结果。若系统让一条 follow-up 进入、却拒绝另一条，还需要引入调用方没有请求的混合状态。
|
||||
|
||||
## 决策
|
||||
|
||||
规则很简单：一次成功的 `send()` 创建一个独立的 FIFO 队列项。该队列项如果运行，就是所在轮次中唯一的普通消息。队列项可能在启动前被丢弃，因此精确保证是最多一个轮次，而不是必定一个轮次；两次 send 绝不会被悄悄合并。
|
||||
|
||||
队列项入队之前，`send()` 会检查 agent 状态，并接受已有标识且经过深度冻结的消息。它会铸造一个仅属于本次入队的 `InboxItemId`，并发布 `agent/inbox/enqueue`；根据[可寻址队列操作](../feature/2026-07-29-addressable-queue-operations.md)决策，在驱动器认领或丢弃该项之前，这次待处理入队始终可以被寻址。
|
||||
消息插入之前，`send()` 会检查 agent 状态，并接受已有标识且经过深度冻结的值。持久 splice 与 `agent/inbox/inserted { message }` 会保留其 `MessageId`；在驱动器领取或丢弃该消息之前，可以通过 `Inbox.replace()` 与 `Inbox.remove()` 寻址。当前生命周期由[已领取 pre-step inbox 决策](../architecture/2026-07-31-claimed-pre-step-inbox-lifecycle.md)规定。
|
||||
|
||||
如果消息 A、B 都进入处理，B 的轮次只能在 A 记录 `turn/end` 且 A 的持久性检查点处理结束后开始。因此，B 的请求能看到 A 在同一会话日志中留下的已关闭结果。检查点错误会照常报告，但处理结束只表示解除这道顺序屏障，不表示失败的写入已经持久化。广义 `cancel()`、dispose（资源释放）或 `turn/start` 之前的失败也可能丢弃尚未启动的队列项，而不打开一个空轮次。
|
||||
如果消息 A、B 都进入处理，B 的轮次只能在 A 记录 `turn/end` 且 A 的持久性检查点处理结束后开始。因此，B 的请求能看到 A 在同一会话日志中留下的已关闭结果。检查点错误会照常报告，但处理结束只表示解除这道顺序屏障，不表示失败的写入已经持久化。面向整个 agent 的 `cancel()`、dispose（资源释放）或 `turn/start` 之前的失败也可能丢弃尚未启动的队列项，而不打开一个空轮次。
|
||||
|
||||
提示词准入会在轮次打开前，每次只决定一条消息。获准提示词成为该轮次的 `user/message`；被阻止的提示词会被丢弃，不打开轮次，也不写入会话历史。实现中不存在混合批次或全阻止批次分支。
|
||||
轮次边界上，循环会先打开轮次，再在待处理 next-step 输入之后领取一条 follow-up。`agent/pre-step` 要么拒绝提案，要么返回进入步骤的完整批次。被拒绝的 follow-up 保持已删除，并关闭一个 blocked 的无步骤轮次，不写入模型可见历史。实现中不存在混合普通 follow-up 分支。
|
||||
|
||||
上述不合批规则只适用于普通 `send()`。agent 运行时，`steer()` 会把输入放入 outbox。只要当前轮次仍然打开，agent loop 就会在下一个步骤边界记录该输入，而 steering（中途引导）会默认让循环再执行一个步骤。在到达该边界前发生失败，会让 steering 保持暂存且不唤醒 agent；请求错误的重试动作或后续提示词会取走它，而取消或 dispose 可以将其丢弃。agent 空闲时，`steer()` 委托给 `send()`，因此会创建一个独立的普通队列项。
|
||||
上述不合批规则只适用于普通 follow-up 输入。`steer()` 会把输入放入 next-step inbox 并唤醒驱动器。在轮次期间，循环可以在后续步骤边界领取它；agent 空闲时，这个会唤醒的 next-step 批次会启动一个新轮次。批次被领取后才到达的输入会等待后续边界，而取消或 dispose 可以将其丢弃。
|
||||
|
||||
`inject()` 继续添加面向模型的上下文，而不提交普通消息。轮次打开时，该上下文会留在 outbox 中，等待安全的步骤边界；agent 空闲时，系统会直接追加一条 `user/message`，既不打开轮次，也不运行模型。持久化层独立负责由此产生的即时排空。`cancel()` 仍是面向整个 agent 的操作，可以清空所有尚未启动的普通输入和 steering，并中止当前步骤。`status` 和 `whenIdle()` 描述的也是整个 agent，而不是某一条消息。多个单消息轮次可以共用一个 `running` 区间，因此 `running` 不表示轮次一定处于打开状态。
|
||||
`inject()` 继续添加面向模型的上下文，但不提交普通输入，也不唤醒驱动器。即使 agent 空闲，它也始终在 next-step inbox 中等待后续 pre-step；AgentLoop 只会在 enter 决策于轮次内返回它时，将其记录为 `user/message`。`cancel()` 仍是面向整个 agent 的操作，可以清空所有尚未启动的普通输入、steering 和注入，并中止当前步骤。`status` 和 `whenIdle()` 描述的也是整个 agent，而不是某一条消息。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
@@ -32,14 +32,14 @@ Status: implemented
|
||||
|
||||
## 验证
|
||||
|
||||
- 单元测试和性质测试从同一调用栈、相邻微任务、不同生产方和重入回调提交 send；每条消息都会得到一个按 FIFO 排序的独立轮次。
|
||||
- 单元测试和基于属性的测试从同一调用栈、相邻微任务、不同生产方和重入回调提交 send；每条消息都会得到一个按 FIFO 排序的独立轮次。
|
||||
- stdio 构建产物测试提交两行输入，并观察到两个模型请求和两个轮次边界。
|
||||
- 延迟和拒绝第一个轮次的检查点，都能让下一个轮次保持等待，并证明其请求可以看到前一条助手结果。
|
||||
- 失败路径测试覆盖提示词否决、监听器失败、广义取消、dispose 和 `turn/start` 之前的失败；准入拒绝不会创建轮次，已记录的轮次保持边界平衡，消息不会合并，仍需处理的排队工作也能继续清空。
|
||||
- 其他测试分别覆盖轮次打开时、轮次失败后和空闲时的 `steer()`，以及 `inject()`、面向整个 agent 的状态和 `whenIdle()`。
|
||||
- 失败路径测试覆盖 pre-step reject、监听器失败、广义取消、dispose 和 `turn/start` 之前的失败；首次 pre-step 的各种退出都会关闭边界平衡的无步骤轮次，消息不会合并，之后仍需处理的工作也能继续清空。
|
||||
- 其他测试分别覆盖轮次打开时、轮次失败后和空闲时的 `steer()`，以及待处理的 `inject()`、面向整个 agent 的状态和 `whenIdle()`。
|
||||
|
||||
## 后果
|
||||
|
||||
普通轮次的边界可预测：消息 A、B 始终分开，B 只能在 A 关闭并到达检查点后运行。调用方仍然拿不到逐次 send 的完成句柄；待处理项可通过其仍有效的 `InboxItemId` 移除，广义取消可以丢弃整个尚未启动的队尾，而状态与完全停稳仍是面向整个 agent 的观察。
|
||||
普通轮次的边界可预测：消息 A、B 始终分开，B 只能在 A 关闭并到达检查点后运行。调用方仍然拿不到逐次 send 的完成句柄；待处理消息可通过其 `MessageId` 移除，广义取消可以丢弃整个尚未启动的队尾，而状态与完全停稳仍是面向整个 agent 的观察。
|
||||
|
||||
代价是模型请求和检查点都会增加。繁忙队列可能需要更长时间才能清空；如果生产方持续提交消息，队列也可能增长。只有建立显式且经过测量的契约后，才能重新引入普通 send 批处理。
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-20-remove-stdio-and-echo-agents.md
|
||||
2026-07-20-remove-stdio-and-echo-agents.md: b578e49f004cabf0e37385ea06a769ced5388217
|
||||
2026-07-20-remove-stdio-and-echo-agents.zh.md: 3d8eecf73f17f3b62dc4b35b0f3c1af03f6f3a86
|
||||
2026-07-20-remove-stdio-and-echo-agents.md: 1c7ff4b5337341ea7678c1455dd7bbd3365df6ba
|
||||
2026-07-20-remove-stdio-and-echo-agents.zh.md: 67445d25f6a0534c92c50568da898a5bffc73a52
|
||||
|
||||
@@ -18,7 +18,7 @@ The stdio and Echo agents are removed without compatibility packages, modes, com
|
||||
|
||||
The remaining application roles are explicit:
|
||||
|
||||
- [`@deepseek-ai/dsh-tui`](../../../../packages/ui/tui/README.md) owns terminal-interactive execution. It rejects non-TTY streams before Loader boot; `apps/cli/config/base.cordis.yml` plus the `tui.cordis.yml` overlay own the complete coding composition, with PTY plus terminal-snapshot coverage in `apps/cli/tests/`.
|
||||
- `@deepseek-ai/dsh-tui` owns terminal-interactive execution. It rejects non-TTY streams before Loader boot; `apps/cli/config/base.cordis.yml` plus the `tui.cordis.yml` overlay own the complete coding composition, with PTY plus terminal-snapshot coverage in `apps/cli/tests/`.
|
||||
- [`@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.
|
||||
|
||||
|
||||
@@ -18,7 +18,7 @@ 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`](../../../../packages/ui/tui/README.md) 负责终端交互式执行。它会在 Loader 启动前拒绝非 TTY 流；`apps/cli/config/base.cordis.yml` 与 `tui.cordis.yml` overlay 拥有完整 coding 组装，PTY 与终端快照覆盖则位于 `apps/cli/tests/`。
|
||||
- `@deepseek-ai/dsh-tui` 负责终端交互式执行。它会在 Loader 启动前拒绝非 TTY 流；`apps/cli/config/base.cordis.yml` 与 `tui.cordis.yml` overlay 拥有完整 coding 组装，PTY 与终端快照覆盖则位于 `apps/cli/tests/`。
|
||||
- [`@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。
|
||||
|
||||
|
||||
@@ -1,6 +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
|
||||
2026-07-20-unwrap-injected-content-envelopes.md: 32642660f7bcea748c349933b99552b1974922c5
|
||||
2026-07-20-unwrap-injected-content-envelopes.zh.md: a01a51e12cecca5bc46526ccca61dbe90eb3136f
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-20-unwrap-injected-content-envelopes.md
|
||||
2026-07-20-unwrap-injected-content-envelopes.md: aff458d76027bf5518497328ee1bd852869793b9
|
||||
2026-07-20-unwrap-injected-content-envelopes.zh.md: 59cbd823da3305185a05c79207fbd7b702a11fba
|
||||
|
||||
@@ -15,7 +15,7 @@ Two problems:
|
||||
|
||||
## Decision
|
||||
|
||||
Injected session content projects verbatim; the caller owns any framing. `deriveEventMessage` renders `user/message`, `context/message`, and `steering/message` through one shared case returning `{ role: 'user', content: event.data.content }`; their content blocks reach the model unchanged. `context/message`'s `source`/`meta` and `steering/message`'s `turn` stay in the durable event log but do not render.
|
||||
Injected session content projects verbatim; the caller owns any framing. `deriveEventMessage` renders `user/message` content blocks to the model unchanged; `source` stays in the durable event log but does not render.
|
||||
|
||||
The `ContextEnvelope` type and every `envelope` field are removed — `context/message` in `SessionEventMap`, `InjectOptions`, `HookContext`, and the `inject()`/`additionalContexts` plumbing in `dsh-agent-loop`. `workspace-context` no longer requests `'raw'`; its self-framed content renders as before. The `renderTagged`/`renderContextEnvelope` helpers are deleted. `context/message.meta` still carries durable, model-hidden JSON state.
|
||||
|
||||
|
||||
@@ -10,32 +10,32 @@ Status: implemented
|
||||
|
||||
两个问题：
|
||||
|
||||
- **没有模型在这些标签上训练过。** `<steering>` 和 `<context>` 是任何模型都未被教会去读的任意标记，因此这层框架只是徒增 token 而没有可靠效果，还可能起反作用——已录制的 transcript 显示，模型会把 `<steering>` 指令当成第三方元数据而拒绝服从，只回答原始提示。
|
||||
- **session 表层是承载框架的错误层次。** 表层的职责是把持久日志投影为模型 transcript；决定内容如何措辞并不是它的事。想要特定框架的调用方可以在注入前自行格式化内容——唯一的重度生产方（`workspace-context`）本就这样做，它自带完整的 `<system-reminder>` 框架，并用 `envelope: 'raw'` 退出 `<context>` 封套。剩下的标签机制（`ContextEnvelope` 类型，以及贯穿 `InjectOptions`、`HookContext`、`context/message` 事件和 agent loop 的 `envelope` 字段）所服务的区分，本应归属调用方。
|
||||
- **没有模型在这些标签上训练过。** `<steering>` 和 `<context>` 是任何模型都未被教会去读的任意标记，因此这层框架只是徒增 token 而没有可靠效果，还可能起反作用——已录制的 transcript 显示，模型会把 `<steering>` 指令当成第三方元数据而拒绝服从，只回答原始提示词。
|
||||
- **会话表层是承载框架的错误层次。** 表层的职责是把持久日志投影为模型 transcript；决定内容如何措辞并不是它的事。想要特定框架的调用方可以在注入前自行格式化内容——唯一的重度生产方（`workspace-context`）本就这样做，它自带完整的 `<system-reminder>` 框架，并用 `envelope: 'raw'` 退出 `<context>` 封套。剩下的标签机制（`ContextEnvelope` 类型，以及贯穿 `InjectOptions`、`HookContext`、`context/message` 事件和 agent loop（智能体循环）的 `envelope` 字段）所服务的区分，本应归属调用方。
|
||||
|
||||
## 决策
|
||||
|
||||
注入的会话内容逐字投影，框架由调用方自行负责。`deriveEventMessage` 通过一个共享分支渲染 `user/message`、`context/message` 和 `steering/message`，都返回 `{ role: 'user', content: event.data.content }`；它们的内容块原样到达模型。`context/message` 的 `source`/`meta` 和 `steering/message` 的 `turn` 保留在持久事件日志中，但不渲染。
|
||||
注入的会话内容逐字投影，框架由调用方自行负责。`deriveEventMessage` 把 `user/message` 的内容块原样送达模型；`source` 保留在持久事件日志中，但不渲染。
|
||||
|
||||
`ContextEnvelope` 类型和所有 `envelope` 字段都被移除——包括 `SessionEventMap` 中的 `context/message`、`InjectOptions`、`HookContext`，以及 `dsh-agent-loop` 中 `inject()`/`additionalContexts` 的相关管线。`workspace-context` 不再请求 `'raw'`；它自带框架的内容渲染方式不变。`renderTagged`/`renderContextEnvelope` 辅助函数被删除。`context/message.meta` 仍携带持久的、对模型隐藏的 JSON 状态。
|
||||
|
||||
封套曾携带的 `source` 归属并未丢失——它仍保留在持久事件上；只是不再渲染进 transcript。
|
||||
封套曾携带的 `source` 来源信息并未丢失——它仍保留在持久事件上；只是不再渲染进 transcript。
|
||||
|
||||
## 权衡的替代方案
|
||||
|
||||
- **保留 `<context>` 封套，只对 steering 去封套** —— 会为一个没有模型会读的框架位保留 `ContextEnvelope`/`envelope` 机制，并保留主要生产方本就退出的那种不一致。
|
||||
- **仅对插件来源的内容保留 envelope 字段** —— 会按 `source.kind` 把一条投影拆成两条，却没有观察到任何收益；插件引导 agent（智能体）时（钩子桥接器的轮次续行原因）同样希望指令被遵从，而不是被贴标签。
|
||||
- **保留 `<context>` 封套，只对 steering（中途引导）去封套** —— 会为一个没有模型会读的框架位保留 `ContextEnvelope`/`envelope` 机制，并保留主要生产方本就退出的那种不一致。
|
||||
- **仅对插件来源的内容保留 envelope 字段** —— 会按 `source.kind` 把一条投影拆成两条，却没有观察到任何收益；插件引导 agent 时（钩子桥接器的轮次续行原因）同样希望指令被遵从，而不是被贴标签。
|
||||
- **把去封套的逻辑移入适配器** —— 规范投影就是模型可见契约（「模型可见 ⟺ 已记录」）；让各适配器在框架上各行其是，会使派生的 transcript 依赖于适配器。调用方确实想要的框架应放进调用方自己的内容里，而不是适配器。
|
||||
|
||||
## 结果
|
||||
|
||||
- 中途引导与注入的 context 以与普通用户提示相同的权重到达模型。
|
||||
- 中途引导与注入的上下文以与普通用户提示词相同的权重到达模型。
|
||||
- transcript 不再区分注入内容与用户消息；需要这一区分的消费方读取持久事件日志，其中事件类型、`source` 和 `meta` 完整保留。
|
||||
- `hook-{cc,codex}-stop-continue` ACP 快照已重新录制：旧录制捕获的是模型把 steering 当作第三方元数据而拒绝服从，正是本次修复针对的失败模式。
|
||||
- `hook-{cc,codex}-stop-continue` ACP（Agent Client Protocol）快照已重新录制：旧录制捕获的是模型把 steering 当作第三方元数据而拒绝服从，正是本次修复针对的失败模式。
|
||||
- [内容块词汇表 Agent Note](../architecture/2026-06-11-content-block-vocabulary.md) 中关于带标签封套的条款已修订为指向本文。
|
||||
|
||||
## 推迟事项
|
||||
|
||||
`workspace-context` 已经自行为内容加框架：它把一个完整的 `<system-reminder>…</system-reminder>` 块作为消息内容发出，而不依赖表层封套。这种调用方自有的模式才是应保留的——表层逐字透传内容，任何框架都住在生产方自己的内容里。
|
||||
|
||||
曾经存在两条框架路径——调用方自行加框架（`workspace-context` 的 `<system-reminder>`），以及表层封套（`deriveEventMessage` 加上的 `<context>`/`<steering>`）。本次变更移除了后者，只留下调用方自有的框架。如果未来又需要带标签的框架，应由事件的 `meta` map（生产方附加、对模型隐藏的元数据字段）来统一它，交给专门的渲染器或适配器消费，而不是在 `deriveEventMessage` 中重新硬编码标签。生产方在 `meta` 中声明所需的框架，由一个渲染器统一施加；session 表层的投影始终保持逐字透传。
|
||||
曾经存在两条框架路径——调用方自行加框架（`workspace-context` 的 `<system-reminder>`），以及表层封套（`deriveEventMessage` 加上的 `<context>`/`<steering>`）。本次变更移除了后者，只留下调用方自有的框架。如果未来又需要带标签的框架，应由事件的 `meta` map（生产方附加、对模型隐藏的元数据字段）来统一它，交给专门的渲染器或适配器消费，而不是在 `deriveEventMessage` 中重新硬编码标签。生产方在 `meta` 中声明所需的框架，由一个渲染器统一施加；会话表层的投影始终保持逐字透传。
|
||||
|
||||
@@ -1,6 +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
|
||||
2026-07-22-plan-specific-collaboration-state.md: fb26d15238f0eb1b63fdccc7e48a6c49a44236cf
|
||||
2026-07-22-plan-specific-collaboration-state.zh.md: 93186eebb263458bc99e7f7562d065fbf9e5d4bf
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-22-plan-specific-collaboration-state.md
|
||||
2026-07-22-plan-specific-collaboration-state.md: d139f3d861244c80acfb604d17172461bbf0cd57
|
||||
2026-07-22-plan-specific-collaboration-state.zh.md: a11ed1f270f58fc49a41be0e475889f84fa7cd40
|
||||
|
||||
@@ -14,7 +14,7 @@ Plan mode also needs a durable stance, a reviewable plan artifact, an explicit h
|
||||
|
||||
## Decision
|
||||
|
||||
Plan mode owns a plan-specific product package: `@deepseek-ai/dsh-plan-mode` at `packages/plan/plan-mode/`. The durable fact is `plan/mode: { active: boolean }`, folded by `foldPlanMode(events)` with `false` as the empty-log value. `ctx.planMode.get(agent)` returns `{ active, pending? }`, and `set(agent, active)` records the boundary-applied selection. The existing prompt-submit, continuation, retry, append-failure, and disposal fences remain unchanged in meaning.
|
||||
Plan mode owns a plan-specific product package: `@deepseek-ai/dsh-plan-mode` at `packages/plan/plan-mode/`. The durable fact is `plan/mode: { active: boolean }`, folded by `foldPlanMode(events)` with `false` as the empty-log value. `ctx.planMode.get(agent)` returns `{ active, pending? }`, and `set(agent, active)` records the boundary-applied selection. The pre-step, retry, append-failure, and disposal fences preserve the same state-transition ownership.
|
||||
|
||||
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.
|
||||
|
||||
@@ -24,7 +24,7 @@ Sandbox mode and approval policy remain separate enforcement axes. Plan mode nei
|
||||
|
||||
### 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.
|
||||
`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 at initial or continuation pre-step, or on 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.
|
||||
|
||||
|
||||
@@ -6,25 +6,25 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
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。
|
||||
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。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。
|
||||
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（Agent Client Protocol）交互投影，这些要求仍归 plan 功能所有。
|
||||
|
||||
## 决策
|
||||
|
||||
Plan mode 拥有一个 plan 专用产品包：位于 `packages/plan/plan-mode/` 的 `@deepseek-ai/dsh-plan-mode`。持久化事实为 `plan/mode: { active: boolean }`，由 `foldPlanMode(events)` 折叠，空日志值为 `false`。`ctx.planMode.get(agent)` 返回 `{ active, pending? }`，`set(agent, active)` 则记录在边界生效的选择。现有的提示词提交、continuation、重试、追加失败和 dispose（资源释放）栅栏在语义上保持不变。
|
||||
Plan mode 拥有一个 plan 专用产品包：位于 `packages/plan/plan-mode/` 的 `@deepseek-ai/dsh-plan-mode`。持久化事实为 `plan/mode: { active: boolean }`，由 `foldPlanMode(events)` 折叠，空日志值为 `false`。`ctx.planMode.get(agent)` 返回 `{ active, pending? }`，`set(agent, active)` 则记录在边界生效的选择。pre-step、重试、追加失败和 dispose（资源释放）栅栏保留相同的状态转换归属。
|
||||
|
||||
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。
|
||||
|
||||
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。
|
||||
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) 取代了那个协议投影，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。
|
||||
`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初始或续步 pre-step、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。
|
||||
|
||||
@@ -32,7 +32,7 @@ Plan mode 拥有一个 plan 专用产品包：位于 `packages/plan/plan-mode/`
|
||||
|
||||
`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。
|
||||
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，均会拒绝退出，并保留手动 `/plan off` 作为人类退出路径。
|
||||
|
||||
## 删除的接口
|
||||
|
||||
@@ -55,7 +55,7 @@ Plan mode 拥有一个 plan 专用产品包：位于 `packages/plan/plan-mode/`
|
||||
|
||||
**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。
|
||||
**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，需要原样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
|
||||
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-23-acp-automation-only-protocol.md
|
||||
2026-07-23-acp-automation-only-protocol.md: 0fe2fc27a963d21e8a24c1682359ab3bc9e7af48
|
||||
2026-07-23-acp-automation-only-protocol.zh.md: 0a471f0bf1b12e835660cdce2d2a2acd761e1b89
|
||||
2026-07-23-acp-automation-only-protocol.zh.md: bd21696c40811adeb8b8883ef57c30f9321f4ea4
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Agent Note：ACP 作为仅面向自动化的协议
|
||||
# Agent Note: ACP 作为仅面向自动化的协议
|
||||
|
||||
Status: implemented
|
||||
|
||||
@@ -6,11 +6,11 @@ Status: implemented
|
||||
|
||||
## 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（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（reasoning）、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。
|
||||
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测试。
|
||||
|
||||
## 决策
|
||||
|
||||
@@ -18,19 +18,19 @@ ACP 仍有一个有用的职责：另一个 agent（智能体）或自动化控
|
||||
|
||||
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。
|
||||
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登记的同一 agent 对象；外部请求或缺少调用标识的请求会继续委派；RPC 失败则映射为拒绝请求的 `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。
|
||||
断开连接与插件 dispose（资源释放）共享同一个经记忆化的完全停稳边界。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。
|
||||
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、基于同一 agent 对象的权限归属、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豁免覆盖率要求，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
|
||||
|
||||
@@ -42,7 +42,7 @@ ACP 快照套件仍会启动组装后的 ACP 示例，并保留用于锁定后
|
||||
|
||||
**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。
|
||||
**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。只有通过已删除的 UI 方法驱动的场景才离开该套件。
|
||||
|
||||
## 结果
|
||||
|
||||
@@ -50,4 +50,4 @@ ACP 具有适合 agent 与自动化的精简契约，而 TUI 和 Web 拥有面
|
||||
|
||||
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。
|
||||
因此，后端快照测试仍与 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。
|
||||
|
||||
@@ -1,6 +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
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-23-collapse-persistence-flush-state.md
|
||||
2026-07-23-collapse-persistence-flush-state.md: a9b0f6847712f47d46adb6b01c57563033738964
|
||||
2026-07-23-collapse-persistence-flush-state.zh.md: acb9f798d86b4ec41d975d9de23f36080d3d7848
|
||||
2026-07-23-collapse-persistence-flush-state.zh.md: e665d164fb66b2bce97a4f6ded88f2ee07324e61
|
||||
|
||||
@@ -16,7 +16,7 @@ Status: implemented
|
||||
|
||||
初始化只进入现有的按 id 操作链一次，并在占有该轮执行权时调用未串行化的核心操作。该操作链与活跃控制器保持分离，因为公共 `create`、`append`、`load` 调用即使没有 `Session` 对象仍可能发生竞态，依然需要按标识串行执行。
|
||||
|
||||
崩溃修复仅适用于冷态标识。对于活跃标识，`load(id)` 会在等待刷新完成前，先对内存中的权威事件生成快照，再将这些事件与 `SessionState.meta`（即持久化写入实际使用的标头）一同返回；若轮次仍打开，则在不读取或修复存储的情况下拒绝该次加载。冷态加载会先在按 id 操作链内同步占用对应标识，再等待读取已存储前缀或执行修复写入；在这项占用解除前，`session/created` 发布边界会拒绝同 id 活跃会话的发布并将其回滚。HMR 接管仍由 `loadStored` 与协调器的 cwd 检查独立处理，会截断撕裂的存储，但不会闭合权威的活跃轮次。
|
||||
崩溃修复仅适用于冷态标识。对于活跃标识，`load(id)` 会在等待刷新完成前，先对内存中的权威事件生成快照，再将这些事件与 `SessionState.meta`（即持久化写入实际使用的标头）一同返回；若轮次仍打开，则在不读取或修复存储的情况下拒绝该次加载。冷态加载会先在按 id 操作链内同步占用对应标识，再等待读取已存储前缀或执行修复写入；在这项占用解除前，`session/created` 发布边界会拒绝同 id 活跃会话的发布并将其回滚。HMR（热模块替换）接管仍由 `loadStored` 与协调器的 cwd 检查独立处理，会截断撕裂的存储，但不会闭合权威的活跃轮次。
|
||||
|
||||
活跃控制器映射同时也是退役注册表。退役成功时，系统排空并移除其控制器；退役失败时，控制器保留在映射中。后端资源销毁会停止接纳事件，刷新所有仍存在的控制器，等待其余按 id 操作完成，然后关闭后端。无需另设退役集合来重新发现未完成的工作。
|
||||
|
||||
@@ -38,7 +38,7 @@ Status: implemented
|
||||
- 共享后端契约会持久化一个仍打开的活跃轮次，证明 `load` 会拒绝且不会写入合成闭合事件，随后完成该轮次并让其所有者退役，最后重新加载完全相同的已完成轮次。
|
||||
- AgentLoop 回归测试让 `resume()` 与一个仍打开的活跃轮次发生竞态，并证明原有的 agent（智能体）仍能完成该轮次并将其持久化，其间不会注入 `interrupted` 边界。
|
||||
- 一个受控后端会阻塞 `loadStored`，在修复操作持有标识占用期间尝试发布同 id 会话，并证明回滚不会留下残留控制器，之后可以成功恢复一个闭合会话。
|
||||
- 无所有者声明契约会为活跃 `Session` 设置不同的 `createdAt`，并证明活跃加载和之后的冷态加载均返回最初存储的标头。
|
||||
- 无所有者认领契约会为活跃 `Session` 设置不同的 `createdAt`，并证明活跃加载和之后的冷态加载均返回最初存储的标头。
|
||||
|
||||
## 后果
|
||||
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-24-agent-loop-observable-state-machine.md
|
||||
2026-07-24-agent-loop-observable-state-machine.md: a25657c6a41e2c0989db620046f44ea3254be151
|
||||
2026-07-24-agent-loop-observable-state-machine.zh.md: 058d89d3cb9e3d30963f95fda1510ef3c5bf281e
|
||||
2026-07-24-agent-loop-observable-state-machine.md: b1773fdf7fd3bcbea9c6b9a4a0ce37c715bce4ca
|
||||
2026-07-24-agent-loop-observable-state-machine.zh.md: a4ed98117b965b445ec0888c55a1b1b27887532a
|
||||
|
||||
@@ -18,16 +18,16 @@ The public contract exposes four orthogonal state dimensions:
|
||||
|
||||
- Registration lifetime is the `agent/created` to `agent/disposed` interval. Disposal is the terminal registry edge, not an `AgentStatus`.
|
||||
- Whole-agent activity is `AgentStatus = 'idle' | 'running'`. Consecutive turns may share one `running` interval.
|
||||
- A FIFO-backed message progresses from `agent/inbox/enqueue` to exactly one `agent/inbox/dequeue` or `agent/inbox/discard`. Enqueue and dequeue correlate an occurrence by `MessageId` plus its queued-or-steering placement; same-placement repeats retire in FIFO order. The inbox events describe acceptance, claim, and removal rather than turn completion.
|
||||
- A claimed turn passes through prompt admission and zero or more request steps. An automatic retry closes the failed turn and immediately opens another; `agent/settled` reports only the terminal turn in that chain and remains distinct from the whole-agent transition to `status === 'idle'`.
|
||||
- A pending message emits `agent/inbox/inserted` when inserted, then either `agent/inbox/claimed` after an atomic pure-deletion claim or `agent/inbox/discarded` after an ordinary removal. `MessageId` correlates the exact message; durable splice coordinates retain placement and cancellation. Inbox events describe insertion, claim, and discard rather than turn completion.
|
||||
- A claimed turn passes through pre-step entry and zero or more request steps. An automatic retry closes the failed turn and immediately opens another; `agent/settled` reports only the terminal turn in that chain and remains distinct from the whole-agent transition to `status === 'idle'`.
|
||||
|
||||
The loop keeps five machine extension events. `agent/prompt-submit` admits, rewrites, or blocks a claimed prompt. `agent/step` is the single awaited between-steps checkpoint and runs before every request is derived. `agent/request` is the waterfall for the frozen call configuration; the configuration comes only from `await next()`, not from a duplicate positional argument. `agent/request-error` serializes ownership of awaited model-request recovery. `agent/turn-stopping` runs when the turn otherwise has no work left; a listener that needs another step records real steering with `agent.steer()`, and the loop decides from that data after all listeners settle.
|
||||
The loop keeps four machine extension events. `agent/pre-step` decides reject or enter for one exclusive claimed batch and runs before every proposed step. `agent/request` is the waterfall for the frozen call configuration; the configuration comes only from `await next()`, not from a duplicate positional argument. `agent/request-error` serializes ownership of awaited model-request recovery. `agent/turn-stopping` runs when the turn otherwise has no work left; a listener that needs another step records real steering with `agent.steer()`, and the loop decides from that data after all listeners settle.
|
||||
|
||||
Continuation and termination are data rather than returned control enums. Tool calls and accepted steering require another step. A tool result carrying `concludesTurn` ends the tool loop at its step. The loop does not expose general `ContinuationDecision` or terminal-stop return channels.
|
||||
|
||||
A model-request failure closes its step, then enters `agent/request-error` with the exact error, normalized `LlmFailure`, and live turn signal. A listener that owns recovery repairs state, returns `{ kind: 'retry' }`, and stops delegating. The loop closes the failed turn and opens one retry turn over that state without an intervening idle notification; retry is not another step inside the failed turn. `agent/settled` reports the terminal outcome, and `agent/error` remains the live error notification for consumers that report failures independently of turn settlement. The [retry-action decision](2026-07-27-request-error-retry-action.md) supersedes the command-shaped part of this design.
|
||||
|
||||
The event taxonomy removes `agent/pre-step`, `agent/post-step`, `agent/session-prefix`, `agent/step-result`, `agent/turn-continuation`, and `agent/turn-stop`. Durable turn and step boundaries remain session events. Model-facing additions use logged message channels, request configuration uses `agent/request`, response content is recorded as assembled, failed-request recovery uses the `agent/request-error` return action, and end-of-turn continuation uses `agent/turn-stopping` plus steering.
|
||||
The event taxonomy removes the legacy prompt preparation/submission and serial step hooks together with `agent/post-step`, `agent/session-prefix`, `agent/step-result`, `agent/turn-continuation`, and `agent/turn-stop`. The single `agent/pre-step` waterfall owns claimed-message entry. Durable turn and step boundaries remain session events. Model-facing additions use logged message channels, request configuration uses `agent/request`, response content is recorded as assembled, failed-request recovery uses the `agent/request-error` return action, and end-of-turn continuation uses `agent/turn-stopping` plus steering.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
@@ -51,7 +51,7 @@ The inbox lifecycle complements, rather than replaces, the durable session log.
|
||||
|
||||
## Related
|
||||
|
||||
- [Unify agent delivery on send(target × wakeup) and coalesce injected context into user/message](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md)
|
||||
- [Unify agent delivery routing and coalesce injected context into user/message](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md)
|
||||
- [Remove implicit batching from ordinary sends](2026-07-17-one-send-one-turn.md)
|
||||
- [Microkernel event taxonomy](../architecture/2026-06-11-microkernel-event-taxonomy.md)
|
||||
- [Bounded LLM request recovery](../architecture/2026-06-21-bounded-llm-request-recovery.md)
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Agent Note: 围绕可观察状态机收拢 agent loop（智能体循环）事件
|
||||
# Agent Note: 围绕可观察状态机收拢 agent loop 事件
|
||||
|
||||
Status: implemented
|
||||
|
||||
@@ -6,7 +6,7 @@ Status: implemented
|
||||
|
||||
## 问题
|
||||
|
||||
agent loop 曾将其控制流暴露为大量 Cordis 事件。`pre-step` 和 `post-step` 两个独立检查点分列步骤前后，`session-prefix` 和 `step-result` 分别变换请求消息与响应消息，`request-error` 决定失败的请求是否在当前轮次内重试，`turn-continuation` 与 `turn-stop` 则组合相互竞争的继续执行决策。
|
||||
agent loop（智能体循环）曾将其控制流暴露为大量 Cordis 事件。`pre-step` 和 `post-step` 两个独立检查点分列步骤前后，`session-prefix` 和 `step-result` 分别变换请求消息与响应消息，`request-error` 决定失败的请求是否在当前轮次内重试，`turn-continuation` 与 `turn-stop` 则组合相互竞争的继续执行决策。
|
||||
|
||||
即使持久会话日志已经记录了对应的轮次与步骤事实，这些事件仍会将内部阶段公开。它们还混用了两种扩展模型：部分监听器观察边界并发出 agent 命令，另一些监听器则返回由循环解释的控制决策。因此，要理解公开状态机，必须同时还原事件顺序、waterfall（瀑布式事件）优先级和特殊的终止覆盖规则。
|
||||
|
||||
@@ -18,22 +18,22 @@ agent 生命周期、agent 整体活动状态、收件箱条目的进度以及
|
||||
|
||||
- 注册生命周期是从 `agent/created` 到 `agent/disposed` 的区间。dispose（资源释放）是注册表的终止边界，而不是一种 `AgentStatus`。
|
||||
- agent 整体活动状态为 `AgentStatus = 'idle' | 'running'`。连续多个轮次可以共用同一个 `running` 区间。
|
||||
- 由 FIFO 支撑的消息从 `agent/inbox/enqueue` 开始，最终必然进入 `agent/inbox/dequeue` 或 `agent/inbox/discard` 二者之一。enqueue 与 dequeue 通过 `MessageId` 加 queued 或 steering（中途引导）放置方式关联一次消息出现；放置方式相同的重复项按 FIFO 顺序结算。收件箱事件描述接受、领取和移除，而不是轮次完成。
|
||||
- 已领取的轮次经过提示词准入和零个或多个请求步骤。自动重试会关闭失败轮次并立即开启另一个轮次；`agent/settled` 只报告该重试链的终态轮次，且仍不同于 agent 整体转换到 `status === 'idle'`。
|
||||
- 待处理消息插入时会发出 `agent/inbox/inserted`，随后要么在原子纯删除领取后发出 `agent/inbox/claimed`，要么在普通删除后发出 `agent/inbox/discarded`。`MessageId` 关联确切消息；持久 splice 坐标保留 placement 与取消信息。inbox 事件描述插入、领取和丢弃，而不是轮次完成。
|
||||
- 已领取的轮次经过 pre-step 进入决策和零个或多个请求步骤。自动重试会关闭失败轮次并立即开启另一个轮次；`agent/settled` 只报告该重试链的终态轮次，且仍不同于 agent 整体转换到 `status === 'idle'`。
|
||||
|
||||
循环保留五个状态机扩展事件。`agent/prompt-submit` 对已领取的提示词执行准入、改写或阻断。`agent/step` 是步骤之间唯一需要等待的检查点，在每次派生请求前运行。`agent/request` 是冻结调用配置所用的 waterfall；配置只能来自 `await next()`，不再通过重复的位置参数提供。`agent/request-error` 串行确定需要等待的模型请求恢复由谁负责。当轮次原本已经没有剩余工作时，`agent/turn-stopping` 运行；需要再执行一个步骤的监听器使用 `agent.steer()` 记录真实的 steering，循环在所有监听器完成后根据这份数据作出决定。
|
||||
循环保留四个状态机扩展事件。`agent/pre-step` 对独占的已领取批次执行 reject 或 enter 决策，并在每个拟议步骤前运行。`agent/request` 是冻结调用配置所用的 waterfall；配置只能来自 `await next()`，不再通过重复的位置参数提供。`agent/request-error` 串行确定需要等待的模型请求恢复由谁负责。当轮次原本已经没有剩余工作时，`agent/turn-stopping` 运行；需要再执行一个步骤的监听器使用 `agent.steer()` 记录真实的 steering，循环在所有监听器完成后根据这份数据作出决定。
|
||||
|
||||
是否继续和终止执行由数据表达，不再由返回的控制枚举表达。工具调用和已接受的 steering 要求再执行一个步骤。携带 `concludesTurn` 的工具结果会在其所属步骤终止工具循环。循环不再暴露通用的 `ContinuationDecision` 或终止停止返回通道。
|
||||
是否继续和终止执行由数据表达，不再由返回的控制枚举表达。工具调用和已接受的 steering 要求再执行一个步骤。携带 `concludesTurn` 的工具结果会在其所属步骤终止工具循环。循环不再暴露通用的 `ContinuationDecision` 或终止返回通道。
|
||||
|
||||
模型请求失败会先关闭当前步骤，再携带准确错误、标准化 `LlmFailure` 和仍有效的轮次信号进入 `agent/request-error`。负责恢复的监听器修复状态、返回 `{ kind: 'retry' }`，并停止继续委托。循环会关闭失败轮次，并基于该状态开启一个重试轮次，中间不发布空闲通知；重试不是失败轮次内的另一个步骤。`agent/settled` 报告终态结果；对于需要脱离轮次结算单独报告失败的消费方，`agent/error` 仍作为实时错误通知保留。[重试动作决策](2026-07-27-request-error-retry-action.md)取代了本设计中命令形式的部分。
|
||||
模型请求失败会先关闭当前步骤，再携带该错误本身、标准化 `LlmFailure` 和仍有效的轮次信号进入 `agent/request-error`。负责恢复的监听器修复状态、返回 `{ kind: 'retry' }`，并停止继续委托。循环会关闭失败轮次，并基于该状态开启一个重试轮次，中间不发布空闲通知；重试不是失败轮次内的另一个步骤。`agent/settled` 报告终态结果；对于需要脱离轮次结算单独报告失败的消费方，`agent/error` 仍作为实时错误通知保留。[重试动作决策](2026-07-27-request-error-retry-action.md)取代了本设计中命令形式的部分。
|
||||
|
||||
事件分类体系移除了 `agent/pre-step`、`agent/post-step`、`agent/session-prefix`、`agent/step-result`、`agent/turn-continuation` 和 `agent/turn-stop`。持久的轮次与步骤边界仍由会话事件记录。面向模型的新增内容使用有日志记录的消息通道，请求配置使用 `agent/request`，响应内容按组装后的原样记录，失败请求恢复使用 `agent/request-error` 返回动作，轮次结束时是否继续则使用 `agent/turn-stopping` 加 steering 表达。
|
||||
事件分类体系移除了旧的提示词准备／提交与串行 step hook，以及 `agent/post-step`、`agent/session-prefix`、`agent/step-result`、`agent/turn-continuation` 和 `agent/turn-stop`。唯一的 `agent/pre-step` waterfall 负责已领取消息能否进入步骤。持久的轮次与步骤边界仍由会话事件记录。面向模型的新增内容使用有日志记录的消息通道，请求配置使用 `agent/request`，响应内容按组装后的原样记录，失败请求恢复使用 `agent/request-error` 返回动作，轮次结束时是否继续则使用 `agent/turn-stopping` 加 steering 表达。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**保留细粒度事件序列。** 这样可以为每个内部阶段保留专用拦截点，包括仅用于请求的前缀、助手消息改写、步骤后处理、轮次内请求恢复以及终止停止覆盖。但这也会使循环的私有执行顺序成为永久的公开契约，并允许相互重叠的 seam 表达彼此冲突的决策。当前决策接受这些拦截点的缺失，以换取每项受支持的扩展职责仅对应一个边界。
|
||||
**保留细粒度事件序列。** 这样可以为每个内部阶段保留专用拦截点，包括仅用于请求的前缀、助手消息改写、步骤后处理、轮次内请求恢复以及终止覆盖。但这也会使循环的私有执行顺序成为永久的公开契约，并允许相互重叠的 seam 表达彼此冲突的决策。当前决策接受这些拦截点的缺失，以换取每项受支持的扩展职责仅对应一个边界。
|
||||
|
||||
**将 dispose 表示为第三种 `AgentStatus`。** 这样会让仍被持有的句柄得到一个终止状态值，但也会重复表达 `agent/disposed` 已经体现的注册表生命周期。当前决策让 `AgentStatus` 只表示活动中 agent 的状态，并将注册生命周期作为独立维度。
|
||||
**将 dispose 表示为第三种 `AgentStatus`。** 这样会让仍被持有的句柄得到一个终止状态值，但也会重复表达 `agent/disposed` 已经体现的注册表生命周期。当前决策让 `AgentStatus` 只表示 agent 存续期间的活动状态，并将注册生命周期作为独立维度。
|
||||
|
||||
**让 `agent/request-error` 返回重试决策。** 这一替代方案已由[重试动作决策](2026-07-27-request-error-retry-action.md)取代；新决策移除了重复命令，并将决策局限于 waterfall 的返回结果。
|
||||
|
||||
@@ -43,7 +43,7 @@ agent 生命周期、agent 整体活动状态、收件箱条目的进度以及
|
||||
|
||||
可观察状态机更小，也更容易组合：注册生命周期、活动状态、条目进度和终态结算可以分别追踪。尤其是，`agent/settled` 并不意味着 `agent.status === 'idle'`；前者报告一次排空链的终态轮次，`agent/status` 则报告整个 agent 是否处于活动状态。
|
||||
|
||||
插件不再能够改写循环的每个阶段。不再提供仅用于请求的消息前缀、助手消息变换、步骤后检查点、通用的继续执行枚举、通用的终止停止结果或轮次内请求重试。扩展改用剩余的归属明确的通道，而不是重新构造这些阶段。
|
||||
插件不再能够改写循环的每个阶段。不再提供仅用于请求的消息前缀、助手消息变换、步骤后检查点、通用的继续执行枚举、通用的终止结果或轮次内请求重试。扩展改用剩余的归属明确的通道，而不是重新构造这些阶段。
|
||||
|
||||
负责继续执行的插件发布可持久化的 steering，而不是返回未记录到日志中的原因。恢复插件在失败步骤结束后处理错误，并返回显式重试动作。这样，每次尝试都会成为完整轮次，同时异步修复和策略归属集中在一个狭窄的 waterfall 边界。
|
||||
|
||||
@@ -51,8 +51,8 @@ agent 生命周期、agent 整体活动状态、收件箱条目的进度以及
|
||||
|
||||
## 相关内容
|
||||
|
||||
- [统一通过 send(target × wakeup) 交付 agent 消息，并将注入上下文合并到 user/message](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md)
|
||||
- [统一 agent 交付路由，并将注入上下文合并到 user/message](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md)
|
||||
- [移除普通发送中的隐式批处理](2026-07-17-one-send-one-turn.md)
|
||||
- [微内核事件分类体系](../architecture/2026-06-11-microkernel-event-taxonomy.md)
|
||||
- [有界 LLM 请求恢复](../architecture/2026-06-21-bounded-llm-request-recovery.md)
|
||||
- [有界 LLM（大语言模型）请求恢复](../architecture/2026-06-21-bounded-llm-request-recovery.md)
|
||||
- [可重建的请求](../architecture/2026-07-05-reconstructable-requests.md)
|
||||
|
||||
@@ -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
|
||||
2026-07-26-eventsource-parser-for-deepseek-sse.md: e7835bc738b3dec5aefd6011848525f6604e852e
|
||||
2026-07-26-eventsource-parser-for-deepseek-sse.zh.md: 933b993d479026d8f2bd2dc3173abd9e60823806
|
||||
@@ -1,28 +0,0 @@
|
||||
# Agent Note: Replace the hand-rolled SSE parser in llm-deepseek with eventsource-parser
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-26-eventsource-parser-for-deepseek-sse.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
`packages/llm/llm-deepseek/src/sse.ts` hand-implemented Server-Sent Events parsing: a streaming `TextDecoder`, event-block splitting on `\r?\n\r?\n`, `data:` payload extraction and joining, comment/field skipping, the `[DONE]` sentinel, a `STREAM_CLOSED` error on EOF without it, and a flush of a final unterminated event block. The file was ~67 lines with ~108 lines of dedicated tests (`tests/sse.spec.ts`) re-proving SSE spec behavior — UTF-8 split across chunks, CRLF handling, multi-`data:` joining, no-space-after-colon — that a maintained parser already guarantees. Its only consumer is `adapter.ts` (`yield* translate(parseSse(response.body))`).
|
||||
|
||||
This is exactly the surface `eventsource-parser` owns: the de-facto standard SSE parser (it underlies the Vercel AI SDK and the MCP SDK), zero-dependency, actively maintained, and already present in this repo's lockfile transitively via `@modelcontextprotocol/sdk` — so adopting it directly adds no new supply-chain surface in practice.
|
||||
|
||||
## Decision
|
||||
|
||||
`sse.ts` delegates SSE framing to `EventSourceParserStream` from `eventsource-parser/stream`: `parseSse` pipes the response body through `new TextDecoderStream()` then `new EventSourceParserStream()` and keeps only the DeepSeek protocol shim — yield each event's `data`, terminate on `[DONE]`, and throw `LlmError('STREAM_CLOSED')` when the stream ends without the sentinel. All required builtins (`TextDecoderStream`, `pipeThrough`, async-iterable `ReadableStream`) exist at the Node ^22.19 engine floor. The spec-conformance tests are gone; `tests/sse.spec.ts` pins only the `[DONE]`/`STREAM_CLOSED`/EOF contract. `eventsource-parser` is `llm-deepseek`'s second runtime dependency after schemastery. The [twin-adapters note](../architecture/2026-06-13-twin-llm-adapters.md) and the `dsh-llm` JSDoc that branded this adapter "hand-rolled fetch + SSE parsing" now describe it as direct fetch with library-framed SSE.
|
||||
|
||||
The library also strips a leading BOM (the hand-rolled parser would fail to match `data:` after one) and offers `maxBufferSize` hardening the hand-rolled parser lacked.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **Keep the hand-rolled parser.** Defensible under the [twin-adapters decision](../architecture/2026-06-13-twin-llm-adapters.md): the adapter is deliberately the hand-rolled design-verification twin of the pi-ai adapter. But that note's load-bearing distinction is owning the fetch/translate internals versus delegating to a full provider SDK; a ~700-byte SSE micro-parser is transport plumbing, not the design under verification. The twin-adapters note now states that reading explicitly.
|
||||
- **`createParser({onEvent})` callback API instead of the stream.** Works fed by a manual `TextDecoder` loop, but the `pipeThrough` composition deletes more of the hand-rolled code.
|
||||
|
||||
## Consequences
|
||||
|
||||
- The remaining shim only encodes the DeepSeek `[DONE]`/`STREAM_CLOSED` protocol; SSE framing edge cases are eventsource-parser's contract and are no longer re-proven here.
|
||||
- One deliberate robustness deviation is dropped: the hand-rolled parser flushed a final event block that lacked its terminating blank line, so a trailing `data: [DONE]` without `\n\n` still yielded DONE. eventsource-parser is spec-strict and only dispatches on the blank line, so that shape is now `STREAM_CLOSED`. Real providers and `dsh-llm-mock-server` always terminate events properly — the flush was a robustness nicety, not an observed provider shape — and `tests/sse.spec.ts` pins the new truncation verdict for that tail.
|
||||
- The documented "hand-rolled" identity of the twin adapter narrows to the fetch/translate internals; the twin-adapters note was updated in the same change rather than leaving the claim stale.
|
||||
@@ -1,28 +0,0 @@
|
||||
# Agent Note: 用 eventsource-parser 替换 llm-deepseek 中手写的 SSE 解析器
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-26-eventsource-parser-for-deepseek-sse.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
`packages/llm/llm-deepseek/src/sse.ts` 曾手写实现 SSE（Server-Sent Events）解析：一个流式 `TextDecoder`、按 `\r?\n\r?\n` 切分事件块、提取并拼接 `data:` 载荷、跳过注释与其他字段、`[DONE]` 哨兵、在未见哨兵即 EOF 时抛出 `STREAM_CLOSED` 错误，以及对最后一个未终结事件块的 flush。该文件约 67 行，另有约 108 行专属测试（`tests/sse.spec.ts`）重复验证 SSE 规范行为——UTF-8 字符被切分到多个分片、CRLF 处理、多条 `data:` 拼接、冒号后无空格——而这些行为，持续维护的解析器早已有保证。它唯一的消费方是 `adapter.ts`（`yield* translate(parseSse(response.body))`）。
|
||||
|
||||
这恰好是 `eventsource-parser` 负责的接口面：事实标准的 SSE 解析器（Vercel AI SDK 和 MCP SDK 都构建在它之上），零依赖，持续维护，并且已通过 `@modelcontextprotocol/sdk` 作为传递依赖出现在本仓库的 lockfile 中——因此直接采用它实际上不增加新的供应链接触面。
|
||||
|
||||
## 决策
|
||||
|
||||
`sse.ts` 将 SSE 分帧委托给 `eventsource-parser/stream` 的 `EventSourceParserStream`：`parseSse` 把响应 body 依次管道接入 `new TextDecoderStream()` 和 `new EventSourceParserStream()`，只保留 DeepSeek 协议垫层——逐个产出事件的 `data`，遇到 `[DONE]` 终止，流在未见哨兵时结束则抛出 `LlmError('STREAM_CLOSED')`。所需的全部内置能力（`TextDecoderStream`、`pipeThrough`、可异步迭代的 `ReadableStream`）在 Node ^22.19 引擎下限即已存在。规范符合性测试已删除；`tests/sse.spec.ts` 只固定 `[DONE]`/`STREAM_CLOSED`/EOF 契约。`eventsource-parser` 是 `llm-deepseek` 继 schemastery 之后的第二个运行时依赖。曾把该适配器标为「手写 fetch + SSE 解析」的[孪生适配器 Agent Note（agent 决策记录）](../architecture/2026-06-13-twin-llm-adapters.md)与 `dsh-llm` JSDoc，现在将其描述为直接 fetch 加库分帧的 SSE。
|
||||
|
||||
该库还会剥离开头的 BOM（手写解析器在 BOM 之后会无法匹配 `data:`），并提供手写解析器缺少的 `maxBufferSize` 加固能力。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
- **保留手写解析器。** 依据[孪生适配器决策](../architecture/2026-06-13-twin-llm-adapters.md)，这一选择有辩护余地：该适配器有意作为 pi-ai 适配器的手写设计验证孪生体。但那份 Agent Note 起支撑作用的区分在于「自行持有 fetch/translate 内部实现」与「委托给完整的提供方 SDK」；一个约 700 字节的 SSE 微型解析器属于传输层管道，不是被验证的设计本身。孪生适配器 Agent Note 现已明确写出这一解读。
|
||||
- **改用 `createParser({onEvent})` 回调 API 而非流。** 配合手动的 `TextDecoder` 循环可以工作，但 `pipeThrough` 组合方式能删除更多手写代码。
|
||||
|
||||
## 后果
|
||||
|
||||
- 剩下的垫层只编码 DeepSeek 的 `[DONE]`/`STREAM_CLOSED` 协议；SSE 分帧边界情形属于 eventsource-parser 的契约，不再在这里重复验证。
|
||||
- 放弃了一处有意为之的健壮性偏离：手写解析器会 flush 缺少终结空行的最后一个事件块，因此末尾的 `data: [DONE]` 即使没有 `\n\n` 也仍产出 DONE。eventsource-parser 严格遵循规范，只在空行处分发事件，所以这种形态现在是 `STREAM_CLOSED`。真实提供方和 `dsh-llm-mock-server` 总是正确终结事件——该 flush 只是健壮性上的锦上添花，并非实际观测到的提供方形态——`tests/sse.spec.ts` 固定了对该尾部的新截断判定。
|
||||
- 孪生适配器有文档记录的「手写」身份收窄到 fetch/translate 内部实现；孪生适配器 Agent Note 在同一次变更中更新，而不是让声明陈旧下去。
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-27-copyable-transcript-no-gutter-bar.md
|
||||
2026-07-27-copyable-transcript-no-gutter-bar.md: 659b7d2f2f85bf7efe6b1006a2c61a14f3044560
|
||||
2026-07-27-copyable-transcript-no-gutter-bar.zh.md: 5c43169ba352ef1fef73ecf2188a32304aba9c26
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-26-merge-subagent-control-service.md
|
||||
2026-07-26-merge-subagent-control-service.md: ae7214f7c8c28f8a4b7d754839acbc938270f125
|
||||
2026-07-26-merge-subagent-control-service.zh.md: dc05e03c5d38df0317291aee2ffc1395dab8e362
|
||||
@@ -0,0 +1,41 @@
|
||||
# Agent Note: Merge subagent control into the subagent service
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-26-merge-subagent-control-service.zh.md)
|
||||
|
||||
The public operation set is refined by [Intent-named subagent continuation operations](2026-07-27-intent-named-subagent-continuation-operations.md) and again by [Continuable subagents](../feature/2026-07-28-continuable-subagent-conversations.md), which keeps the single merged service while removing provider `resume` dispatch and the Task-backed continuation lifecycle.
|
||||
|
||||
## Problem
|
||||
|
||||
Continuable-child orchestration originally lived in a separate `ctx.subagentControl` service above the raw `ctx.subagents` provider seam. That split kept provider dispatch independent of Tasks and persistence, and gave model and human adapters one orchestration contract. In practice the two services described one capability family, every continuable caller needed both, and the provider-bound delegation tool had to infer policy from `provider.resume` and inspect whether the control service and `send_message` tool happened to be loaded. This made sibling plugin presence decide execution semantics and coupled starting continuable work to an optional follow-up surface.
|
||||
|
||||
## Decision
|
||||
|
||||
`SubagentService` is the only public service. It exposes ordinary `start(name, request)`, Task-backed `startContinuable(spec)`, and intent-named `followup(...)`; provider resume dispatch remains private to its continuation manager. The standalone `@deepseek-ai/dsh-subagent-control` package and `ctx.subagentControl` key are absent; the optional `@deepseek-ai/dsh-tool-subagent-control` package injects `ctx.subagents` directly.
|
||||
|
||||
The merged service and its providers expose one `SubagentError` taxonomy. Stable codes distinguish provider lookup and capability failures from continuation routing, authorization, cancellation, persistence, and delivery failures; the removed service does not retain a separate error class.
|
||||
|
||||
The continuation implementation remains an internal manager rather than expanding the provider registry's core state. `SubagentService` creates it through `ctx.inject(['tasks', 'agents'], ...)`, so the injected Cordis child fiber owns its Task completion listener and teardown effects. Loading the provider registry does not require Tasks or persistence. The manager exists only while Tasks and Agents are available, and each continuation operation resolves session persistence at the point it needs durability. Disposing that fiber cancels and settles active continuations before releasing their associations.
|
||||
|
||||
`startContinuable` remains distinct from raw `start` because it has a different ownership and timing contract: it allocates the durable child id, creates the Task, and returns both ids synchronously while startup continues inside the Task. Raw `start` instead awaits provider publication and transfers a holder-owned run. Folding the method onto `start` through flags or return unions would broaden the low-level contract and create more change than keeping the existing explicit entry.
|
||||
|
||||
Each `@deepseek-ai/dsh-tool-subagent` instance selects `backgroundMode: 'one-shot' | 'continuable'`, defaulting to `one-shot`. This configuration is policy; `provider.resume` is only the capability check for configured continuable mode. A resumable provider can therefore still run one-shot background work. The `send_message` tool is an independent adapter: loading or omitting it neither enables nor disables `startContinuable`.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep the separate service.** This preserves the strongest dependency separation, but every production continuable path composes both services and the extra public key exposes an architectural distinction callers do not need. The internal manager preserves optional Task and persistence dependencies without a second service.
|
||||
|
||||
**Infer continuable mode from `provider.resume`.** Method presence correctly states cold-resume capability but not deployment policy. It forced every resumable provider into continuable background semantics and made missing sibling plugins a runtime error. Explicit tool configuration separates choice from capability.
|
||||
|
||||
**Register continuation access or inspect the follow-up tool.** A registry could tell the delegation tool whether a continuation surface exists, but starting durable work does not require any follow-up adapter. Such a registry would encode UI composition into execution policy and recreate the sibling dependency under another name.
|
||||
|
||||
**Merge raw and continuable starts into one method.** A flag on `start` would return either a published one-shot run or immediate Task and child identities, weakening a simple ownership boundary. Keeping `startContinuable` is the smaller change and preserves both contracts explicitly.
|
||||
|
||||
## Consequences
|
||||
|
||||
- The service topology has one public key and one package fewer while raw provider dispatch remains usable without Tasks or persistence.
|
||||
- Continuable mode fails at provider mount when the configured provider lacks `resume`; missing Tasks, Agents, or persistence still fail at the earliest operation that requires them.
|
||||
- Follow-up delivery remains optional. Deployments may start and collect continuable work through Task tools without exposing `send_message`.
|
||||
- The continuation manager is still Task- and persistence-aware inside the `dsh-subagent` package, so the package declares optional peer dependencies on those services even though ordinary `start` callers do not need them.
|
||||
- Existing continuation races, authorization, durability, cancellation, and settle-then-dispose semantics are unchanged and remain pinned by the migrated `subagent` tests.
|
||||
@@ -0,0 +1,41 @@
|
||||
# Agent Note: 将 subagent 控制合并到 subagent 服务
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-26-merge-subagent-control-service.md) | 中文
|
||||
|
||||
公开操作集合由[以意图命名的 subagent 继续执行操作](2026-07-27-intent-named-subagent-continuation-operations.md)进一步细化，并由[可继续的 subagent](../feature/2026-07-28-continuable-subagent-conversations.md)再次细化——后者保留这一个合并后的服务，同时移除提供方 `resume` 派发和基于 Task 的继续执行生命周期。
|
||||
|
||||
## 问题
|
||||
|
||||
可继续 child 的编排最初位于原始 `ctx.subagents` 提供方 seam 之上的独立 `ctx.subagentControl` 服务中。该拆分使提供方分发与 Task 和持久化无关，并为模型与人工适配器提供统一的编排契约。实践中，两个服务属于同一组功能，每个可继续调用方都需要二者，而绑定提供方的委派工具必须根据 `provider.resume` 推断策略，并检查控制服务与 `send_message` 工具是否碰巧已加载。如此一来，配套插件是否存在会决定执行语义，并将可继续工作的启动耦合到可选的后续操作接口。
|
||||
|
||||
## 决策
|
||||
|
||||
`SubagentService` 是唯一的公开服务。它公开普通的 `start(name, request)`、由 Task 支撑的 `startContinuable(spec)`，以及按意图命名的 `followup(...)`；提供方的 resume 分发仍封装在其继续执行管理器内部。独立的 `@deepseek-ai/dsh-subagent-control` 包（package）和 `ctx.subagentControl` 键均不存在；可选的 `@deepseek-ai/dsh-tool-subagent-control` 包则直接注入 `ctx.subagents`。
|
||||
|
||||
合并后的服务及其提供方公开一套 `SubagentError` 分类体系。稳定错误码把提供方查找失败和功能检查失败，与继续执行路由、鉴权、取消、持久化和送达失败区分开来；已移除的服务不保留单独的错误类。
|
||||
|
||||
继续执行的实现仍是内部管理器，不会扩展提供方注册表的核心状态。`SubagentService` 通过 `ctx.inject(['tasks', 'agents'], ...)` 创建该管理器，因此注入的 Cordis child fiber 拥有自身的 Task 完成监听器和拆卸 effect。加载提供方注册表不要求 Task 或持久化。只有 Task 和 Agent 可用时，该管理器才会存在；每项继续执行操作都在需要持久性时解析会话持久化服务。dispose（资源释放）该 fiber 会先取消并结算活跃的继续执行，再释放其关联。
|
||||
|
||||
`startContinuable` 与底层 `start` 保持分离，因为二者的所有权与时序契约不同：前者分配持久化 child id、创建 Task，并同步返回两个 id，而启动过程继续在 Task 内运行；底层 `start` 则等待提供方发布，并移交一个由持有方负责的 run。若通过标志或返回值联合类型将该方法并入 `start`，会扩大底层契约，改动反而多于保留现有的显式入口。
|
||||
|
||||
每个 `@deepseek-ai/dsh-tool-subagent` 实例都会选择 `backgroundMode: 'one-shot' | 'continuable'`，默认值为 `one-shot`。这项配置表示策略；`provider.resume` 只用于检查所配置的可继续模式是否受提供方支持。因此，可恢复的提供方仍可执行一次性后台工作。`send_message` 工具是独立适配器：加载或省略该工具既不会启用也不会禁用 `startContinuable`。
|
||||
|
||||
## 已考虑的替代方案
|
||||
|
||||
**保留独立服务。** 这样能保持最严格的依赖分离，但每条生产环境中的可继续路径都要组合两个服务，而额外的公开键会暴露调用方并不需要的架构差异。内部管理器无需第二个服务，也能保留可选的 Task 和持久化依赖。
|
||||
|
||||
**根据 `provider.resume` 推断可继续模式。** 方法是否存在可以准确表示从持久化存储恢复的功能，却不能表示部署策略。这会迫使每个可恢复的提供方都采用可继续后台语义，并使配套插件缺失成为运行时错误。显式的工具配置将选择与功能分离。
|
||||
|
||||
**注册继续执行访问入口，或检查后续操作工具。** 注册表可以告诉委派工具继续执行接口是否存在，但启动具备持久性的工作不需要任何后续操作适配器。这样的注册表会把 UI 组合编码进执行策略，并以另一个名称重新建立插件间依赖关系。
|
||||
|
||||
**将底层启动与可继续启动合并为一个方法。** `start` 上的标志会使该方法返回已发布的一次性 run，或立即返回 Task 和 child 标识，从而削弱简单的所有权边界。保留 `startContinuable` 改动更小，也能明确保留两项契约。
|
||||
|
||||
## 影响
|
||||
|
||||
- 服务拓扑少了一个公开键和一个包，同时底层提供方分发仍可在没有 Task 或持久化时使用。
|
||||
- 配置的提供方缺少 `resume` 时，可继续模式会在提供方挂载阶段失败；缺少 Task、Agent 或持久化时，仍会在需要它们的最早操作处失败。
|
||||
- 后续消息投递仍为可选功能。部署可以通过 Task 工具启动并收集可继续工作，而不公开 `send_message`。
|
||||
- `dsh-subagent` 包内的继续执行管理器仍然感知 Task 和持久化，因此该包会将这些服务声明为可选的对等依赖（peer dependency），即使普通的 `start` 调用方并不需要它们。
|
||||
- 现有的继续执行竞态、授权、持久性、取消及先结算再 dispose 的语义均保持不变，并继续由迁移后的 `subagent` 测试固定。
|
||||
@@ -1,37 +0,0 @@
|
||||
# Agent Note: Replace tool-web's regex HTML-to-markdown converter with turndown
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-26-turndown-for-tool-web-html-markdown.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
`dsh-tool-web`'s `src/html.ts` (~86 lines, ~40 lines of dedicated tests; deleted by this change) converted fetched HTML to markdown with regexes: strip script/style/noscript/comments, convert `<a>`/`<h1-6>`/`<li>`, decode numeric entities plus a 12-entry named-entity table, collapse whitespace. The module's own JSDoc said "A richer converter can replace it without changing the seam or tool schema", and the README's Known Limitations documented it as "a minimal regex converter, not an HTML parser — tables, images, and nested formatting are lost." The [web capability seam note](../architecture/2026-06-24-web-capability-seam.md) assigns HTML→markdown to this package as presentation, so the swap point was exactly here. The converter's output is model-visible on every fetched HTML page; no keyless snapshot exercised `web_fetch`, so no expected outputs pinned it.
|
||||
|
||||
## Decision
|
||||
|
||||
`packages/web/tool-web/src/fetch.ts` owns a module-level [`turndown`](https://github.com/mixmark-io/turndown) instance (`headingStyle: 'atx'`, `codeBlockStyle: 'fenced'`, `bulletListMarker: '-'` — fixed model-facing presentation, not deployment tunables) with `@joplin/turndown-plugin-gfm`'s composite `gfm` plugin for tables/strikethrough and `remove(['script', 'style', 'noscript'])` replacing the old wholesale drops. `formatFetchOutput` limits both the source prefix converted synchronously and the complete rendered output with `fetchMaxOutputChars` (default 200,000), so a custom provider cannot make conversion work unbounded before the output cap applies. The HTML arm then guards conversion twice: a conservative linear lexical pass treats comment contents conservatively, skips raw-text bodies, honors quoted tag text, and passes a body through raw when its stack crosses 512 levels; a try/catch also falls back to raw HTML when turndown rejects markup the guard cannot model. The GFM cell rule is overridden to ignore `colspan`, which Markdown cannot represent, rather than letting an untrusted numeric attribute synthesize arbitrary empty cells. `html.ts` and its conversion tests are deleted; the source/output bounds, fallback, and status-header/truncation-footer formatting are tested in `tests/tool-web.spec.ts`, and the README's Known Limitations trades the regex-converter caveat for the bounded degradation cases. The gfm plugin ships no types; `src/turndown-plugin-gfm.d.ts` declares the one imported export over `@types/turndown` (a devDependency).
|
||||
|
||||
The dependency-weight question the proposal flagged resolves in favor of the swap: `@deepseek-ai/dsh-tool-web` is in the single-file-executable closure ([single-exe note](../architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md)), and the exe's asset globs would pack ~7.9 MB of the three packages as published — but ~6 MB of that is `@mixmark-io/domino`'s test corpus (`test/**`), with runtime `lib/` at ~550 KB against a ~174 MB artifact, under 0.5% either way.
|
||||
|
||||
## Snapshot coverage
|
||||
|
||||
The previously-missing keyless `web_fetch` snapshot ships with the change as the acp-agent scenario `web-fetch`: `examples/acp-agent/web.cordis.yml` composes the web seam, the real `dsh-web-fetch-local` provider, `tool-web` with `search: false`, and `web-fetch-fixture-server.mjs` — a loopback HTTP fixture on a fixed port (the fetched URL is part of the recorded transcript) serving deterministic HTML with named entities, a GFM table, and nested formatting. Recording and keyless replay both drive the real HTTP fetch and conversion; the pinned tool result is the turndown output, and the scenario pins the `web` header class (the `web_fetch` schema and guidance).
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **`@mozilla/readability` + a DOM.** Solves a different problem (content extraction, not conversion) and drags a heavier DOM dependency; the seam only asks for markdown rendering of whatever the fetch returned.
|
||||
- **Keep the regex converter.** It was an explicit v1 placeholder per its own JSDoc; keeping it meant model-visible quality (tables, images, nested formatting) stayed lost for the cost of maintaining bespoke entity tables.
|
||||
- **The minimal `entities`-only variant.** The proposal's fallback position: replace only the entity-decoding third of `html.ts` with the zero-dependency `entities` package, deleting less but avoiding the dependency-weight question. Not taken because the closure math above made the weight immaterial while the full swap deletes the whole hand-rolled converter and its documented quality gaps.
|
||||
- **`turndown-plugin-gfm` (the original) instead of `@joplin/turndown-plugin-gfm`.** The original is unmaintained (last publish 2018); the Joplin fork is current against turndown 7 and actively released.
|
||||
|
||||
## Consequences
|
||||
|
||||
- **Bought**: standards-based model-visible markdown — ordinary tables, images, strikethrough, nested emphasis, fenced code blocks, and the complete named-entity set — plus the deletion of the bespoke converter and its entity tables.
|
||||
- **Paid**: two runtime dependencies (`turndown` → `@mixmark-io/domino`) enter tool-web and therefore the exe closure (~550 KB of runtime code as measured above); overlong input is converted only through a bounded prefix, pathological nesting falls back to raw HTML, and spanning table cells are flattened because GFM has no corresponding syntax.
|
||||
- Model-visible output changed on every fetched HTML page; nothing pinned the old output, and the new snapshot pins the new one.
|
||||
|
||||
## Testing
|
||||
|
||||
- `packages/web/tool-web/tests/tool-web.spec.ts` covers the turndown conversion surface (entities, links, tables, nesting, script/style/noscript removal), ignored table spans, source-prefix and complete-output bounds, fast raw-HTML passthrough for deep or deceptively closed nesting, linear handling of malformed tags, the residual converter-throw fallback, and exact and tiny output budgets; per-file coverage on the package src is 100%.
|
||||
- The `web-fetch` acp-agent snapshot pins the assembled behavior keylessly end to end (real Loader composition, real HTTP fetch, real conversion).
|
||||
@@ -1,37 +0,0 @@
|
||||
# Agent Note: 用 turndown 替换 tool-web 的正则 HTML 转 markdown 转换器
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-26-turndown-for-tool-web-html-markdown.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
`dsh-tool-web` 的 `src/html.ts`（约 86 行，另有约 40 行专属测试；已由本变更删除）曾用正则表达式把抓取到的 HTML 转成 markdown：剥离 script、style、noscript 标签与注释，转换 `<a>`/`<h1-6>`/`<li>`，解码数字实体外加一张 12 项的命名实体表，并折叠空白。该模块自身的 JSDoc 写明「A richer converter can replace it without changing the seam or tool schema」，README 的 Known Limitations 章节也把它记载为「a minimal regex converter, not an HTML parser — tables, images, and nested formatting are lost」。[web 能力 seam 决策记录](../architecture/2026-06-24-web-capability-seam.md)把 HTML 转 markdown 作为呈现职责划归本包（package），因此替换点恰好就在这里。每个抓取到的 HTML 页面上，该转换器的输出都对模型可见；此前没有任何无密钥快照执行到 `web_fetch`，因此没有预期输出固定它的行为。
|
||||
|
||||
## 决策
|
||||
|
||||
`packages/web/tool-web/src/fetch.ts` 持有一个模块级 [`turndown`](https://github.com/mixmark-io/turndown) 实例（`headingStyle: 'atx'`、`codeBlockStyle: 'fenced'`、`bulletListMarker: '-'`——固定的面向模型呈现方式，不是部署可调项），配合 `@joplin/turndown-plugin-gfm` 的组合 `gfm` 插件提供表格／删除线支持，并用 `remove(['script', 'style', 'noscript'])` 替代旧实现的整体剥离。`formatFetchOutput` 通过 `fetchMaxOutputChars`（默认 200,000）同时限制同步转换的源前缀和完整渲染输出，因此自定义提供方无法在输出上限生效前造成无界的转换工作。随后，HTML 分支对转换做双重防护：保守的线性词法扫描会保守处理注释内容，跳过原始文本元素的内容，正确处理标签内的引号文本，并在栈深超过 512 层时将主体作为原始 HTML 直接透传；当 turndown 拒绝守卫无法建模的标记时，try/catch 同样回退为原始 HTML。GFM 单元格规则被覆写为忽略 `colspan`；Markdown 无法表示它，这也避免了不受信任的数值属性凭空合成任意数量的空单元格。`html.ts` 及其转换测试已删除；源／输出上限、回退以及状态头／截断页脚格式化均在 `tests/tool-web.spec.ts` 中有测试覆盖，README 的 Known Limitations 用有界降级情形替换了正则转换器警示。gfm 插件不带类型声明；`src/turndown-plugin-gfm.d.ts` 基于 `@types/turndown`（devDependency）声明了唯一被导入的导出。
|
||||
|
||||
提案标记的依赖体积问题的裁决结果支持替换：`@deepseek-ai/dsh-tool-web` 在单文件可执行文件闭包内（[single-exe 决策记录](../architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md)），可执行文件的资产 glob 会把这三个包按发布原样打入约 7.9 MB——但其中约 6 MB 是 `@mixmark-io/domino` 的测试语料（`test/**`），运行时 `lib/` 仅约 550 KB，相对约 174 MB 的产物，两种口径都不到 0.5%。
|
||||
|
||||
## 快照覆盖
|
||||
|
||||
此前缺失的无密钥 `web_fetch` 快照随本变更以 acp-agent 场景 `web-fetch` 落地：`examples/acp-agent/web.cordis.yml` 组合了 web seam、真实的 `dsh-web-fetch-local` 提供方、`search: false` 的 `tool-web`，以及 `web-fetch-fixture-server.mjs`——一个固定端口（抓取的 URL 是录制 transcript（文本记录）的一部分）上的回环 HTTP fixture，提供包含命名实体、GFM 表格与嵌套格式的确定性 HTML。录制与无密钥回放都驱动真实的 HTTP 抓取与转换；固定住的工具结果就是 turndown 的输出，该场景同时固定 `web` header 类（`web_fetch` 的 schema 与指引）。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
- **`@mozilla/readability` 加一个 DOM。** 它解决的是另一个问题（内容提取，而非格式转换），还会拖入更重的 DOM 依赖；这个 seam 只要求把抓取返回的内容渲染成 markdown。
|
||||
- **保留正则转换器。** 按其自身 JSDoc 的说法，它本来就是明确的 v1 占位实现；保留它意味着模型可见的质量（表格、图片、嵌套格式）继续缺失，代价还是维护一套自制实体表。
|
||||
- **仅引入 `entities` 的最小变体。** 提案中的退守方案：只用零依赖的 `entities` 包替换 `html.ts` 中的实体解码部分，删得更少但完全避开依赖体积问题。未采纳：上述闭包测算表明体积无关紧要，而完整替换能删掉整个手写转换器及其记录在案的质量缺口。
|
||||
- **用原版 `turndown-plugin-gfm` 而非 `@joplin/turndown-plugin-gfm`。** 原版已无人维护（最后发布于 2018 年）；Joplin 分叉与 turndown 7 保持同步并持续发布。
|
||||
|
||||
## 后果
|
||||
|
||||
- **收益**：基于标准的模型可见 markdown——普通表格、图片、删除线、嵌套强调、围栏代码块以及完整的命名实体集——并删除了自制转换器及其实体表。
|
||||
- **代价**：两个运行时依赖（`turndown` → `@mixmark-io/domino`）进入 tool-web 进而进入可执行文件闭包（如上实测约 550 KB 运行时代码）；超长输入只转换有界前缀，病态嵌套回退为原始 HTML，跨列表格单元格会被展平，因为 GFM 没有对应语法。
|
||||
- 每个抓取到的 HTML 页面上模型可见的输出都已变化；旧输出本无任何固定，新快照固定了新输出。
|
||||
|
||||
## 测试
|
||||
|
||||
- `packages/web/tool-web/tests/tool-web.spec.ts` 覆盖 turndown 转换面（实体、链接、表格、嵌套、script/style/noscript 移除）、被忽略的表格跨列、源前缀与完整输出上限、深层或带欺骗性闭合嵌套的快速原始 HTML 透传、畸形标签的线性处理、残余的转换器抛错回退，以及恰好达到上限和极小的输出预算；该包 src 的逐文件覆盖率为 100%。
|
||||
- acp-agent 的 `web-fetch` 快照无密钥地端到端固定组装后的行为（真实 Loader 组合、真实 HTTP 抓取、真实转换）。
|
||||
@@ -1,34 +0,0 @@
|
||||
# Agent Note: Copyable TUI transcript without gutter bars
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-27-copyable-transcript-no-gutter-bar.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The TUI grouped user prompts and tool cards behind a colored left-gutter bar (`▌ `) prepended to every body line, and indented assistant and system blocks by one column. Both are per-line prefixes: a terminal mouse drag-select over the scrollback captures the leading `▌ ` or the leading space on each line, so copy-paste of a message, a tool's output, or a code block pulls in decoration the user must strip by hand. The bar was the transcript's only per-message separator, so it could not simply be dropped without another way to tell messages apart.
|
||||
|
||||
## Decision
|
||||
|
||||
The scrollback carries no per-line prefix. Messages are separated only by a bold, underlined role header in the role color and blank-line spacing, both of which the terminal already inserts around each block. The underline gives each role a distinct visual band without a background fill, so it reads on any terminal theme and never enters the clipboard:
|
||||
|
||||
- User and steering prompts (`UserMessageComponent`) are a plain `Container`: a bold, underlined accent `You` / `Steering` header line (via the shared `messageHeader` helper), then the prompt body at column 0.
|
||||
- Assistant blocks render a bold, underlined `Assistant` header, then reasoning and text at column 0, with the timing line at the end of the block (the former `paddingX = 1` indent is gone).
|
||||
- Tool cards drop the `GutterBox` wrapper. The card status (pending / error / success) colors the whole title line — the status glyph (`◌` / `✕` / `✓`) plus the title text share one color, bold and underlined to match the role headers — instead of a colored bar beside an uncolored title. The body renders unprefixed; body lines still pass through `Text` at the terminal width so overlong raw tool output wraps rather than overflowing.
|
||||
- The `GutterBox` class is deleted; nothing else used it.
|
||||
|
||||
A drag-select over any of these regions now copies exactly the message text.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **Keep the bar only on user messages, drop it on tool cards** — leaves tool output, the most-copied region, still polluted. Rejected: the goal is a wholly copyable transcript.
|
||||
- **A single top rule or bar on the header line only** — the body copies clean, but selecting the header still captures a glyph, and it reintroduces a decoration character for no distinguishing gain over the underlined role header.
|
||||
- **Indent grouped bodies instead of a bar** — leading spaces still enter the clipboard, so it does not solve the copy problem; explicitly ruled out.
|
||||
- **A filled background band on the header** (reverse video, or a 256-color muted background) — gives each role a strong color block, but the saturated ANSI fill reads as too heavy and the 256-color shades are fixed rather than theme-remapped. The underline gives per-role distinction with a far lighter footprint.
|
||||
|
||||
## Consequences
|
||||
|
||||
- Copy-paste from the scrollback is clean with no user post-processing. This was the motivating win.
|
||||
- The transcript is flatter than the gutter-bar layout, but each role's bold, underlined header in the role color plus blank-line spacing keeps message boundaries clear without any left-edge fill. Tool-card status stays legible through the colored, underlined glyph and title.
|
||||
- Box-drawing borders (`│`) on transient overlays — status panel, model selector, resume list — are untouched. They are not scrollback message content and are rarely copied.
|
||||
- The affected keyless TUI `*.expected.txt` snapshots were re-recorded by fixture replay (no API key needed; the recorded LLM sessions are unchanged, only the render differs). Interactive boot and a round-trip prompt were verified in tmux.
|
||||
@@ -1,34 +0,0 @@
|
||||
# Agent Note: 无 gutter bar 的可复制 TUI transcript
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-27-copyable-transcript-no-gutter-bar.md) | 中文
|
||||
|
||||
## Problem
|
||||
|
||||
TUI 此前把用户提示词和工具卡片分组在一条彩色左侧 gutter bar（`▌ `）之后，该竖条被逐行加在每一行正文前面，并把 assistant 与系统块整体缩进一列。两者都是逐行前缀：在 transcript 上用鼠标框选时，每一行开头的 `▌ ` 或前导空格都会被一并选中，因此复制一条消息、一段工具输出或一个代码块时都会带上装饰字符，用户必须手动清理。该竖条又是 transcript 中唯一的逐条消息分隔标记，所以不能在没有其他区分方式的情况下直接删掉。
|
||||
|
||||
## Decision
|
||||
|
||||
transcript 不再带任何逐行前缀。消息仅通过以角色色渲染的粗体带下划线角色标题和空行分隔，而这两者本就由终端在每个块前后自动插入。下划线让每个角色获得清晰的视觉分带，且无需背景填充，因此在任何终端配色下都可读，也绝不会进入剪贴板：
|
||||
|
||||
- 用户提示词与 steering 提示词（`UserMessageComponent`）改为普通 `Container`：一行粗体带下划线的强调色 `You` / `Steering` 标题（经共享的 `messageHeader` 辅助函数生成），随后是位于第 0 列的提示词正文。
|
||||
- Assistant 块渲染一行粗体带下划线的 `Assistant` 标题，随后 reasoning 与文本均在第 0 列渲染，timing 行位于块末尾（原先的 `paddingX = 1` 缩进已移除）。
|
||||
- 工具卡片去掉 `GutterBox` 包装层。卡片状态（进行中 / 错误 / 成功）对整行标题着色——状态字形（`◌` / `✕` / `✓`）与标题文本共用一种颜色，并同角色标题一样加粗且带下划线——而不再是未着色标题旁的一条彩色竖条。正文无前缀渲染；正文行仍按终端宽度经 `Text` 处理，使过长的原始工具输出换行而非溢出。
|
||||
- `GutterBox` 类被删除；没有其他地方使用它。
|
||||
|
||||
现在对上述任一区域框选，复制得到的正是消息文本本身。
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **仅在用户消息上保留竖条、在工具卡片上去掉** —— 会让最常被复制的工具输出仍然带有污染。已否决：目标是让整个 transcript 都可复制。
|
||||
- **仅在标题行上加一条顶部横线或竖条** —— 正文复制干净，但选中标题时仍会带上一个字形，且相比带下划线的角色标题并未带来额外的区分收益，却重新引入了装饰字符。
|
||||
- **用缩进代替竖条对分组正文缩进** —— 前导空格仍会进入剪贴板，无法解决复制问题；已明确排除。
|
||||
- **在标题上使用填充背景带**（反色，或 256 色柔和背景）—— 能给每个角色一块强烈的色块，但饱和的 ANSI 填充观感过重，且 256 色是固定色而非随主题重映射。下划线以远更轻的方式提供了同样的逐角色区分。
|
||||
|
||||
## Consequences
|
||||
|
||||
- 从 transcript 复制粘贴无需用户做任何后处理。这正是本次改动的核心收益。
|
||||
- transcript 比 gutter bar 布局更扁平，但每个角色以角色色渲染的粗体带下划线标题加空行分隔，无需任何左缘填充即可让消息边界保持清晰。工具卡片状态仍通过彩色带下划线的字形与标题保持可读。
|
||||
- 临时浮层（状态面板、模型选择器、恢复列表）上的制表符边框（`│`）保持不变。它们不属于 transcript 消息内容，且很少被复制。
|
||||
- 受影响的 keyless TUI `*.expected.txt` 快照均通过 fixture 回放重新记录（无需 API 密钥；所记录的 LLM 会话未变，仅渲染不同）。交互式启动与一次往返提示已在 tmux 中验证。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-27-intent-named-subagent-continuation-operations.md
|
||||
2026-07-27-intent-named-subagent-continuation-operations.md: e74d62b7582e92f8e5ce68327a677259c8453d24
|
||||
2026-07-27-intent-named-subagent-continuation-operations.zh.md: dae4dd37fa9950f0b8d1ba6ec5c46b99977a7201
|
||||
@@ -0,0 +1,38 @@
|
||||
# Agent Note: Intent-named subagent continuation operations
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-27-intent-named-subagent-continuation-operations.zh.md)
|
||||
|
||||
The current activation-based realization is owned by [Continuable subagents](../feature/2026-07-28-continuable-subagent-conversations.md). It retains the `followup` operation this record names, returns the accepted `MessageId`, uses the bare `Agent` parameter as exact live-direct-parent authority, and limits provider participation in continuable children to `prepareContinuable`.
|
||||
|
||||
## Problem
|
||||
|
||||
Merging continuable-child orchestration into `ctx.subagents` left provider dispatch and caller intent on the same public service. `resume(name, request)` accepted a descriptor, authorized parent, durable child id, and activation signal that only the internal continuation manager could resolve correctly. `sendMessage(...)` exposed transport wording rather than the `followup` intent already used by `Agent`, and its separate source and signal parameters widened an operation every caller had to use atomically.
|
||||
|
||||
The durability boundary also exposed both `SessionStore.flush()` and `flushRequired()`. They performed the same scoped parallel dispatch and differed only in whether an empty listener snapshot was accepted, so the session interface encoded one consumer's policy as a second operation.
|
||||
|
||||
## Decision
|
||||
|
||||
`SubagentService` separates four execution intents: `start(name, request)` returns an ordinary holder-owned one-shot run; `startContinuable(spec)` establishes a durable child and returns its id plus the accepted initial `MessageId`; `followup(parent, childId, content, { source, signal })` sends later parent content; and `reportFrom(child, content, { delivery, signal })` sends selected child content to its direct parent. `followup` matches `Agent.followup()`, while `SubagentRun.steer()` remains the narrower confirmed live-run capability. The model-facing tools keep their stable `send_message` and `report` names and delegate routing to the corresponding intent methods.
|
||||
|
||||
Caller and provider requests are distinct. `SubagentStartRequest` contains caller-supplied one-shot data; `ResolvedSubagentStartRequest` adds the service-resolved descriptor before `SubagentProvider.start()`. For continuable creation, the manager passes a `ContinuableCreateRequest` to optional `SubagentProvider.prepareContinuable()` and receives detached creation data only. `SubagentService.resume()` and provider resume dispatch are absent: the continuation manager loads the descriptor, authorizes the parent, and owns Agent materialization, prompt delivery, cold resume, and teardown.
|
||||
|
||||
`SessionStore.flush(session)` is the single durability barrier and returns `Promise<boolean>`. It resolves `true` after at least one scoped listener participates successfully, resolves `false` for an empty listener snapshot, and rejects with the first registered listener failure after all listeners settle. Participation cannot identify whether a selected persistence backend stored the state. Ordinary checkpoints may ignore the boolean; the continuation manager also treats its final flush as a best-effort barrier, deliberately ignores participation, logs rejection, and still disposes the child and releases ownership.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep public provider resume dispatch.** No production caller outside the continuation manager owns descriptor lookup, direct-parent authorization, Agent materialization, Activation ownership, and child-first teardown. A public method would expose resolved implementation data without a valid independent intent; providers instead contribute detached first-creation data through `prepareContinuable` and never participate in cold resume.
|
||||
|
||||
**Keep `sendMessage` on the service.** The model tool sends a message, but the service operation represents a follow-up that may steer or cold-resume. `followup` aligns with the structural `Agent` interface and does not promise a particular route.
|
||||
|
||||
**Keep `flushRequired()`.** A second method hides only an empty-listener check. Returning participation from the existing barrier keeps dispatch in one implementation and lets each caller state whether absence is acceptable.
|
||||
|
||||
**Fold ordinary and continuable starts together.** A flag would make one method return either an awaited holder-owned one-shot run or immediate durable child and message identities. Separate intent methods preserve the ownership and timing distinction without a return union.
|
||||
|
||||
## Consequences
|
||||
|
||||
- The Cordis service catalog contains only caller operations; a provider can opt into continuable first creation through `SubagentProvider.prepareContinuable?()` without receiving Agent lifecycle authority or a public resume operation.
|
||||
- Follow-up source and cancellation travel in one options object, matching the intent-helper shape on `Agent` while retaining the existing live-delivery and cold-resume semantics.
|
||||
- Session durability has one barrier operation. Its participation result remains observable, but no continuable-child path treats arbitrary listener participation as proof that a persistence backend stored the state.
|
||||
- The `send_message` and `report` schemas, accepted message identities, `AgentHandle` ownership, durable event vocabulary, and model-visible transcript follow the activation-based realization linked above.
|
||||
@@ -0,0 +1,38 @@
|
||||
# Agent Note: 按意图命名的 subagent 继续执行操作
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-27-intent-named-subagent-continuation-operations.md) | 中文
|
||||
|
||||
当前基于 Activation 的实现由[可继续的 subagent](../feature/2026-07-28-continuable-subagent-conversations.md)负责。它保留本记录命名的 `followup` 操作，返回已接受的 `MessageId`，使用裸 `Agent` 参数作为确切的在线直属父级权限，并将提供方对可继续 child 的参与限制为 `prepareContinuable`。
|
||||
|
||||
## 问题
|
||||
|
||||
将可继续 child 的编排合并到 `ctx.subagents` 后，提供方分发与调用方意图共存于同一个公开服务中。`resume(name, request)` 接受描述符、已鉴权的 parent、持久化 child id 与激活信号，而只有内部继续执行管理器才能正确解析这些数据。`sendMessage(...)` 暴露的是传输层措辞，而不是 `Agent` 已采用的 `followup` 意图；它还将来源与信号拆成独立参数，扩大了操作接口，而每个调用方都必须以原子方式同时使用二者。
|
||||
|
||||
持久性边界还同时公开了 `SessionStore.flush()` 与 `flushRequired()`。二者执行相同的作用域内并行分发，唯一差别是是否接受空的监听器快照，因此会话接口将一个消费方的策略编码为第二项操作。
|
||||
|
||||
## 决策
|
||||
|
||||
`SubagentService` 分离四种执行意图：`start(name, request)` 返回普通的、由持有方负责的 one-shot run；`startContinuable(spec)` 建立持久化 child，并返回其 id 与已接受的初始 `MessageId`；`followup(parent, childId, content, { source, signal })` 发送后续 parent 内容；`reportFrom(child, content, { delivery, signal })` 将选定的 child 内容发送给其直接 parent。`followup` 与 `Agent.followup()` 一致，而 `SubagentRun.steer()` 仍是范围更窄的能力，仅向已确认仍在运行的 run 提供 steering。面向模型的工具保留稳定的 `send_message` 与 `report` 名称，并将路由委托给对应的意图方法。
|
||||
|
||||
调用方请求与提供方请求相互分离。`SubagentStartRequest` 包含调用方提供的 one-shot 数据；`ResolvedSubagentStartRequest` 会在调用 `SubagentProvider.start()` 前加入由服务解析的描述符。创建可继续 child 时，管理器将 `ContinuableCreateRequest` 传给可选的 `SubagentProvider.prepareContinuable()`，且只接收分离的创建数据。`SubagentService.resume()` 与提供方恢复分发均不存在：继续执行管理器加载描述符、对 parent 进行鉴权，并负责 Agent 实体化、提示词投递、冷恢复与 teardown。
|
||||
|
||||
`SessionStore.flush(session)` 是唯一的持久性屏障，并返回 `Promise<boolean>`。至少一个作用域内监听器成功参与后，它解析为 `true`；监听器快照为空时解析为 `false`；所有监听器结算后，如有失败，则以注册顺序最靠前的监听器错误拒绝。参与结果无法表明所选的持久化后端是否已经存储状态。普通检查点可以忽略该布尔值；继续执行管理器同样将最终 flush 视为 best-effort 屏障，有意忽略参与结果，记录拒绝日志，并仍会对 child 执行 dispose（资源释放）并释放所有权。
|
||||
|
||||
## 已考虑的替代方案
|
||||
|
||||
**保留公开的提供方恢复分发。** 继续执行管理器之外，没有任何生产调用方同时负责安全调用所需的描述符查找、直接 parent 鉴权、Agent 实体化、Activation 所有权与 child-first teardown。公开方法会暴露已解析的实现数据，却没有合理的独立调用意图；提供方改为通过 `prepareContinuable` 贡献分离的首次创建数据，且从不参与冷恢复。
|
||||
|
||||
**在服务上保留 `sendMessage`。** 面向模型的工具发送消息，但服务操作表达的是后续操作，既可能对运行中的激活执行 steering，也可能从持久化存储恢复。`followup` 与结构化 `Agent` 接口保持一致，也不承诺特定路由。
|
||||
|
||||
**保留 `flushRequired()`。** 第二个方法只封装了空监听器检查。由现有屏障返回是否有监听器参与，可以让分发只保留一套实现，并让每个调用方自行判定缺少监听器是否可接受。
|
||||
|
||||
**合并普通启动与可继续启动。** 一个标志会让同一方法要么等待由持有方负责的 one-shot run 就绪后返回，要么立即返回持久化 child 与消息标识。按意图拆分的方法无需返回值联合类型即可保留所有权与时序差异。
|
||||
|
||||
## 影响
|
||||
|
||||
- Cordis 服务目录只包含调用方操作；提供方可以通过 `SubagentProvider.prepareContinuable?()` 选择参与可继续 child 的首次创建，但不会获得 Agent 生命周期权限或公开恢复操作。
|
||||
- 后续操作的来源与取消信号通过同一个选项对象传递，与 `Agent` 上按意图命名的辅助方法形态一致，同时保留在线投递与从持久化存储恢复的语义。
|
||||
- 会话持久性只有一个屏障操作。参与结果仍可观测，但任何可继续 child 路径都不会将任意监听器参与视为持久化后端已存储状态的证明。
|
||||
- `send_message` 与 `report` schema、已接受的消息标识、`AgentHandle` 所有权、持久化事件词汇与模型可见的 transcript（文本记录）遵循上文链接的基于 Activation 的实现。
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-27-request-error-retry-action.md
|
||||
2026-07-27-request-error-retry-action.md: 3057b9fa28cf203c9374930fe97421918b4c1a6f
|
||||
2026-07-27-request-error-retry-action.zh.md: bcb4e592f0c3d86f896e279cf0e3ea400741a1bb
|
||||
2026-07-27-request-error-retry-action.md: 18ae9bc4ba26d1ad3cb7d1328d9e9d3e8de8377c
|
||||
2026-07-27-request-error-retry-action.zh.md: 5b36a38b6baed9b954fb04bca9e49c24edf9a969
|
||||
|
||||
@@ -14,7 +14,7 @@ Model-request recovery was decided inside `agent/request-error` but communicated
|
||||
|
||||
The loop reads the action after the waterfall settles, closes the failed turn, and opens one retry turn from durable history. It rechecks the turn signal when consuming the action, so cancellation or disposal during recovery prevents the retry even if a listener returns it afterward. A thrown recovery never produces an action.
|
||||
|
||||
`Agent` and `ReactLoopAgent` expose no `retry()` method. Ordinary new work enters through `send()` and its `followup()`, `steer()`, and `inject()` presets; only a handled model-request failure can open a promptless retry turn.
|
||||
`Agent` and `ReactLoopAgent` expose no `retry()` method. Ordinary new work enters through `followup()`, `steer()`, and `inject()`; only a handled model-request failure can open a promptless retry turn.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
|
||||
@@ -12,13 +12,13 @@ Status: implemented
|
||||
|
||||
`agent/request-error` 返回 `RequestErrorAction`，其中负责处理的动作是 `{ kind: 'retry' }`；默认的 `undefined` 会让失败轮次保持终态。不拥有该失败的监听器调用 `next()`。拥有该失败的监听器执行所有需要等待的修复，然后直接返回重试动作而不继续委托。
|
||||
|
||||
waterfall 结算后，循环读取该动作，关闭失败轮次，并从持久历史开启一个重试轮次。循环在使用该动作时会再次检查轮次信号，因此即使监听器随后返回重试动作，恢复期间发生的取消或资源释放仍会阻止重试。抛出异常的恢复不会产生动作。
|
||||
waterfall 结算后，循环读取该动作，关闭失败轮次，并从持久历史开启一个重试轮次。循环在使用该动作时会再次检查轮次信号，因此即使监听器随后返回重试动作，恢复期间发生的取消或 dispose（资源释放）仍会阻止重试。抛出异常的恢复不会产生动作。
|
||||
|
||||
`Agent` 与 `ReactLoopAgent` 均不暴露 `retry()` 方法。普通新工作通过 `send()` 及其 `followup()`、`steer()` 和 `inject()` 预设进入；只有已处理的模型请求失败才能开启没有提示词的重试轮次。
|
||||
`Agent` 与 `ReactLoopAgent` 均不暴露 `retry()` 方法。普通新工作通过 `followup()`、`steer()` 和 `inject()` 进入；只有已处理的模型请求失败才能开启没有提示词的重试轮次。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**保留 `Agent.retry()` 作为恢复命令。** 运行时防护检查可以将该命令限制在请求错误窗口内，但接口仍会暴露一个没有生产消费方的空闲无提示词再运行操作，循环也仍需通过可变的旁路状态恢复已经由 waterfall 决定的结果。
|
||||
**保留 `Agent.retry()` 作为恢复命令。** 运行时防护检查可以将该命令限制在请求错误窗口内，但接口仍会暴露一个没有生产消费方的空闲无提示词再运行操作，循环也仍需通过可变的旁路状态取回已由 waterfall 承载的决策。
|
||||
|
||||
**返回显式终态动作。** `undefined` 已经表示 waterfall 未处理时的默认值，并可直接通过 `next()` 组合。再添加一个 `{ kind: 'fail' }` 值不会提供不同的行为或归属信息。
|
||||
|
||||
@@ -26,4 +26,4 @@ waterfall 结算后，循环读取该动作，关闭失败轮次，并从持久
|
||||
|
||||
恢复归属、异步修复和重试决策共用一条类型化返回路径。活跃 agent 接口与具体循环不再具有空闲无提示词再运行能力和重试窗口状态。调用方如果不提交后续提示词，就无法重启任意失败的非请求工作；瞬时策略与上下文溢出策略则保留编号重试轮次、从持久历史重建、有限的策略私有预算和取消优先级。
|
||||
|
||||
聚焦的 agent-loop 测试固定了重试链、未处理失败保持终态、恢复失败和取消竞态。llm-retry 与 compact-basic 测试套件固定其策略自有的动作返回，而 ACP、goal-session 和 plan-mode 集成测试固定后继轮次承接。
|
||||
聚焦的 agent-loop 测试固定了重试链、未处理失败保持终态、恢复失败和取消竞态。llm-retry 与 compact-basic 测试套件固定其策略自有的动作返回，而 ACP（Agent Client Protocol）、goal-session 和 plan-mode 集成测试固定后继轮次承接。
|
||||
|
||||
@@ -3,4 +3,4 @@
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-28-local-json-tree-renderer.md
|
||||
2026-07-28-local-json-tree-renderer.md: 5e8e6d6968319ada53e65a48e9e03b297dc0141c
|
||||
2026-07-28-local-json-tree-renderer.zh.md: 221880391da05551d05ad9e3dadb909e2e74fa68
|
||||
2026-07-28-local-json-tree-renderer.zh.md: f3a9faa5f9de264d0b7eadac631f7f4afff166b2
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Agent Note：本地 JSON 树渲染器
|
||||
# Agent Note: 本地 JSON 树渲染器
|
||||
|
||||
Status: implemented
|
||||
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-28-remove-synthetic-log-only-turns.md
|
||||
2026-07-28-remove-synthetic-log-only-turns.md: fc76667924ec839301aad993efd996112c9a6b09
|
||||
2026-07-28-remove-synthetic-log-only-turns.zh.md: 7520c33e2219c5fe7ab7d8da6312247e42cc69b0
|
||||
2026-07-28-remove-synthetic-log-only-turns.md: 720aa6bbe246ea960a61f6d6f67d36d1f85af08a
|
||||
2026-07-28-remove-synthetic-log-only-turns.zh.md: 587165a58ae322192b9ed1a9b3312f61e0b21ce6
|
||||
|
||||
@@ -8,7 +8,7 @@ English | [中文](2026-07-28-remove-synthetic-log-only-turns.zh.md)
|
||||
|
||||
The session store exposed `appendOutOfBand()` so a plugin could publish a late log-only event while no agent turn was running. The method wrapped that event in `turn/start` and `turn/end`, then flushed it. This preserved the old rule that every durable event had to live inside a turn, but it made one identifier mean both a model-loop execution and a persistence-only update.
|
||||
|
||||
That rule was introduced when persistence recovery treated the last `turn/end` as the only committed boundary. The persistence scanners now preserve every valid contiguous event, and crash repair reacts only to an actually open turn. Idle context already uses the same capability by appending `user/message` between turns. Retaining synthetic turns for title updates therefore inflated turn counts, produced execution outcomes for work that never ran the model, and let a late metadata write consume the next turn number.
|
||||
That rule was introduced when persistence recovery treated the last `turn/end` as the only committed boundary. The persistence scanners now preserve every valid contiguous event, and crash repair reacts only to an actually open turn. Retaining synthetic turns for title updates therefore inflated turn counts, produced execution outcomes for work that never ran the model, and let a late metadata write consume the next turn number.
|
||||
|
||||
The generic seam also duplicated domain policy. Its marker map said which plugin events were eligible, while the title capability already owned cancellation, liveness, and stale-result rules. Replacing it with another generic or title-specific append wrapper would preserve the same type indirection for two literal event types.
|
||||
|
||||
@@ -20,7 +20,7 @@ Core session invariants continue to enforce core-owned execution relations: turn
|
||||
|
||||
The title service appends `session/title` directly after its existing service, revision, cancellation, and live-session checks. The bundled model helper appends its literal `session/title-llm-request` record before dispatch. Persistence observes both through the eager `session/event` path and drains them at ordinary checkpoints and lifecycle teardown; neither append forces a flush merely because it is between turns. A fallback, auxiliary request record, or accepted provider title may therefore appear after `turn/end` and before the next `turn/start`. Manual compaction uses the same between-turn capability for a `compact/* { turn: null }` bracket, but explicitly flushes the closed attempt because `/compact` promises durability before releasing queued prompt admission.
|
||||
|
||||
A session fork may end at any stable event position outside an open turn, not only at `turn/end`. This preserves standalone title and context records in a default fork while still rejecting a prefix cut through active execution.
|
||||
A session fork may end at any stable event position outside an open turn, not only at `turn/end`. This preserves standalone title and other plugin-owned log-only records in a default fork while still rejecting a prefix cut through active execution.
|
||||
|
||||
The historical [universal turn-enclosure decision](../../archived/architecture/2026-06-15-turn-enclosure-invariant.md) remains useful only as the reason the synthetic mechanism was introduced. The [context-injection decision](../architecture/2026-07-24-separate-context-injection-from-turn-execution.md) established the current meaning: one turn represents one model-loop execution. The [queued manual compaction decision](../feature/2026-07-30-queued-manual-compaction.md) applies that rule to a durable multi-event bracket and owns its marker and admission semantics.
|
||||
|
||||
|
||||
@@ -8,7 +8,7 @@ Status: implemented
|
||||
|
||||
会话存储曾暴露 `appendOutOfBand()`，让插件可以在没有 agent（智能体）轮次运行时发布延迟到达的纯日志事件。该方法会用 `turn/start` 和 `turn/end` 包住事件，再将其刷写。这保留了「每个持久事件都必须位于轮次内」的旧规则，却让同一个标识符既表示模型循环执行，又表示仅持久化更新。
|
||||
|
||||
引入该规则时，持久化恢复曾将最后一个 `turn/end` 视为唯一的已提交边界。如今，持久化扫描器会保留每个合法且连续的事件，崩溃修复也只处理确实处于开放状态的轮次。空闲上下文早已采用同一机制，在轮次之间追加 `user/message`。因此，为标题更新保留合成轮次会夸大轮次计数、为从未运行模型的工作产生执行结果，还会让延迟到达的元数据写入占用下一个轮次编号。
|
||||
引入该规则时，持久化恢复曾将最后一个 `turn/end` 视为唯一的已提交边界。如今，持久化扫描器会保留每个合法且连续的事件，崩溃修复也只处理确实处于开放状态的轮次。因此，为标题更新保留合成轮次会夸大轮次计数、为从未运行模型的工作产生执行结果，还会让延迟到达的元数据写入占用下一个轮次编号。
|
||||
|
||||
通用 seam 还重复了领域策略。它的标记映射说明哪些插件事件符合条件，而标题功能本就拥有取消、活跃性和陈旧结果处理规则。改用另一个通用或标题专属追加包装层，仍会为两个字面量事件类型保留同一层类型间接性。
|
||||
|
||||
@@ -20,7 +20,7 @@ Status: implemented
|
||||
|
||||
标题服务会在完成既有的服务状态、修订、取消和实时会话检查后，直接追加 `session/title`。随附模型辅助函数会在发起调用前追加其字面量 `session/title-llm-request` 记录。持久化通过尽快处理的 `session/event` 路径观察两者，并在常规检查点与生命周期 teardown 时排空；二者都不会仅因为位于轮次之间就强制 flush。因此，回退标题、辅助请求记录或已接受的提供方标题可以出现在 `turn/end` 之后、下一个 `turn/start` 之前。手动压缩（compaction）利用同一项轮次间能力记录 `compact/* { turn: null }` 标记对，但会显式 flush 已闭合的尝试，因为 `/compact` 承诺在释放排队提示词接纳预留前完成持久化。
|
||||
|
||||
会话 fork 可以结束于开放轮次之外的任意稳定事件位置，而不限于 `turn/end`。这样，默认 fork 会保留独立标题和上下文记录，同时仍拒绝在活跃执行过程中截断前缀。
|
||||
会话 fork 可以结束于开放轮次之外的任意稳定事件位置，而不限于 `turn/end`。这样，默认 fork 会保留独立标题和其他插件所属的纯日志记录，同时仍拒绝在活跃执行过程中截断前缀。
|
||||
|
||||
历史上的[通用轮次封闭决策](../../archived/architecture/2026-06-15-turn-enclosure-invariant.md)如今只适合用于解释为何曾引入合成机制。[上下文注入决策](../architecture/2026-07-24-separate-context-injection-from-turn-execution.md)确立了当前语义：一个轮次表示一次模型循环执行。[排队手动压缩决策](../feature/2026-07-30-queued-manual-compaction.md)将该规则应用于持久多事件标记对，并拥有其标记与接纳语义。
|
||||
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-29-shared-base-config-overlays.md
|
||||
2026-07-29-shared-base-config-overlays.md: ee642cbc786bef708791fb58e655c5a3f0e9c4e7
|
||||
2026-07-29-shared-base-config-overlays.zh.md: b7fc9c6b121b8d0eb94d734af6bda6df45e25b1d
|
||||
2026-07-29-shared-base-config-overlays.md: 9fae9775e0b37acc99ce2eb5fc630e6f4f52bf1d
|
||||
2026-07-29-shared-base-config-overlays.zh.md: 20cc3e522c369315a1fafc84778fced0260581d0
|
||||
|
||||
@@ -22,7 +22,7 @@ Precedence is list order, last write winning per row: base, then the surface ove
|
||||
|
||||
`--config <path>` now applies an overlay **instead of** the personal overlay, so a demo or test tree never inherits the user's provider and model. `--config-replace <path>` boots a file as the entire tree, bypassing base, surface overlay, and personal overlay alike; that is what the old `--config` did, so trees like `examples/web-cordis` moved to the new flag. Both flags survive the `/resume` execve handoff, or resuming would silently change the agent.
|
||||
|
||||
A patch replaces its target row's whole `config` rather than merging, which shapes the split: a row whose value differs per surface lives in the overlays, never in the base, so no row is patched by three layers at once. Session identity therefore cannot ride a config key at all — it moved to `dsh-agent-loop`'s `CONFIGURED_AGENT_IDENTITIES_KEY`, as [the launcher-owned identity note](../architecture/2026-07-28-launcher-owned-resume-identity.md) now records.
|
||||
A patch replaces its target row's whole `config` rather than merging, which shapes the split: a row whose value differs per surface lives in the overlays, never in the base, so no row is patched by three layers at once. Session identity therefore cannot ride a config key at all — it moved to `dsh-agent-loop`'s `CONFIGURED_AGENT_IDENTITIES_KEY`, as the launcher-owned identity record documented.
|
||||
|
||||
`examples/tui-agent`, `examples/cordis-agent`, `examples/code-mode`, and `packages/examples/tui-demo` are deleted. The TUI tests move to `apps/cli/tests/`, the cordis-toolset e2e to `packages/cordis/tool-cordis/tests/`, and the supported Code Mode demo remains the ACP overlay at `examples/acp-agent/code-mode.cordis.yml`.
|
||||
|
||||
|
||||
@@ -22,7 +22,7 @@ Status: implemented
|
||||
|
||||
`--config <path>` 现在应用一个 overlay 来**取代**个人 overlay，因此 demo 或测试用的树绝不会继承用户的 provider 与 model。`--config-replace <path>` 则把某个文件作为整棵树启动，同时绕过 base、surface overlay 与个人 overlay；这正是旧 `--config` 的行为，所以像 `examples/web-cordis` 这样的树改用了新 flag。两个 flag 都会在 `/resume` 的 execve 交接中保留，否则 resume 会静默更换 agent。
|
||||
|
||||
patch 会整体替换目标配置项的 `config` 而不合并，这决定了拆分方式：取值因 surface 而异的配置项住在 overlay 中，绝不住在 base 里，从而没有任何配置项会被三层同时 patch。因此会话身份根本不能经由配置键传递——它迁移到了 `dsh-agent-loop` 的 `CONFIGURED_AGENT_IDENTITIES_KEY`，如[启动器持有身份的 note](../architecture/2026-07-28-launcher-owned-resume-identity.md) 现在所记录。
|
||||
patch 会整体替换目标配置项的 `config` 而不合并，这决定了拆分方式：取值因 surface 而异的配置项住在 overlay 中，绝不住在 base 里，从而没有任何配置项会被三层同时 patch。因此会话身份根本不能经由配置键传递——它迁移到了 `dsh-agent-loop` 的 `CONFIGURED_AGENT_IDENTITIES_KEY`，正如启动器持有身份的记录所述。
|
||||
|
||||
`examples/tui-agent`、`examples/cordis-agent`、`examples/code-mode` 与 `packages/examples/tui-demo` 均被删除。TUI 测试迁往 `apps/cli/tests/`，cordis 工具集的 e2e 迁入 `packages/cordis/tool-cordis/tests/`，受支持的 Code Mode demo 则保留为 `examples/acp-agent/code-mode.cordis.yml` 中的 ACP overlay。
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-30-sidebar-resize-without-visible-pill.md
|
||||
2026-07-30-sidebar-resize-without-visible-pill.md: cc41898990fa23ff2937140186a8324217911d2e
|
||||
2026-07-30-sidebar-resize-without-visible-pill.zh.md: 9f1f521df2848b15f5015719bfa6f0e0e9b7be0c
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-30-private-agent-send.md
|
||||
2026-07-30-private-agent-send.md: 43353c309f98eab1bff91b8fbe0c9cfaa2bbc69b
|
||||
2026-07-30-private-agent-send.zh.md: 49f7c999ec60c7305bdca570a4d59039c7c1923b
|
||||
@@ -0,0 +1,27 @@
|
||||
# Agent Note: Keep agent routing private
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-30-private-agent-send.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The public `Agent.send()` method exposed the concrete loop's routing matrix even though production callers use only the semantic `followup()`, `steer()`, and `inject()` operations. Its fourth combination, `next-turn` with `wakeup: false`, had no consumer beyond tests. Keeping that latent capability public also required alternate `Agent` implementations and test fakes to accept implementation-level routing policy.
|
||||
|
||||
## Decision
|
||||
|
||||
`Agent` exposes `followup()`, `steer()`, and `inject()` as its complete delivery contract. `ReactLoopAgent` keeps a private `send()` helper that shares routing mechanics among those methods, while `SendTarget` and `SendOptions` are no longer exported from `dsh-agent`.
|
||||
|
||||
The public interface cannot queue a turn without waking the driver. A follow-up always requests execution, steering requests the nearest step, and injection supplies model-facing context without requesting execution. This partially supersedes the public-surface portion of the [unified delivery decision](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md) while retaining its internal routing and unified `user/message` representation.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep the routing matrix public.** This preserves the unused quiet-queue combination, but exposes mechanism instead of caller intent and imposes it on every alternate driver.
|
||||
|
||||
**Add a public quiet-queue method.** A named method would be clearer than raw routing flags, but no production workflow currently needs work that remains parked until an unrelated delivery wakes it.
|
||||
|
||||
## Consequences
|
||||
|
||||
Plugins choose among three semantic operations instead of constructing routing options. Alternate drivers and structural test fakes implement a smaller contract, and the Cordis API catalog no longer advertises `send`, `SendTarget`, or `SendOptions`.
|
||||
|
||||
The removed quiet-queue capability can return only with a named consumer and explicit lifecycle semantics. `cancel({ keepInbox: true })` still preserves work already pending through the supported delivery paths.
|
||||
@@ -0,0 +1,27 @@
|
||||
# Agent Note: 将 agent 路由保留为私有实现
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-30-private-agent-send.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
公开的 `Agent.send()` 方法暴露了实体循环的路由矩阵，但生产调用方只使用语义明确的 `followup()`、`steer()` 和 `inject()` 操作。第四种组合，即 `next-turn` 配合 `wakeup: false`，除测试外没有消费方。将这项潜在能力保留为公开接口，还会迫使其他 `Agent` 实现和测试替身接受实现层的路由策略。
|
||||
|
||||
## 决策
|
||||
|
||||
`Agent` 将 `followup()`、`steer()` 和 `inject()` 作为完整的交付契约公开。`ReactLoopAgent` 保留私有的 `send()` 辅助方法，供这三个方法共用路由机制；`dsh-agent` 不再导出 `SendTarget` 和 `SendOptions`。
|
||||
|
||||
公开接口无法在不唤醒驱动器的情况下让一个轮次入队。`followup()` 始终请求执行，`steer()` 请求最近的步骤，`inject()` 则提供面向模型的上下文而不请求执行。本决策部分取代[统一交付决策](../architecture/2026-07-22-unified-send-and-coalesced-user-messages.md)中关于公开接口的内容，同时保留其内部路由与统一的 `user/message` 表示。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**让路由矩阵保持公开。** 这会保留未使用的无唤醒排队组合，但也会暴露机制而非调用方意图，并要求每个替代驱动器都支持该机制。
|
||||
|
||||
**添加公开的无唤醒排队方法。** 使用具名方法会比原始路由标志更清晰，但目前没有生产工作流需要让工作持续处于等待状态，直到无关的交付将其唤醒。
|
||||
|
||||
## 后果
|
||||
|
||||
插件从三种语义操作中选择，不再自行构造路由选项。其他驱动器和结构型测试替身只需实现更小的契约，Cordis API 目录也不再列出 `send`、`SendTarget` 或 `SendOptions`。
|
||||
|
||||
只有出现明确的消费方并定义显式的生命周期语义后，才能恢复已移除的无唤醒排队能力。`cancel({ keepInbox: true })` 仍会保留已通过受支持交付路径进入待处理状态的工作。
|
||||
@@ -1,25 +0,0 @@
|
||||
# Agent Note: Sidebar resize without a visible pill
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-30-sidebar-resize-without-visible-pill.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The AppFrame exposed identical floating pills on both column borders. The left pill added unnecessary visual weight beside primary navigation, but the sidebar's resize interaction remains useful.
|
||||
|
||||
## Decision
|
||||
|
||||
AppFrame keeps the sidebar's 8px resize hit strip, `col-resize` cursor, pointer capture, animation-frame throttling, and width updates, but does not generate the sidebar handle's pill pseudo-element. The details boundary retains both its hit strip and floating pill.
|
||||
|
||||
The layout component test continues to pin sidebar dragging and both handles' collapse lifecycle. A keyless browser scenario reads the generated pseudo-elements from the shipped composition and drags the invisible sidebar boundary to prove the interaction remains live.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Remove the sidebar drag interaction with the pill.** Rejected because the requested change is visual; removing a working geometry control would unnecessarily narrow the interaction.
|
||||
|
||||
**Keep the pill but reduce its emphasis.** A smaller or lower-contrast pill still leaves an unwanted object on the sidebar boundary.
|
||||
|
||||
## Consequences
|
||||
|
||||
The sidebar boundary is visually quiet while pointer resizing remains available from the boundary and retains the resize cursor. Unlike the details control, that interaction has no visible pill.
|
||||
@@ -1,25 +0,0 @@
|
||||
# Agent Note: 侧边栏缩放不显示胶囊
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-30-sidebar-resize-without-visible-pill.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
AppFrame 在两个栏位边界都显示相同的浮动胶囊。左侧胶囊在主导航旁增加了不必要的视觉负担，但侧边栏的缩放交互仍有用。
|
||||
|
||||
## 决策
|
||||
|
||||
AppFrame 保留侧边栏宽 8px 的缩放命中条带、`col-resize` 光标、指针捕获、动画帧节流和宽度更新，但不再生成侧边栏手柄的胶囊形伪元素。详情栏边界同时保留命中条带和浮动胶囊。
|
||||
|
||||
布局组件测试继续固定侧边栏拖动行为，以及两个手柄随面板折叠时的生命周期。一个无密钥浏览器场景读取实际交付组合所生成的伪元素，并拖动不可见的侧边栏边界，证明该交互仍然有效。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**随胶囊一并移除侧边栏拖动交互。** 不予采纳，因为本次要求只改视觉表现；移除正常工作的几何控制会不必要地缩减交互方式。
|
||||
|
||||
**保留胶囊，但降低其视觉强调。** 更小或对比度更低的胶囊仍会在侧边栏边界留下一个不需要的物体。
|
||||
|
||||
## 后果
|
||||
|
||||
侧边栏边界在视觉上保持简洁，同时仍可在边界处通过指针调整宽度，并保留缩放光标。与详情栏控件不同，该交互没有可见胶囊。
|
||||
@@ -1,38 +0,0 @@
|
||||
# Agent Note: Web UI drops steer entry and interjection chrome
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-31-web-ui-no-steer-entry-or-interjection-chrome.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Mid-turn steering is a host/agent-loop capability (`mode:'steer'`, durable `steering/message`). The Web product already locked the composer while a turn runs and never shipped a queue/steer menu, yet the client still threaded `'queue' | 'steer'` through the input machine, `conversation.send`, and locale keys, and rendered consumed steering as a badged 「插话」/「Interjection」 bubble. That left a half-built UI surface: an unused submit mode, a product label for a gesture users cannot perform, and e2e goldens that pinned chrome the product does not own.
|
||||
|
||||
## Decision
|
||||
|
||||
Keep host and runtime steering intact. Remove only the Web UI entry and chrome:
|
||||
|
||||
- `InputMachine` / `SessionInput` / `InputActions.submit` / hub `defaultSink` are queue-only; they always call `session.prompt(..., 'queue')`.
|
||||
- `ConversationService.send(text)` drops its mode argument and always queues.
|
||||
- `MessageItem`'s `steering` arm still folds durable `steering/message` content into a plain right-aligned bubble (no badge, no user IconActions) so external/host steers stay visible on replay.
|
||||
- Delete `message.steering` locale strings and the unused badge CSS.
|
||||
- The web steering e2e still POSTs `mode:'steer'` over `/api/session.prompt` and asserts durable + model-visible obedience; it no longer expects interjection chrome. Update [web input machine note](../architecture/2026-07-25-web-input-machine-and-slash-pipeline.md) fact lines to match.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Delete host steering entirely.** Out of scope; the user asked only for Web UI display and entry. Agent-loop drain, session events, and the wire mode remain load-bearing for ACP/TUI/automation.
|
||||
|
||||
**Hide `steering/message` from the transcript.** Would lie on replay when an external client steers; rejected in favor of a plain bubble.
|
||||
|
||||
**Keep the mode parameter but only ever pass `'queue'`.** Leaves dead API surface and tests that invent `'steer'` paths the composer cannot reach.
|
||||
|
||||
## Consequences
|
||||
|
||||
- Web users cannot steer from the composer or `ctx.conversation.send`; stop/cancel and Queue remain the only mid-turn controls.
|
||||
- Host-wire and non-Web clients can still steer; the Web client shows those messages without labeling them as interjections.
|
||||
- Reintroducing a dedicated steer UI would need a new product decision; do not revive the mode union or badge without one.
|
||||
|
||||
## Testing
|
||||
|
||||
- `packages/client/ui-conversation` unit/jsdom coverage: input machine enter/sink, ConversationService routing, MessageItem steering arm (no 「插话」), InputBar submit.
|
||||
- `apps/web/tests/steering.e2e.ts` keyless replay plus updated `settled.expected.md` (steer text without badge).
|
||||
@@ -1,38 +0,0 @@
|
||||
# Agent Note: Web UI 去掉 steer 入口与插话 chrome
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-31-web-ui-no-steer-entry-or-interjection-chrome.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
中途 steering 是 host／agent-loop 能力（`mode:'steer'`、持久 `steering/message`）。Web 产品已在 turn 运行中锁定 composer，且从未交付排队／steer 菜单，但客户端仍把 `'queue' | 'steer'` 穿进 input machine、`conversation.send` 与 locale 键，并把已消费的 steering 渲染成带「插话」／「Interjection」徽章的气泡。这留下半成品 UI：用不到的提交 mode、用户做不到的手势却有产品文案，以及把产品并不拥有的 chrome 钉死在 e2e golden 上。
|
||||
|
||||
## 决策
|
||||
|
||||
保留 host 与 runtime 的 steering。只去掉 Web UI 入口与 chrome：
|
||||
|
||||
- `InputMachine`／`SessionInput`／`InputActions.submit`／hub `defaultSink` 仅 queue；始终调用 `session.prompt(..., 'queue')`。
|
||||
- `ConversationService.send(text)` 去掉 mode 参数，始终排队。
|
||||
- `MessageItem` 的 `steering` 分支仍把持久 `steering/message` 内容折成右对齐普通气泡（无徽章、无用户 IconActions），以便外部／host steer 在回放时仍可见。
|
||||
- 删除 `message.steering` locale 字符串与未使用的徽章 CSS。
|
||||
- web steering e2e 仍通过 `/api/session.prompt` POST `mode:'steer'`，并断言持久化与模型可见服从；不再期望插话 chrome。同步更新 [web input machine note](../architecture/2026-07-25-web-input-machine-and-slash-pipeline.md) 中的事实行。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**整段删除 host steering。** 超出范围；用户只要求清 Web UI 展示与入口。agent-loop 排空、session 事件与线缆 mode 对 ACP／TUI／自动化仍是承重能力。
|
||||
|
||||
**在 transcript 中隐藏 `steering/message`。** 外部客户端 steer 时回放会撒谎；改为普通气泡。
|
||||
|
||||
**保留 mode 参数但永远只传 `'queue'`。** 留下死 API 面与只会虚构 composer 到不了的 `'steer'` 路径的测试。
|
||||
|
||||
## 后果
|
||||
|
||||
- Web 用户无法从 composer 或 `ctx.conversation.send` steer；中途控制只剩停止／取消与 Queue。
|
||||
- Host 线缆与非 Web 客户端仍可 steer；Web 客户端展示这些消息时不再标成插话。
|
||||
- 若要重新引入专用 steer UI，需要新的产品决策；没有决策就不要复活 mode 联合类型或徽章。
|
||||
|
||||
## 测试
|
||||
|
||||
- `packages/client/ui-conversation` unit／jsdom 覆盖：input machine enter／sink、ConversationService 路由、MessageItem steering 分支（无「插话」）、InputBar submit。
|
||||
- `apps/web/tests/steering.e2e.ts` 无密钥回放，以及更新后的 `settled.expected.md`（有 steer 正文、无徽章）。
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-26-turndown-for-tool-web-html-markdown.md
|
||||
2026-07-26-turndown-for-tool-web-html-markdown.md: 0e387021e3d3be3011cc0d64d37864b30aec4fdf
|
||||
2026-07-26-turndown-for-tool-web-html-markdown.zh.md: 44e7d08c1db40a1203e8cda955b521774335e774
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-03-explicit-config-dsh-entrypoint.md
|
||||
2026-08-03-explicit-config-dsh-entrypoint.md: e0d1e954d9cef472ea59345a3d2ef5a67bd03ae8
|
||||
2026-08-03-explicit-config-dsh-entrypoint.zh.md: b5b464e3b45a6f3909bbf087f7005ad3f819424a
|
||||
@@ -0,0 +1,45 @@
|
||||
# Agent Note: Explicit-config dsh entrypoint
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-03-explicit-config-dsh-entrypoint.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Bare `dsh` selected a product TUI implicitly. That made one command own terminal lifecycle, session identity and resume handoff, onboarding, source-workspace shortcuts, guided upgrade sessions, personal config watching, and a large app-level PTY and transcript snapshot suite. The default also hid the actual composition boundary: `--config` was an optional third layer over a TUI overlay rather than the deployment definition a raw launcher needs.
|
||||
|
||||
The shared base is intentionally neutral: it provides capabilities but creates no startup agent or interaction front door. A neutral base paired with an implicit application made raw config composition less explicit and kept product policy in the CLI rather than in the caller-selected overlay.
|
||||
|
||||
## Decision
|
||||
|
||||
Raw executable use is `dsh --config <path>`. The named file must be an Include patch list and is applied directly over `apps/cli/config/base.cordis.yml` at the same include level. It is required for boot, is not a complete replacement tree, and does not inherit `apps/cli/config/web.cordis.yml` or `$DSH_HOME/config.yaml`. Relative paths resolve from the invoking directory. Boot errors fail loud; SIGINT and SIGTERM dispose the root before exit.
|
||||
|
||||
The raw diagnostic forms remain boot-free: `dsh --dump-default-config` prints the base, while `dsh --config <path> --dump-config` prints base plus the required overlay. The dump uses the Include implementation's patch algorithm and YAML dialect.
|
||||
|
||||
The CLI no longer ships a TUI application. Its TUI overlay, launcher, first-run onboarding assets, app-level TUI fixtures, PTY harness, terminal journeys, and snapshots are deleted. The `meta` and `upgrade` subcommands, their experimental gate, default-surface resume, and full-tree `--config-replace` path are deleted with that application. The installer builds and launches Web without an interface selector.
|
||||
|
||||
`dsh web` retains the shared base plus Web overlay and personal-or-explicit user layer. `dsh -p` retains the one-shot Web/headless composition. The reusable TUI package initially remained after this entrypoint change, then [the package-wide removal decision](2026-08-04-remove-tui-package.md) deleted it and its SDK interface.
|
||||
|
||||
This decision supersedes the `dsh`-specific parts of the [dedicated TUI front door](../../archived/feature/2026-07-17-dedicated-full-screen-tui-front-door.md), [personal config](../feature/2026-07-20-dsh-cli-personal-config.md), [guided skill commands](../../archived/feature/2026-07-28-dsh-guided-skill-session-commands.md), [meta workspace](../../archived/feature/2026-07-28-dsh-meta-source-workspace.md), [shared config overlays](2026-07-29-shared-base-config-overlays.md), [config dump](../../archived/feature/2026-07-30-dsh-dump-config.md), [first-run welcome](../../archived/feature/2026-07-30-versioned-tui-first-run-welcome.md), and [experimental subcommand gate](../../archived/feature/2026-07-31-experimental-subcommand-gate.md) notes. The later [package-wide removal decision](2026-08-04-remove-tui-package.md) supersedes their reusable-package decisions and consolidates the deleted launcher-identity record.
|
||||
|
||||
## Verification
|
||||
|
||||
Parser tests require `--config` for raw boot and reject the removed command names and incompatible option combinations. Built-bin acceptance runs the published JavaScript entry without tsx, checks base-only and base-plus-overlay dumps, and drives an invalid raw provider overlay to prove a boot failure settles and exits rather than hanging. Source-launch compatibility checks the same required-config diagnostic through `bin/dsh`. No `apps/cli` TUI demo or test remains.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep bare `dsh` as a TUI and add an explicit config subcommand.** Rejected because the CLI would still own two unrelated application policies and retain the TUI-only launcher, onboarding, and test infrastructure.
|
||||
|
||||
**Allow bare `dsh` to boot the neutral base.** Rejected because the base creates no agent or interaction front door. A process that settles successfully but has no usable entry point hides a missing deployment decision.
|
||||
|
||||
**Keep `--config-replace` for complete trees.** Rejected because raw execution now has one composition contract: a required overlay over the product base. Complete-tree deployments can use the generic Cordis loader or a dedicated application bin without adding a second meaning to `dsh --config`.
|
||||
|
||||
**Delete the TUI package with the product entrypoint.** Initially rejected because removing one shipped application did not by itself require removing a reusable UI implementation. Once no shipped composition or independent consumer remained, [the package-wide removal decision](2026-08-04-remove-tui-package.md) accepted this alternative.
|
||||
|
||||
## Consequences
|
||||
|
||||
Invoking `dsh` without a mode or raw config is a usage error. Existing TUI startup, `meta`, `upgrade`, resume, and full-tree replacement invocations stop working without compatibility aliases. This is acceptable under the pre-release compatibility stance and keeps the supported grammar small.
|
||||
|
||||
Raw deployments must state their agent and front-door rows in an overlay, which makes the application boundary reviewable and keeps base updates available underneath. They do not receive personal config implicitly; deployments that want that policy must compose it themselves. Web remains the installed interactive product surface, while headless and automation entries remain separate.
|
||||
|
||||
Reintroducing a shipped terminal application requires a concrete product need, a named entry mode rather than an implicit raw default, and its own current snapshot and lifecycle acceptance surface.
|
||||
@@ -0,0 +1,45 @@
|
||||
# Agent Note：显式配置的 dsh 入口
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-03-explicit-config-dsh-entrypoint.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
裸 `dsh` 会隐式选择产品 TUI。这使一条命令负责终端生命周期、会话身份与恢复移交、onboarding、源码 workspace 快捷入口、引导式升级会话、个人配置监听，以及一整套规模庞大的应用级 PTY 和 transcript（文本记录）快照测试。该默认行为还隐藏了真实的组合边界：`--config` 是 TUI overlay 之上的可选第三层，而不是 raw 启动器所需的部署定义。
|
||||
|
||||
共享 base 有意保持中性：它提供各项能力，但不会创建启动 agent（智能体）或交互入口。将中性 base 与隐式应用配对，使 raw 配置组合缺乏明确边界，也使产品政策留在 CLI（命令行界面）中，而不是由调用方选定的 overlay 持有。
|
||||
|
||||
## 决策
|
||||
|
||||
raw 执行方式为 `dsh --config <path>`。指定文件必须是一份 Include 补丁列表，并在同一 include 层级直接应用到 `apps/cli/config/base.cordis.yml` 之上。启动时必须提供该文件；它不是完整的替换配置树，也不会继承 `apps/cli/config/web.cordis.yml` 或 `$DSH_HOME/config.yaml`。相对路径从调用目录解析。启动错误会明确报错；SIGINT 和 SIGTERM 会先对根上下文执行 dispose（资源释放），再退出。
|
||||
|
||||
raw 诊断形式仍然无需启动：`dsh --dump-default-config` 打印 base，`dsh --config <path> --dump-config` 则打印 base 与必需 overlay 的合成结果。转储过程使用 Include 实现的补丁算法和 YAML 方言。
|
||||
|
||||
CLI 不再交付 TUI 应用。TUI overlay、启动器、首次运行 onboarding 资产、应用级 TUI fixture（测试前置数据）、PTY harness、终端流程和快照均已删除。`meta` 与 `upgrade` 子命令、对应的实验功能门禁、默认 surface 恢复入口，以及整棵配置树的 `--config-replace` 路径也随该应用一并删除。安装器不再提供界面选择器，只构建并启动 Web。
|
||||
|
||||
`dsh web` 保留共享 base、Web overlay 与个人或显式用户层。`dsh -p` 保留一次性 Web/headless 组合。可复用 TUI 包（package）在本入口变更后起初保留，随后[全包移除决策](2026-08-04-remove-tui-package.md)将其及 SDK 接口删除。
|
||||
|
||||
本决策取代以下记录中专用于 `dsh` 的部分：[独立 TUI 入口](../../archived/feature/2026-07-17-dedicated-full-screen-tui-front-door.md)、[个人配置](../feature/2026-07-20-dsh-cli-personal-config.md)、[引导式 skill 命令](../../archived/feature/2026-07-28-dsh-guided-skill-session-commands.md)、[meta workspace](../../archived/feature/2026-07-28-dsh-meta-source-workspace.md)、[共享配置 overlay](2026-07-29-shared-base-config-overlays.md)、[配置转储](../../archived/feature/2026-07-30-dsh-dump-config.md)、[首次运行欢迎页](../../archived/feature/2026-07-30-versioned-tui-first-run-welcome.md)和[实验性子命令门禁](../../archived/feature/2026-07-31-experimental-subcommand-gate.md)。后续的[全包移除决策](2026-08-04-remove-tui-package.md)取代了其中关于可复用包的决策，并整合了已删除的启动器身份记录。
|
||||
|
||||
## 验证
|
||||
|
||||
解析器测试要求 raw 启动提供 `--config`，并拒绝已删除的命令名和不兼容的选项组合。构建后二进制验收测试在不使用 tsx 的情况下运行已发布的 JavaScript 入口，检查仅含 base 及 base 加 overlay 的转储，并传入无效的 raw 提供方 overlay，以证明启动失败能够结束并退出，而不会挂起。源码启动兼容性测试通过 `bin/dsh` 检查同一条缺少必需配置的诊断。`apps/cli` 不再包含任何 TUI demo 或测试。
|
||||
|
||||
## 曾考虑的替代方案
|
||||
|
||||
**保留裸 `dsh` 作为 TUI，并新增显式配置子命令。** 不予采纳，因为 CLI 仍需持有两套互不相关的应用政策，并保留仅供 TUI 使用的启动器、onboarding 和测试基础设施。
|
||||
|
||||
**允许裸 `dsh` 启动中性 base。** 不予采纳，因为 base 不会创建 agent 或交互入口。进程成功结束启动但没有可用入口，会掩盖缺失的部署决策。
|
||||
|
||||
**保留 `--config-replace` 以支持完整配置树。** 不予采纳，因为 raw 执行现在只有一份组合契约：在产品 base 之上施加一份必需的 overlay。完整配置树部署可以使用通用 Cordis loader 或专用应用二进制文件，无需为 `dsh --config` 增加第二种含义。
|
||||
|
||||
**随产品入口一并删除 TUI 包。** 最初不予采纳，因为仅移除一项已交付应用本身并不要求移除可复用 UI 实现。当已交付组合与独立消费方均不复存在后，[全包移除决策](2026-08-04-remove-tui-package.md)采纳了这一方案。
|
||||
|
||||
## 后果
|
||||
|
||||
调用 `dsh` 时如果既不指定模式，也不提供 raw 配置，将产生用法错误。既有的 TUI 启动、`meta`、`upgrade`、恢复和整棵配置树替换调用会停止工作，且不提供兼容别名。根据发布前兼容性立场，这是可以接受的，并能使支持的命令语法保持精简。
|
||||
|
||||
raw 部署必须在 overlay 中声明 agent 和入口配置项，使应用边界可供评审，并能继续吸收底层 base 的更新。它们不会隐式接收个人配置；需要该政策的部署必须自行组合。Web 仍是安装后提供的交互式产品 surface，headless 与自动化入口仍保持独立。
|
||||
|
||||
重新引入已交付的终端应用，需要有具体产品需求，采用具名入口模式而非隐式 raw 默认值，并建立自身当前有效的快照与生命周期验收面。
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-07-31-web-ui-no-steer-entry-or-interjection-chrome.md
|
||||
2026-07-31-web-ui-no-steer-entry-or-interjection-chrome.md: 6f397668106a6c74f327fc799327752c54824d8c
|
||||
2026-07-31-web-ui-no-steer-entry-or-interjection-chrome.zh.md: f127bd57006747465a1ece87406f5f97086b6a34
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-03-omit-invariants-from-shipped-config.md
|
||||
2026-08-03-omit-invariants-from-shipped-config.md: ff9a3b0ab2b4797ca4e96bea9e6b961b2e38501f
|
||||
2026-08-03-omit-invariants-from-shipped-config.zh.md: 526ce0756e69b36b6f54b46b23e1814322d6a1fd
|
||||
@@ -0,0 +1,30 @@
|
||||
# Agent Note: Omit runtime invariants from shipped dsh config
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-03-omit-invariants-from-shipped-config.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
`@deepseek-ai/dsh-invariants` and package-owned `./invariant` companions are optional development diagnostics. The shipped TUI mounted the service and four stateful companions while the shipped Web tree omitted them, so the two product surfaces had different diagnostic cost and failure behavior. A relational assertion failure could terminate an ordinary TUI run even though the always-on product boundary remained responsible for session validation and immutable history.
|
||||
|
||||
## Decision
|
||||
|
||||
The shipped `dsh` configuration trees under `apps/cli/config/` mount neither `@deepseek-ai/dsh-invariants` nor any package-owned `./invariant` companion. The CLI package therefore carries no direct dependency on the invariant service.
|
||||
|
||||
Invariant support remains available for focused tests, example bundles, generated SDK compositions, and custom deployments that opt into diagnostics explicitly. Session validation, snapshotting, freezing, and provenance remain always on and do not depend on the optional service, as defined by the [source-owned immutability decision](../architecture/2026-06-11-dev-invariants-over-deep-readonly.md).
|
||||
|
||||
The built CLI config-dump test checks both shipped surfaces and rejects either the service entry or any `@deepseek-ai/dsh-*/invariant` entry.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
- **Mount the service with `enabled: false`.** Rejected because the shipped tree and CLI dependency would still carry diagnostics that install no checks.
|
||||
- **Keep the TUI-only mount.** Rejected because the shipped surfaces would retain different diagnostic and failure behavior.
|
||||
- **Remove invariant support from the repository.** Rejected because package-owned checks remain useful in tests, examples, generated SDKs, and explicit development compositions; only the default product config is out of scope.
|
||||
|
||||
## Consequences
|
||||
|
||||
- Ordinary `dsh` TUI and Web runs install no invariant listeners or trace state and cannot fail through `InvariantError`.
|
||||
- Development and custom compositions retain explicit access to the invariant service and companions.
|
||||
- The shipped config absence is verified from the built CLI's composed output for both surfaces.
|
||||
- Always-on session integrity remains unchanged.
|
||||
@@ -0,0 +1,30 @@
|
||||
# Agent Note: 从交付的 dsh 配置中省略运行时不变式
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-03-omit-invariants-from-shipped-config.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
`@deepseek-ai/dsh-invariants` 与各包（package）拥有的 `./invariant` 伴随插件是可选的开发诊断。交付的 TUI 挂载了该服务和四个有状态伴随插件，而交付的 Web 配置树省略了这些条目，导致两个产品 surface 的诊断成本和失败行为不同。即使始终启用的产品边界仍负责会话验证与不可变历史，关系断言失败也可能终止普通的 TUI 运行。
|
||||
|
||||
## 决策
|
||||
|
||||
`apps/cli/config/` 下交付的 `dsh` 配置树既不挂载 `@deepseek-ai/dsh-invariants`，也不挂载任何包拥有的 `./invariant` 伴随插件。因此，CLI 包不再直接依赖不变式服务。
|
||||
|
||||
不变式支持仍可供聚焦测试、示例组合包、生成的 SDK 组合，以及显式选择诊断的自定义部署使用。会话验证、快照、冻结和 provenance 始终启用，且不依赖可选服务，具体由[源端拥有的不可变性决策](../architecture/2026-06-11-dev-invariants-over-deep-readonly.md)规定。
|
||||
|
||||
构建后 CLI 的配置转储测试会检查两个交付的 surface，并拒绝服务条目或任何 `@deepseek-ai/dsh-*/invariant` 条目。
|
||||
|
||||
## 已考虑的替代方案
|
||||
|
||||
- **挂载服务并设置 `enabled: false`。** 不予采纳，因为交付的配置树和 CLI 依赖仍会携带不安装任何检查的诊断。
|
||||
- **保留仅由 TUI 挂载的方案。** 不予采纳，因为两个交付的 surface 仍会保留不同的诊断和失败行为。
|
||||
- **从仓库中移除不变式支持。** 不予采纳，因为包拥有的检查在测试、示例、生成的 SDK 及显式开发组合中仍然有用；只有默认产品配置不在其范围内。
|
||||
|
||||
## 后果
|
||||
|
||||
- 普通的 `dsh` TUI 与 Web 运行不安装不变式监听器或 trace 状态，也不会因 `InvariantError` 失败。
|
||||
- 开发和自定义组合仍可显式使用不变式服务及伴随插件。
|
||||
- 构建后 CLI 的组合输出会验证两个 surface 的交付配置中均不存在这些条目。
|
||||
- 始终启用的会话完整性保持不变。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-04-drop-windows-powershell-picker-fallback.md
|
||||
2026-08-04-drop-windows-powershell-picker-fallback.md: 619afd31d9ec78cdb8565e29fa942b7db8749365
|
||||
2026-08-04-drop-windows-powershell-picker-fallback.zh.md: e14904db46a955d4cf40da195a39bf62cbef96ff
|
||||
@@ -0,0 +1,38 @@
|
||||
# Agent Note: Drop the Windows PowerShell picker fallback
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-04-drop-windows-powershell-picker-fallback.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The win32 branch of the native directory picker kept a two-tier PowerShell fallback under the koffi `IFileOpenDialog` child process: `pwsh.exe` first, then `powershell.exe` (Windows PowerShell 5.1), both running the same WinForms script with a `SetProcessDPIAware` opt-in. The chain existed to keep a working chooser when the koffi tier was "unavailable", but every trigger it plausibly protected was a failure of our own packaging or deployment, not of the operating system:
|
||||
|
||||
- koffi's native binary ships as an ordinary optional dependency (`@koromix/koffi-win32-x64`, no install script); a host that installs the package at all has the binary, and a host that cannot install it fails the package install loudly — the fallback code never loads either.
|
||||
- "Ancient Windows" cannot occur: the Node versions this repo supports run on Windows generations far newer than the Vista-era `IFileOpenDialog` ABI the dialog needs.
|
||||
- A koffi/COM defect crashes only the dialog child process (crash isolation); the correct response to our own bug is a surfaced failure, not a silent downgrade to a legacy dialog.
|
||||
|
||||
The chain also cost real complexity: two spawn tiers running one identical script, a fallback trigger widened from `ENOENT` to any pwsh failure to close the PowerShell 6 (no WinForms) regression, a triple-miss `AggregateError` carrying all three causes, and per-tier abort re-checks. The seam already owns the only fallback that matters — the `browse` backend at the composition level, chosen once at boot by `directory-picker-auto`.
|
||||
|
||||
## Decision
|
||||
|
||||
The win32 tier is exactly the koffi `IFileOpenDialog` child process; any failure surfaces as-is with no fallback. The PowerShell chain — the `pwsh` → Windows PowerShell 5.1 cascade, the DPI-corrected WinForms script, the `AggregateError` aggregation — is deleted, and `pickNativeDirectory`'s win32 branch is a single call. `dsh-native-command` remains a dependency for the POSIX tiers.
|
||||
|
||||
The fallback criterion the rest of the package already followed now applies uniformly: a fallback tier exists only for tools the OS/desktop environment provides and may omit (`zenity` → `kdialog` on Linux, which the boot-time probe also samples); tools our own package ships (`koffi`) fail loud. macOS `osascript` stays fallback-free as before.
|
||||
|
||||
This change consolidates and deletes the pwsh-first DPI picker-fix note: its decision is fully reversed here, and its preserved rationale no longer guides future work on a koffi-only tier. What it kept that was real: PowerShell 7 renders the modern `IFileDialog`-based folder picker where 5.1's `FolderBrowserDialog` is hardwired to the legacy `SHBrowseForFolder` tree; the script's `SetProcessDPIAware` corrected the spawn's system-DPI ceiling; the pwsh→5.1 hop existed because a resolvable PowerShell 6 has no WinForms (exit 1, not `ENOENT`). Its rejected alternatives (requiring PowerShell 7, importing `resolvePwshPath`, setting DPI awareness in the harness process) are moot with the chain gone.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep the chain but drop the pwsh quality tier (`koffi` → Windows PowerShell 5.1).** Rejected: the remaining tier still defends our own packaged dependency, still costs the script, the widened trigger, and the aggregation, and still hides our own vtable/COM defects behind a legacy dialog. The criterion "fallback only for externally provided tools" admits no Windows tier at all.
|
||||
|
||||
**Keep the chain as-is.** Rejected: it was the only two-level runtime fallback in the picker surface, its triggers were deployment-side failures that fail loud anyway, and it degraded a failed pick into an `AggregateError` whose most actionable entry was a PowerShell host.
|
||||
|
||||
**Fall back to `browse` at runtime when the native pick fails.** Rejected: the seam's flow holes are `single`-kind and the `-auto` composition already picks one backend at boot; a runtime cross-kind hop would double-mount both backends and blur the capability boundary.
|
||||
|
||||
## Consequences
|
||||
|
||||
- The win32 picker's failure surface is one error from one tier; callers see the real cause (koffi load failure, COM refusal, dialog crash) instead of a chain-aggregated error.
|
||||
- `pwsh`/`powershell.exe` are no longer invoked by this package; the WinForms script, its `SetProcessDPIAware` correction, and the `-STA` flags are gone with them.
|
||||
- Tests shrink accordingly: the pwsh/5.1 cascade and triple-miss cases are replaced by one "failure surfaces with no fallback" case; the default-adapter test now drives the Linux tier.
|
||||
- Reintroduction condition: a future win32 mechanism outside our packaging chain (a system-provided dialog host we do not ship) would justify a single fallback tier under the same criterion.
|
||||
@@ -0,0 +1,38 @@
|
||||
# Agent Note：删除 Windows PowerShell 选择器回退
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-04-drop-windows-powershell-picker-fallback.md) | 中文
|
||||
|
||||
## Problem
|
||||
|
||||
原生目录选择器的 win32 分支在 koffi `IFileOpenDialog` 子进程之下保留了一条两级 PowerShell 回退：先 `pwsh.exe`，再 `powershell.exe`（Windows PowerShell 5.1），两者运行同一个带 `SetProcessDPIAware` 开关的 WinForms 脚本。该链的存在是为了在 koffi 层"不可用"时仍能给出一个可用的选择器，但它可能保护的每一个触发条件都是我们自己打包或部署的失败，而不是操作系统的：
|
||||
|
||||
- koffi 的原生二进制作为普通 optional 依赖（`@koromix/koffi-win32-x64`，无 install script）分发；能装上该包的宿主就一定有二进制，装不上的宿主会在安装期大声失败——回退代码同样不会加载。
|
||||
- "上古 Windows"不可能出现：本仓库支持的 Node 版本运行在远比 Vista 时代 `IFileOpenDialog` ABI 新的 Windows 世代上。
|
||||
- koffi/COM 缺陷只崩对话框子进程（crash isolation）；对我们自己 bug 的正确反应是上报失败，而不是静默降级到旧版对话框。
|
||||
|
||||
这条链还付出了真实的复杂度：两个 spawn 层运行同一脚本、把回退触发从 `ENOENT` 拓宽为 pwsh 的任何失败以关闭 PowerShell 6（无 WinForms）回归、携带全部三个原因的三连败 `AggregateError`，以及每层的 abort 重检。seam 早已拥有唯一重要的回退——组合层面的 `browse` 后端，由 `directory-picker-auto` 在启动时选择一次。
|
||||
|
||||
## Decision
|
||||
|
||||
win32 层恰好就是 koffi `IFileOpenDialog` 子进程；任何失败原样上报，无回退。PowerShell 链——`pwsh` → Windows PowerShell 5.1 级联、DPI 修正的 WinForms 脚本、`AggregateError` 聚合——被删除，`pickNativeDirectory` 的 win32 分支成为单次调用。`dsh-native-command` 仍为 POSIX 层保留依赖。
|
||||
|
||||
本包其余部分早已遵循的回退判据现在统一适用：回退层只存在于操作系统/桌面环境提供且可能缺失的工具（Linux 的 `zenity` → `kdialog`，启动探针同样采样它们）；我们自己打包的工具（`koffi`）失败即大声报错。macOS `osascript` 与之前一样保持无回退。
|
||||
|
||||
本次变更合并并删除了 pwsh 优先的 DPI 选择器修复 Note：其决策在此被完全反转，其保留的 rationale 对只含 koffi 的层不再指导未来工作。其中真实的部分：PowerShell 7 呈现基于 `IFileDialog` 的现代文件夹选择器，而 5.1 的 `FolderBrowserDialog` 被硬连到旧版 `SHBrowseForFolder` 树；脚本的 `SetProcessDPIAware` 修正了 spawn 的系统 DPI 上限；pwsh→5.1 的跳转存在是因为可解析的 PowerShell 6 没有 WinForms（退出码 1，而非 `ENOENT`）。其被拒绝的替代方案（要求 PowerShell 7、导入 `resolvePwshPath`、在 harness 进程设置 DPI 感知）随链删除而失去意义。
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**保留链但去掉 pwsh 质量层（`koffi` → Windows PowerShell 5.1）。** 拒绝：剩下的层仍在为我们自己打包的依赖辩护，仍要付出脚本、拓宽的触发与聚合的代价，仍会把我们自己的 vtable/COM 缺陷藏到旧版对话框后面。"仅对外部提供的工具回退"的判据不接受任何 Windows 层。
|
||||
|
||||
**原样保留链。** 拒绝：它是选择器面上唯一的二级运行时回退，其触发条件是本就大声失败的部署侧失败，并且它把失败的 pick 降级成一个最具可操作性的条目是 PowerShell 宿主的 `AggregateError`。
|
||||
|
||||
**原生 pick 失败时在运行时回退到 `browse`。** 拒绝：seam 的流程洞是 `single` kind，`-auto` 组合已在启动时选择一个后端；运行时跨 kind 跳转会双挂两个后端并模糊能力边界。
|
||||
|
||||
## Consequences
|
||||
|
||||
- win32 选择器的失败面是来自单一层的一个错误；调用方看到真实原因（koffi 加载失败、COM 拒绝、对话框崩溃），而不是链式聚合的错误。
|
||||
- 本包不再调用 `pwsh`/`powershell.exe`；WinForms 脚本、其 `SetProcessDPIAware` 修正与 `-STA` 标志随之消失。
|
||||
- 测试相应缩减：pwsh/5.1 级联与三连败用例被一个"失败原样上报、无回退"用例取代；默认适配器测试改驱动 Linux 层。
|
||||
- 重新引入条件：未来出现在我们打包链之外的 win32 机制（我们不随包分发的系统提供的对话框宿主）才值得在同一判据下保留一层回退。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-04-remove-tui-package.md
|
||||
2026-08-04-remove-tui-package.md: 7f7a0dd86ddd36e940ed8b7d6154185d9740341c
|
||||
2026-08-04-remove-tui-package.zh.md: 36cb4b6a5e4eddd152eccee92c913bcca5b7fbae
|
||||
@@ -0,0 +1,39 @@
|
||||
# Agent Note: Remove the TUI package
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-04-remove-tui-package.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Removing the implicit `dsh` terminal application left `@deepseek-ai/dsh-tui` without a shipped composition. The package still carried a terminal renderer, interactive command and question adapters, extension overlays, snapshot fixtures, a patched `pi-tui` dependency, and SDK scaffolding that advertised TUI as a supported application interface. Keeping that surface required maintaining a product-sized frontend whose only remaining consumer was the project generator itself.
|
||||
|
||||
The package also made the repository's supported application inventory misleading. Current runnable products use Web, ACP, JSON-RPC, or one-shot CLI front doors, while the SDK continued to offer a terminal choice that no example or product command exercised.
|
||||
|
||||
## Decision
|
||||
|
||||
The `packages/ui/tui` package is deleted without a compatibility package or alias. Its source, package tests, terminal snapshots, dependency declarations, patched `pi-tui` artifact, workspace references, generated service catalog entry, and documentation are removed together. Generic host and agent-loop capabilities remain unchanged.
|
||||
|
||||
The SDK run-interface union now contains only `acp` and `embed`. `create-sdk` defaults to ACP, generated templates contain no terminal startup, resume, session-environment, or model-argument branch, and the builtin `ask-user` feature is removed because neither remaining generated interface supplies a `UserInteractionProvider`. Host applications may still mount the provider-neutral `dsh-user-interaction`, `dsh-commands`, and presentation seams directly.
|
||||
|
||||
This decision supersedes the reusable-package retention in [the explicit-config `dsh` entrypoint decision](2026-08-03-explicit-config-dsh-entrypoint.md) and the current applicability of the archived TUI implementation notes. Their historical records remain frozen, but they are not authority for the supported package or application inventory.
|
||||
|
||||
This note consolidates the deleted package-only records that could not remain current after removal. The terminal UI had kept session identity visible during long conversations, removed duplicate model labels, attached elapsed timing and phase status to messages, showed workspace and branch context beside the prompt, and conservatively parsed complete XML wrappers for human-readable fallback output. Those choices improved one terminal frontend but do not justify retaining it without a deployment. A future XML fallback must still use a real parser rather than regular expressions.
|
||||
|
||||
## Verification
|
||||
|
||||
Repository searches and generated catalogs contain no TUI package, dependency patch, SDK interface option, service key, or package link. Focused SDK tests cover ACP and embedded creation, configuration, templates, and snapshots. The ordinary source build, typecheck, lint, hygiene, documentation gates, and remaining assembled snapshot suites run without the deleted workspace.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Keep the package unshipped.** Rejected because it preserves the maintenance cost and continues to present an unsupported terminal frontend as reusable product surface without a real composition proving its lifecycle.
|
||||
|
||||
**Keep the SDK option for external consumers.** Rejected because the generator would be the package's only in-repository consumer and would scaffold an application the repository no longer accepts end to end. The pre-release compatibility stance does not require preserving that option.
|
||||
|
||||
**Move the package to an examples or experimental group.** Rejected because moving code does not provide a current product need, a maintained deployment, or assembled acceptance. A future terminal frontend should start from its actual host and interaction requirements rather than inherit this implementation by default.
|
||||
|
||||
## Consequences
|
||||
|
||||
DeepSeek Harness has no terminal UI package or generated TUI application. Existing imports, `cordis.yml` rows, SDK `--interface=tui` requests, and projects that depend on the package fail instead of being translated. Web remains the shipped interactive surface; ACP, JSON-RPC, and one-shot CLI remain the non-Web front doors.
|
||||
|
||||
The provider-neutral command, user-interaction, approval, tool-presentation, PTY, and session-projection capabilities remain available to other hosts. Reintroducing a terminal frontend requires a named product or deployment, an explicit package boundary, a concrete interaction provider, and assembled lifecycle and transcript acceptance for that frontend.
|
||||
@@ -0,0 +1,39 @@
|
||||
# Agent Note: 移除 TUI 包（package）
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-04-remove-tui-package.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
移除隐式的 `dsh` 终端应用后，`@deepseek-ai/dsh-tui` 不再拥有任何已交付的组合。该包仍包含终端渲染器、交互式命令与问答适配器、扩展浮层、快照 fixture（测试前置数据）、已打补丁的 `pi-tui` 依赖，以及仍将 TUI 宣称为受支持应用接口的 SDK 脚手架。保留这一表面意味着继续维护一个产品规模的前端，而其唯一剩余消费方就是项目生成器本身。
|
||||
|
||||
该包还会使仓库所支持的应用清单产生误导。当前可运行产品使用 Web、ACP（Agent Client Protocol）、JSON-RPC 或一次性 CLI（命令行界面）入口，SDK 却仍提供终端选项，而没有任何示例或产品命令会实际运行它。
|
||||
|
||||
## 决策
|
||||
|
||||
删除 `packages/ui/tui` 包，不提供兼容包或别名。其源码、包测试、终端快照、依赖声明、已打补丁的 `pi-tui` 产物、workspace 引用、生成的服务目录条目和文档会一并移除。通用宿主能力与 agent-loop 能力保持不变。
|
||||
|
||||
SDK 的运行接口联合类型现在只包含 `acp` 与 `embed`。`create-sdk` 默认使用 ACP，生成的模板不再包含终端启动、恢复、会话环境或模型参数分支；内置的 `ask-user` 功能也被移除，因为剩余两个生成接口都不提供 `UserInteractionProvider`。宿主应用仍可直接挂载提供方无关的 `dsh-user-interaction`、`dsh-commands` 和呈现 seam。
|
||||
|
||||
本决策取代[显式配置 `dsh` 入口决策](2026-08-03-explicit-config-dsh-entrypoint.md)中保留可复用包的决定，也使已归档 TUI 实现记录不再适用于当前状态。这些历史记录继续保持冻结，但不再作为受支持包或应用清单的依据。
|
||||
|
||||
本记录汇总了删除后无法继续保持当前状态的仅限包记录。终端 UI 曾在长对话期间保持会话身份可见、移除重复模型标签、为消息附加耗时与阶段状态、在提示词旁显示 workspace 与分支上下文，并保守地解析完整 XML 包装层，以生成人类可读的回退输出。这些选择改善了一个终端前端，但没有部署时不足以证明应保留它。未来的 XML 回退仍必须使用真实解析器而非正则表达式。
|
||||
|
||||
## 验证
|
||||
|
||||
仓库搜索结果与生成目录中不再包含 TUI 包、依赖补丁、SDK 接口选项、服务键或包链接。聚焦的 SDK 测试覆盖 ACP 与嵌入式创建、配置、模板和快照。常规源码构建、类型检查、lint、hygiene、文档门禁以及其余组装快照测试套件均可在没有已删除 workspace 的情况下运行。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**保留未交付的包。** 不予采纳，因为这会保留维护成本，并继续将一个没有真实组合证明其生命周期的、不受支持的终端前端呈现为可复用产品表面。
|
||||
|
||||
**为外部消费方保留 SDK 选项。** 不予采纳，因为生成器会成为该包在仓库内唯一的消费方，并会搭建一个仓库不再进行端到端验收的应用。预发布兼容性立场不要求保留该选项。
|
||||
|
||||
**将包移入 examples 或 experimental 组。** 不予采纳，因为移动代码无法提供当前产品需求、受维护的部署或组装验收。未来的终端前端应以其实际宿主和交互需求为起点，而不是默认继承此实现。
|
||||
|
||||
## 后果
|
||||
|
||||
DeepSeek Harness 不再提供终端 UI 包或生成的 TUI 应用。现有 import、`cordis.yml` 条目、SDK `--interface=tui` 请求以及依赖该包的项目会直接失败，不会得到兼容转换。Web 仍是已交付的交互表面；ACP、JSON-RPC 与一次性 CLI 仍是 Web 之外的入口。
|
||||
|
||||
提供方无关的命令、用户交互、审批、工具呈现、PTY 与会话投影能力仍可供其他宿主使用。重新引入终端前端时，必须为其提供具名产品或部署、显式包边界、具体交互提供方，以及组装后的生命周期与 transcript（文本记录）验收。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-06-buffer-free-feedback-telemetry.md
|
||||
2026-08-06-buffer-free-feedback-telemetry.md: 008bebdcb59f7ef4fe49f8e731aad77861368d5c
|
||||
2026-08-06-buffer-free-feedback-telemetry.zh.md: 7052e075921f4470864f5ea1c4aed5cf6201becf
|
||||
@@ -0,0 +1,29 @@
|
||||
# Agent Note: Buffer-free feedback telemetry
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-06-buffer-free-feedback-telemetry.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Feedback-only telemetry must upload the session-log prefix only after recorded feedback. Retaining a deep-copied, redacted record for every projected event until that trigger duplicates the canonical session log and grows without a bound for a long-lived session that never records feedback.
|
||||
|
||||
## Decision
|
||||
|
||||
The telemetry coordinator provides `live` and `on-demand` capture. On-demand capture registers no session, flush, or operational-event listeners and retains no projected records. `captureSession(session, throughSeq?)` reads the canonical session log after the handoff cursor through an optional inclusive sequence boundary, applies the fixed projection, deep-copies each accepted event, runs the current `telemetry/record` waterfall, and hands the result to the backend.
|
||||
|
||||
`FEEDBACK_ONLY` invokes that method with the `feedback/record` event's sequence. The append is already committed when `session/event` listeners run, so the replay contains the feedback event and cannot include a later suffix. The existing handoff cursor distinguishes later replays without another pending-record index.
|
||||
|
||||
Because on-demand capture reads only the canonical log, it emits no `agent-error` or `shutdown` operational records. Redaction is evaluated at feedback time rather than append time. The [feedback mode decision](../feature/2026-08-05-feedback-gated-session-telemetry.md) owns the public sharing behavior; this note owns its buffer-free realization.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Retain capture-time redacted records.** This preserves the exact redaction policy and operational records observed when each event occurs, but duplicates the unbounded session prefix. The mode promises feedback-triggered session-log upload, not capture-time policy snapshots or pre-feedback operational telemetry.
|
||||
|
||||
**Retain session event references or sequence numbers.** Rejected because the canonical log already supplies both order and identity. A second index saves payload copies but adds lifecycle state without enabling any required behavior.
|
||||
|
||||
**Write a durable pre-feedback spool.** Deferred until a deployment requires crash recovery before feedback. It adds storage, cleanup, and confidentiality policy to a mode whose intended behavior is to upload nothing when the process exits before feedback.
|
||||
|
||||
## Consequences
|
||||
|
||||
A no-feedback session consumes no telemetry-owned memory proportional to its event count; the canonical session log remains the only pre-feedback copy. Feedback handling performs projection, cloning, and redaction synchronously before the backend's non-blocking enqueue, so its cost scales with the unreleased prefix. A redaction-policy change before feedback affects that replay, and a crash before feedback uploads nothing. Later feedback processes only events beyond the handoff cursor.
|
||||
@@ -0,0 +1,29 @@
|
||||
# Agent Note: 无缓冲反馈遥测
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-06-buffer-free-feedback-telemetry.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
仅反馈遥测必须只在记录反馈后上传会话日志前缀。若在触发前为每个已投影事件保留一份已深拷贝、已脱敏的记录，就会复制权威会话日志；对于长期运行但从不记录反馈的会话，这份副本会无限增长。
|
||||
|
||||
## 决策
|
||||
|
||||
遥测协调器提供 `live` 与 `on-demand` 捕获。按需捕获不注册会话、flush 或运维事件监听器，也不保留投影记录。`captureSession(session, throughSeq?)` 从 handoff 游标之后读取权威会话日志，直至可选的序列号边界（含边界），应用固定投影、深拷贝每个已接受事件、运行当前的 `telemetry/record` waterfall（瀑布式事件），并将结果交给后端。
|
||||
|
||||
`FEEDBACK_ONLY` 以 `feedback/record` 事件的序列号调用该方法。`session/event` 监听器运行时，追加已经提交，因此回放包含该反馈事件，且无法包含后续后缀。现有 handoff 游标可区分后续回放，无需另一个待处理记录索引。
|
||||
|
||||
按需捕获只读取权威日志，因此不会发出 `agent-error` 或 `shutdown` 运维记录。脱敏在反馈时而非追加时求值。[反馈模式决策](../feature/2026-08-05-feedback-gated-session-telemetry.md)规定公开的共享行为；本记录规定其无缓冲实现。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**保留捕获时的已脱敏记录。** 该方案会保留每个事件发生时观察到的确切脱敏策略与运维记录，但也会复制无上限的会话前缀。该模式承诺在反馈触发后上传会话日志，而非保留捕获时策略快照或反馈前运维遥测。
|
||||
|
||||
**保留会话事件引用或序列号。** 已否决，因为权威日志已同时提供顺序与身份。第二个索引可以省去载荷副本，但会增加生命周期状态，且无法实现任何必需行为。
|
||||
|
||||
**写入持久化的反馈前 spool。** 推迟到有部署要求反馈前的崩溃恢复时再实现。该方案会为一个预期在进程于反馈前退出时不上传任何内容的模式增加存储、清理与保密策略。
|
||||
|
||||
## 后果
|
||||
|
||||
没有反馈的会话不会消耗随事件数量增长的遥测自有内存；权威会话日志仍是反馈前的唯一副本。反馈处理会在后端非阻塞入队前同步执行投影、深拷贝与脱敏，因此其开销随未释放前缀增长。反馈前的脱敏策略变更会影响该次回放，而反馈前发生崩溃时什么都不上传。后续反馈只处理 handoff 游标之后的事件。
|
||||
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-08-06-user-bubbles-drop-the-branch-action.md
|
||||
2026-08-06-user-bubbles-drop-the-branch-action.md: 817b5e72b7e18b03ddb3a160e6f7a86b04f02764
|
||||
2026-08-06-user-bubbles-drop-the-branch-action.zh.md: dab3890818d872d7bbb3ac9bcab014ce9b829a61
|
||||
@@ -0,0 +1,27 @@
|
||||
# Agent Note: User and steering bubbles drop the branch action
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-08-06-user-bubbles-drop-the-branch-action.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
Every user and consumed-steering bubble rendered the branch control under the completed-turn-tail gate of the [completed-turn-tail decision](../bug-fix/2026-08-02-message-fork-actions-require-completed-turn-tail.md). On those bubbles the gate is effectively permanent: a turn-opening user message is followed by its own turn's nodes, and a consumed steering message is mid-turn by construction, so the control could enable only when the turn ended with no node after the message at all — a cancel before the first model event. Readers therefore saw a control that never enables, with a tooltip promising a state the button cannot reach. The affordance also misled when read at all: a fork at a message seq cuts at the containing `turn/end`, so "branch at my message" includes the answer below it — the opposite of the branch-to-re-ask reading a control on one's own bubble suggests.
|
||||
|
||||
## Decision
|
||||
|
||||
User and steering bubbles render no branch action. `MessageItem` loses its fork props, `PendingSteeringBubble` loses its `showBranch` special case, and `messageBranchSeqs` narrows to `assistantBranchSeqs`: only a completed turn's transcript tail that is the turn's own content-text assistant may fork. The branch affordance lives solely under the settled answer.
|
||||
|
||||
A turn containing a steer keeps its fork point unchanged: fork is a log-prefix cut at `turn/end`, and the steer is model-visible history the child must inherit, so the settled answer of a steered turn forks like any other. The assistant-side gate and its visible-but-unavailable presentation are also unchanged — under an answer, unavailable is a transient, reachable state (a trailing tool or error row currently owns the tail), which is exactly what the tooltip is for.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Hide the control on message bubbles only while ineligible.** Rejected: it preserves the near-unreachable enabled case at the cost of an icon that appears on one's own bubble only when a turn died before producing anything, an inconsistency not worth the case it serves.
|
||||
|
||||
**Keep the visible-but-unavailable control (status quo).** Rejected: the [completed-turn-tail decision](../bug-fix/2026-08-02-message-fork-actions-require-completed-turn-tail.md) chose visibility so the tooltip could explain a boundary the reader can reach; on user and steering bubbles the boundary is unreachable in practice, so the explanation props up a control that should not exist there.
|
||||
|
||||
**Branch-before-the-message semantics on user bubbles.** Out of scope: re-asking from one's own prompt needs a cut before the message plus composer prefill, a different Host operation. Removing the current control keeps that seat free for such a feature instead of squatting on it with opposite semantics.
|
||||
|
||||
## Consequences
|
||||
|
||||
The only fork handles are the enabled branch controls under settled answers. A turn cancelled before any node followed its message loses its only handle and has no fork point, matching turns whose tail is a content-free interrupted node. Web aria goldens across `apps/web` drop the user-bubble disabled-branch row and its hidden explanation text. Package tests pin that user and steering bubbles render no branch control and that a steering-tail turn leaves the narration's control unavailable.
|
||||
@@ -0,0 +1,27 @@
|
||||
# Agent Note：user 与 steering 气泡移除分支操作
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-08-06-user-bubbles-drop-the-branch-action.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
每个 user 气泡和已消费的 steering（中途引导）气泡都渲染分支控件，受[已完成轮次尾部决策](../bug-fix/2026-08-02-message-fork-actions-require-completed-turn-tail.md)的门禁约束。在这些气泡上，该门禁实际上是永久性的：开轮的 user 消息后面必然跟着本轮自己的节点，已消费的 steering 消息按构造就处在轮次中间，因此只有当轮次结束时该消息之后一个节点都没有——即在第一个模型事件之前就取消——控件才可能启用。读者因此看到一个永远不会启用的控件，tooltip 许诺的是这个按钮到达不了的状态。这个操作入口本身也有误导：在消息 seq 处 fork 会切在所在轮次的 `turn/end`，"在我的消息处分支"实际会把下方的回答一并带走，与在自己气泡上看到分支时"分叉重问"的直觉预期恰好相反。
|
||||
|
||||
## 决策
|
||||
|
||||
user 与 steering 气泡不再渲染分支操作。`MessageItem` 移除其 fork props，`PendingSteeringBubble` 移除其 `showBranch` 特例，`messageBranchSeqs` 收窄为 `assistantBranchSeqs`：只有已完成轮次的 transcript 尾部、且该尾部是本轮自己的带 text 内容 assistant 节点才可 fork。分支入口只存在于已定稿的回答之下。
|
||||
|
||||
含有 steer 的轮次的 fork 点保持不变：fork 是切在 `turn/end` 上的日志前缀，steer 是子会话必须继承的模型可见历史，因此被引导过的轮次的已定稿回答与其他轮次一样可以 fork。assistant 侧的门禁及其可见但不可用的呈现也保持不变——在回答之下，不可用是一个短暂且可到达的状态（当前尾部被后续工具行或错误行占据），这正是 tooltip 的用武之地。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**仅在不可用时隐藏消息气泡上的控件。** 否决：它保住了那个几乎不可达的启用场景，代价是图标只在轮次尚未产出任何东西就中止时才出现在自己的气泡上，这种不一致不值得为它服务的场景付出。
|
||||
|
||||
**保留可见但不可用的控件（现状）。** 否决：[已完成轮次尾部决策](../bug-fix/2026-08-02-message-fork-actions-require-completed-turn-tail.md)选择可见，是为了让 tooltip 解释一个读者可以到达的边界；在 user 与 steering 气泡上这个边界实际不可达，解释文本是在为一个不该存在于此的控件打补丁。
|
||||
|
||||
**在 user 气泡上采用切在消息之前的分支语义。** 不在本次范围内：从自己的提示词重问需要切在消息之前并预填输入框，是另一个 Host 操作。移除当前控件恰好为这样的功能留出位置，而不是让语义相反的控件占着它。
|
||||
|
||||
## 后果
|
||||
|
||||
唯一的 fork 入口是已定稿回答下方启用的分支控件。在任何节点跟上其消息之前就被取消的轮次失去了它唯一的入口，从此没有 fork 点，与尾部是无内容 interrupted 节点的轮次一致。`apps/web` 的 aria golden 全部移除 user 气泡的禁用分支行及其隐藏说明文本。包测试钉住：user 与 steering 气泡不渲染分支控件，steering 作为尾部的轮次让叙述节点的控件保持不可用。
|
||||
Reference in New Issue
Block a user