Merge commit 'ead58a4a476200de2a2f2549ef6a02e73752b618' into codex/dsh-sdk-dynamic-subagent-routing

# Conflicts:
#	.agents/notes/implemented/feature/2026-07-27-typescript-sdk-and-sdk-subagent-backend.i18n.yaml
#	.agents/notes/implemented/feature/2026-07-27-typescript-sdk-and-sdk-subagent-backend.md
#	.agents/notes/implemented/feature/2026-07-27-typescript-sdk-and-sdk-subagent-backend.zh.md
#	examples/python-sdk-agent/tests/sdk.snapshot.ts
#	packages/sdk/client/README.i18n.yaml
#	packages/sdk/client/README.md
#	packages/sdk/client/README.zh.md
#	packages/sdk/client/src/types.ts
#	packages/sdk/protocol/README.i18n.yaml
#	packages/sdk/protocol/README.md
#	packages/sdk/protocol/README.zh.md
#	packages/sdk/server/src/server.ts
#	packages/subagent/subagent-dsh-sdk/tests/fixtures/loader/snapshot.cordis.yml
#	packages/subagent/subagent-dsh-sdk/tests/fixtures/loader/snapshot.replay.cordis.yml
This commit is contained in:
pku-xht 2026-08-25 18:01:08 +08:00
commit c2fb21d13a
2168 changed files with 34789 additions and 13242 deletions

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-08-24-system-prompt-section-order-ties.md
2026-08-24-system-prompt-section-order-ties.md: 673c3b3cd668115ead9f9b5478c2bc432b78f930
2026-08-24-system-prompt-section-order-ties.zh.md: 96a6843a0db48e52a2132ad9f8caf6243dfbbcd2
2026-08-24-system-prompt-section-order-ties.md: d92756e751e893b1d03b8892ef71ff9faac9d2c6
2026-08-24-system-prompt-section-order-ties.zh.md: 4a822b7925a38feb254dbc534fc6153c76a93e19

View file

@ -1,6 +1,7 @@
# Agent Note: Equal-order system-prompt sections render in activation order
Status: implemented
Archived: 2026-08-25
English | [中文](2026-08-24-system-prompt-section-order-ties.zh.md)

View file

@ -1,6 +1,7 @@
# Agent Note: 等序系统提示词分段按激活顺序渲染
Status: implemented
Archived: 2026-08-25
[English](2026-08-24-system-prompt-section-order-ties.md) | 中文

View file

@ -112,6 +112,9 @@
"bug-fix/2026-08-12-collapsed-sidebar-shared-entry-motion.i18n.yaml": "sha256:3ce4f6e39e173fc304bf64deca9c95bcddc1dbb492e065ca8c267a7a40788588",
"bug-fix/2026-08-12-collapsed-sidebar-shared-entry-motion.md": "sha256:7b169aa4543edfc965de5a8b7b9e60aa9d9d5218693cd0b57908e2d482280723",
"bug-fix/2026-08-12-collapsed-sidebar-shared-entry-motion.zh.md": "sha256:88db36c698800bf55c3c7531d6f92665576d978c29c15ff7d74215fb93376cb1",
"bug-fix/2026-08-24-system-prompt-section-order-ties.i18n.yaml": "sha256:f7a20bddd4544738ec0dbbfc52ea931f42317defa1674beb9a3c0daebd52fc2d",
"bug-fix/2026-08-24-system-prompt-section-order-ties.md": "sha256:108a97346eb7a62f1ab01f48dbb9fdd965e8991f53e382b0f501b916af0e9e23",
"bug-fix/2026-08-24-system-prompt-section-order-ties.zh.md": "sha256:3deaddfcf9736b3ff8d61b51093d7e46fdcc86103705033e4aa4c9d043794b16",
"feature/2026-06-14-acp-agent-client-protocol.i18n.yaml": "sha256:006795baa43ae962a8d125cc0f1e9f134bc2ee9fb758b6e7669e3fa0126e1918",
"feature/2026-06-14-acp-agent-client-protocol.md": "sha256:6828c0af74bb3fb96206ca6b21c0e56a000b50e4744aad4bc2c05092f3a5a31b",
"feature/2026-06-14-acp-agent-client-protocol.zh.md": "sha256:ba104e841a1fb84edbd3b6c8119d50445b7785255a7a8d13bb9ac8a2cb4d2e69",

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md
2026-07-05-prompt-variables-and-tool-guidance-ownership.md: 9a53619d9510e3f4fa561f8420b2da3bedbbf4bb
2026-07-05-prompt-variables-and-tool-guidance-ownership.zh.md: dc1164198df0f92b843c75b618f140d8aef86e4f
2026-07-05-prompt-variables-and-tool-guidance-ownership.md: 361184fa7dbdaccd49ac19235c016daf5eb5ca53
2026-07-05-prompt-variables-and-tool-guidance-ownership.zh.md: 6b462572883fb69ca64f2babf28974ae59e7bd74

View file

@ -32,7 +32,7 @@ Plugins register `{{name}}` values through `ctx.systemPrompt.variable(name, prov
### Persona as the order-0 section
`dsh-system-prompt` owns `harness:identity` at order `-100` and the configured `deployment:persona` at order 0, so both survive a replacement loop. Prompt rendering has one path, `renderPrompt(assembly)`, and the routed request header therefore records the exact prompt later replayed by `ctx.tokenMeter` for compaction pressure. An agent-scoped `deployment:persona` shadows the global default and lets subagent providers install a persona before publication. The conventional order bands are identity `-100`, persona `0`, and tool guidance `100–199`.
`dsh-system-prompt` owns `harness:identity` at first-party order `-1000` and the configured `deployment:persona` at order 0, so both survive a replacement loop. Prompt rendering has one path, `renderPrompt(assembly)`, and the routed request header therefore records the exact prompt later replayed by `ctx.tokenMeter` for compaction pressure. An agent-scoped `deployment:persona` shadows the global default and lets subagent providers install a persona before publication. The [first-party order allocation](2026-08-25-sparse-first-party-prompt-section-orders.md) owns the sparse named placements for identity, policy, tool guidance, generated protocol, and final-output obligations.
### Tool guidance ownership

View file

@ -32,7 +32,7 @@ Status: implemented
### Persona 作为 order-0 section
`dsh-system-prompt` 拥有 order 为 `-100` 的 `harness:identity` 和 order 为 0 的配置 `deployment:persona`,因此两者在循环被替换时仍然存活。提示词渲染只有一条路径 `renderPrompt(assembly)`,已路由请求 header 因此会记录准确的提示词,稍后由 `ctx.tokenMeter` 为压缩(compaction)压力回放。agent 作用域的 `deployment:persona` 遮蔽全局默认值,允许 subagent 提供方在发布前安装 persona。约定的 order 区间为:identity `-100`、persona `0`、工具指导 `100–199`。
`dsh-system-prompt` 拥有 first-party order 为 `-1000` 的 `harness:identity` 和 order 为 0 的配置 `deployment:persona`,因此两者在循环被替换时仍然存活。提示词渲染只有一条路径 `renderPrompt(assembly)`,已路由请求 header 因此会记录准确的提示词,稍后由 `ctx.tokenMeter` 为压缩(compaction)压力回放。agent 作用域的 `deployment:persona` 遮蔽全局默认值,允许 subagent 提供方在发布前安装 persona。[first-party 顺序分配](2026-08-25-sparse-first-party-prompt-section-orders.zh.md)规定身份、策略、工具指导、生成协议和最终输出义务的稀疏具名位置。
### 工具指导归属

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md
2026-07-05-reconstructable-requests.md: 3f49ba71a6b98a84b05530c900e902b0cf9f6449
2026-07-05-reconstructable-requests.zh.md: 7b8a9df65b60f975bc3ae60b2c1b0c3a8cc22e95
2026-07-05-reconstructable-requests.md: 3786de02d06c0b6c094297ae89ac3f84053e408d
2026-07-05-reconstructable-requests.zh.md: 851045aca7dababd0da859f3b04b721c65382fc3

View file

@ -22,9 +22,9 @@ Prefix-cache stability is corollary #1, not the headline: an append-only log pro
**Messages.** `Session.deriveMessages()` is cached: each surface entry is projected exactly once, when first seen, through the public per-event function `deriveEventMessage(event)`; a surface rewrite (a compaction `replace` — `SurfaceManager.replaceGeneration`) rebuilds. Callers get a fresh array per call over shared, deep-frozen messages: mutating logged history through a projection is unrepresentable (it throws), replacing the old clone-per-call isolation. External reconstructors fold the same public function over a log prefix, so no two paths can disagree.
`EpochHeader` records the request's non-history state: call config, rendered system prompt, and tool schemas, with empty values canonicalized to absence. `request/header` always writes a full snapshot: the first loop instance uses reason `initial`, later instances use `resume`, and an in-instance change uses `change`. `foldRequestHeader` selects the latest snapshot. Legacy `request/header-delta` events and the removed `fallback` reason are rejected when appended or loaded.
`EpochHeader` records the request's non-history state: call config, rendered system prompt, and tool schemas, with empty values canonicalized to absence. Adapter-supplied effort and token defaults retain their `adapterDefaults` provenance; a Web model selection restored from the log omits an adapter-owned effort so the next resolution cannot reclassify the same effective config as an explicit selection and a false change. `request/header` always writes a full snapshot: the first loop instance uses reason `initial`, later instances use `resume`, an in-instance change uses `change`, and an unchanged envelope beginning an explicitly declared message series or following a surface replacement uses `series`. A `change` snapshot carries `startsSeries: true` when the changed request also starts a series, preserving the two independent facts without a duplicate header. Ordinary append-only later Turns, further same-series Steps, and retries inherit the latest snapshot. `foldRequestHeader` selects the latest snapshot. Legacy `request/header-delta` events and the removed `fallback` reason are rejected when appended or loaded.
Each proposed step first claims its inbox batch and runs `agent/pre-step`. Rejection opens no step; enter opens `step/start` and records the final message batch as `user/message` events. The step then assembles the system prompt and tools, while `agent/request` may replace only the frozen call-config seed. The loop records the owed full header snapshot, builds `GenerateOptions` from derived messages and that header, and deep-freezes it while leaving `AbortSignal` live. The first call config starts from explicit `AgentOptions`, preserving fork overrides and resume reconfiguration; later calls start from the folded header.
Each proposed step first claims its inbox batch and runs `agent/pre-step`. Rejection opens no step; enter opens `step/start`, records the final message batch as `user/message` events, and may use `startsRequestSeries: true` to declare a distinct series. The step then assembles the system prompt and tools, while `agent/request` may replace only the frozen call-config seed. The loop records the owed initial, resume, change, or series full snapshot, builds `GenerateOptions` from derived messages and that header, and deep-freezes it while leaving `AbortSignal` live. The first call config starts from explicit `AgentOptions`, preserving fork overrides and resume reconfiguration; later calls start from the folded header.
**The open step is the reconstruction boundary.** Its entered `user/message` batch and any newly written `request/header` precede request dispatch. Injection after the atomic claim joins a later request, while a listener that must affect this request returns messages through `agent/pre-step`. Header reconstruction selects the step's `request/header`, or carries the prior snapshot when no new header is written.
@ -42,6 +42,7 @@ Like MiniCode, the conversation advances append-only and resets only when model-
- **Detect-and-report** (compare consecutive requests, warn on divergence): catches violations after the fact; a violating request is still constructible and ships. Rejected for interface-level unrepresentability.
- **Event-driven assembly** (re-render only on change signals): a missed-signal bug class — a tool registered mid-session emits `tools/change`, not `system-prompt/change`, and a third-party provider may emit nothing. Per-step render + value compare is robust with zero signal discipline.
- **A custom header-delta codec** (system line edits, name-keyed tool edits, whole config/prefix replacements): reduced repeated bytes but duplicated the representation and its diff/apply/fallback machinery. Full snapshots retain one replay representation.
- **A lightweight series marker referencing the previous header**: reduced repeated prompt and tool bytes, but a window beginning at that marker could not render or reconstruct the request without fetching its predecessor. A self-contained full snapshot preserves one representation for persistence, partial history, and snapshot pinning.
- **Narrative changed-field lists on header snapshots**: derivable by comparing consecutive snapshots. The `reason` remains because an instance boundary is not derivable from the snapshot values.
## Consequences
@ -52,5 +53,5 @@ Like MiniCode, the conversation advances append-only and resets only when model-
- `agent/pre-step` is the current-request message channel; direct inbox mutation is the eventual later-request channel.
- Tool-result trimming needs no new mechanism: a logged single-entry surface replace (`start === end`) carrying a trimmed `tool/result` under the same `callId` — compaction-family, replay-correct, cache-bust batched by the same pressure logic.
- Unreadable referenced attachment objects still fail model requests; [automatic attachment quarantine](../../proposed/bug-fix/2026-08-20-attachment-read-quarantine.md) records the proposed recovery without weakening byte-exact reconstruction.
- Session logs grow one `request/header` snapshot per loop instance plus snapshots on real changes. This is larger than a delta codec but small beside chunk-heavy logs and retains one replay representation. `SESSION_FORMAT_VERSION` stays `0`; legacy delta events are rejected rather than migrated.
- Snapshot expected outputs changed once (every transcript gains its header events); the fs-writing fixtures are stored in the normalized authored form with cwd-relative tool arguments, because replay only round-trips cwd-independent argument paths.
- Session logs grow one `request/header` snapshot per loop instance, real change, and later model-message series. Repeating the full system prompt and tool catalog is larger than a delta codec but small beside chunk-heavy logs and retains one self-contained replay representation. `SESSION_FORMAT_VERSION` stays `0`; legacy delta events are rejected rather than migrated.
- Snapshot fixtures include each repeated series header. Keyless refresh owns those deterministic log changes, while the snapshot harness pins prompt and tool sidecars only for the initial and actual change revisions and reuses the current revision for `series` snapshots. Filesystem-writing fixtures remain in normalized authored form with cwd-relative tool arguments because replay only round-trips cwd-independent argument paths.

View file

@ -22,9 +22,9 @@ Status: implemented
**消息。** `Session.deriveMessages()` 带缓存:每个 surface 条目在首次出现时通过公开的逐事件函数 `deriveEventMessage(event)` 精确投影一次;surface 重写(压缩的 `replace`,即 `SurfaceManager.replaceGeneration`)触发重建。调用方每次获得一个新数组,底层是共享的深度冻结消息:通过投影变异已记录的历史是不可表达的(会抛异常),取代了旧的逐次调用克隆隔离。外部重建器对日志前缀折叠同一个公开函数,因此不可能有两条路径产生分歧。
`EpochHeader` 记录请求的非历史状态:调用配置、渲染后的系统提示词和工具 schema,空值规范化为缺失。`request/header` 始终写入完整快照:首个循环实例使用 reason `initial`,后续实例使用 `resume`,实例内变更使用 `change`。`foldRequestHeader` 选择最新快照。旧的 `request/header-delta` 事件和已移除的 `fallback` reason 在追加或加载时都会被拒绝。
`EpochHeader` 记录请求的非历史状态:调用配置、渲染后的系统提示词和工具 schema,空值规范化为缺失。适配器提供的推理强度与 token 默认值会保留其 `adapterDefaults` 来源信息;Web 从日志恢复模型选择时会省略适配器持有的推理强度,因此下一次解析不会把相同的有效配置重新归类为显式选择并产生虚假变更。`request/header` 始终写入完整快照:首个循环实例使用 reason `initial`,后续实例使用 `resume`,实例内变更使用 `change`,内容未变的封装显式开启消息序列或跟随表层替换时使用 `series`。如果发生变化的请求同时开启序列,`change` 快照会携带 `startsSeries: true`,无需重复 header 即可保留这两个独立事实。普通的仅追加后续 Turn、同一序列内后续的 Step 与重试沿用最新快照。`foldRequestHeader` 选择最新快照。旧的 `request/header-delta` 事件和已移除的 `fallback` reason 在追加或加载时都会被拒绝。
每个拟议步骤先领取其 inbox 批次,再运行 `agent/pre-step`。reject 不打开步骤;enter 打开 `step/start`,并把最终消息批次记录为 `user/message` 事件。随后步骤组装系统提示词与工具,`agent/request` 只能替换冻结的调用配置种子。循环记录所需的完整 header 快照,从派生消息与该 header 构建 `GenerateOptions`,对其深度冻结但保持 `AbortSignal` 活跃。首次调用配置从显式的 `AgentOptions` 出发,保留 fork 覆盖和恢复重配置;后续调用从折叠后的 header 出发。
每个拟议步骤先领取其 inbox 批次,再运行 `agent/pre-step`。reject 不打开步骤;enter 打开 `step/start`,把最终消息批次记录为 `user/message` 事件,并可使用 `startsRequestSeries: true` 声明独立序列。随后步骤组装系统提示词与工具,`agent/request` 只能替换冻结的调用配置种子。循环记录所需的 initial、resume、change 或 series 完整快照,从派生消息与该 header 构建 `GenerateOptions`,对其深度冻结但保持 `AbortSignal` 活跃。首次调用配置从显式的 `AgentOptions` 出发,保留 fork 覆盖和恢复重配置;后续调用从折叠后的 header 出发。
**已打开步骤是重建边界。** 进入步骤的 `user/message` 批次与任何新写入的 `request/header` 都位于请求分派之前。原子领取后发生的注入加入后续请求;必须影响本次请求的监听器则通过 `agent/pre-step` 返回消息。header 重建选择该步骤的 `request/header`,或在无新 header 写入时沿用前一个快照。
@ -42,6 +42,7 @@ Status: implemented
- **检测并报告**(比较连续请求,发散时告警):事后捕获违规;违规请求仍可构造并发出。因违规必须在接口层面不可表达而否决。
- **事件驱动组装**(仅在变更信号时重新渲染):存在漏信号的 bug 类别——会话中途注册的工具发出 `tools/change` 而非 `system-prompt/change`,第三方提供方可能什么都不发。逐步骤渲染加值比较在零信号纪律下即可稳健工作。
- **自定义 header-delta 编解码器**(系统行编辑、按名称键控的工具编辑、完整配置/前缀替换):减少了重复字节,却复制了表示及其 diff/apply/fallback 机制。完整快照只保留一种回放表示。
- **引用前一个 header 的轻量 series 标记**:减少重复的提示词与工具字节,但从该标记开始的窗口若不再读取前序,就无法渲染或重建请求。自包含的完整快照让持久化、局部历史和快照固定共用一种表示。
- **Header 快照上的叙事性变更字段列表**:可以通过比较连续快照推导。`reason` 仍保留,因为实例边界无法从快照值推导。
## 后果
@ -52,5 +53,5 @@ Status: implemented
- `agent/pre-step` 是当前请求的消息通道;直接修改 inbox 则是最终进入后续请求的通道。
- 工具结果裁剪无需新机制:一个已记录的单条目 surface replace(`start === end`),携带同一 `callId` 下裁剪后的 `tool/result`——属压缩家族,回放正确,缓存失效由相同的压力逻辑批量处理。
- 无法读取的被引用附件对象仍会让模型请求失败;[附件自动隔离](../../proposed/bug-fix/2026-08-20-attachment-read-quarantine.zh.md)记录了不削弱字节精确重建的拟议恢复方案。
- 会话日志每个循环实例增长一个 `request/header` 快照,并在真正变更时增加快照。它比 delta 编解码器更大,但相对分片密集型日志仍然很小,并只保留一种回放表示。`SESSION_FORMAT_VERSION` 保持 `0`;旧的 delta 事件被拒绝而非迁移。
- 快照预期输出变更一次(每个 transcript(文本记录)增加其 header 事件);写入文件系统的 fixture(测试前置数据)以规范化的撰写形式存储,工具参数使用 cwd 相对路径,因为回放只对 cwd 无关的参数路径做往返。
- 会话日志会为每个循环实例、真实变更和后续模型消息序列增加一个 `request/header` 快照。重复完整系统提示词与工具目录比 delta 编解码器更大,但相对分片密集型日志仍然很小,并保留一种自包含的回放表示。`SESSION_FORMAT_VERSION` 保持 `0`;旧的 delta 事件被拒绝而非迁移。
- 快照 fixture 包含每个重复的 series header。无密钥 refresh 负责这些确定性日志变化;快照 harness 只为 initial 与真实 change 修订固定提示词和工具 sidecar,并让 `series` 快照复用当前修订。写入文件系统的 fixture 继续以规范化的撰写形式存储,工具参数使用 cwd 相对路径,因为回放只对 cwd 无关的参数路径做往返。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md
2026-07-23-client-plugin-loading-model.md: dfa9f34276f20ffa99541db1544539d693313a2f
2026-07-23-client-plugin-loading-model.zh.md: 68fe9b912c60aceb2ecea315ed0121f9f96c1ecf
2026-07-23-client-plugin-loading-model.md: bd6f6e58c571102afc789ef57085db1e302158cc
2026-07-23-client-plugin-loading-model.zh.md: 256b57102bbec6f793d48d0bdaf60445b194ecdf

View file

@ -28,7 +28,7 @@ The first-generation client loader (`createClientLoader`) hand-wrote both layers
The [client shell layering note](2026-08-15-client-shells-and-dynamic-packages.md) defines the current static and dynamic package sets and the import rules between them. The loading machinery treats every `dsh.client` package as a host-graph row with one ordinary `lib/client.js` factory bundle. Its declaration carries Cordis `inject` edges, synchronous module-table `external` requests, and the optional `immediately` prefetch mark; the composing app owns only the mounted roster.
The web kernel remains framework-free and imports no dynamic package value. Modules is itself a dynamic row, but the host parser delivers its factory before the Vite main module. The HTML-installed `__ModuleLoader__` facade uses that factory to construct the module system when the kernel calls `create()`. Every other dynamic row arrives through the application batch; static React, Cordis, and UI library identities come from the shell seed.
The web kernel remains framework-free and imports no dynamic package value. Modules is itself a dynamic row, but the host parser delivers its factory before the Vite main module. The HTML-installed `__ModuleLoader__` facade uses that factory to construct the module system when the kernel calls `create()`. Every other dynamic row belongs to an application combo script; static React, Cordis, and UI library identities come from the shell seed.
### One module system, one plugin governor
@ -38,13 +38,13 @@ The browser mirrors the host's division of labor. `dsh-client-modules` (`ClientM
The vendored Loader consumes the module system through its `internal` contract — the only call site is `tree.import` — and owns everything entry-shaped: entry creation, fiber activation through cordis service waiting (PENDING until injected services exist, cascading when a service is provided), update/refresh, teardown. The governance code is byte-identical to the host side, per vendor policy. Browserization is compile-time mapping in the shell's vite config: a `node:module` stub alias plus `process.*` defines make `ModuleLoader.fromInternal()` return undefined — exactly the empty slot the shell fills. The module system mounts as `ctx.modules`.
### Batched external-script arrival and source maps
### Combo external-script arrival and source maps
The Host snapshots every built plugin artifact and concatenates its factory registration into one of two same-origin classic scripts. The parser-blocking `bootstrap` batch contains the modules row; the HTML preloads the `application` batch containing every other graph row while bootstrap executes. The module system keys in-flight transport by batch URL, so concurrent row arrivals execute one application script. Successful settlement still requires each requested row's factory id to exist in the module table, and registration does not run the factory, so the side-effect boundary remains first materialization.
The Host snapshots every built plugin artifact and partitions each scheduling phase's ordered rows into one or more same-origin classic scripts. It greedily fills each group while the longer map-form request URL remains within 3 KiB, preserving graph order and allowing another request instead of emitting an oversized URL. Each script is addressed by its package resources, for example `/plugins/??<package-a>/client.js,<package-b>/client.js&rev=<rev>`. The `bootstrap` and `application` values are scheduling phases in the graph, not URL components: HTML preloads every application URL before executing every parser-blocking bootstrap URL. The module system keys in-flight transport by combo URL, so concurrent row arrivals within one group execute one script. Successful settlement still requires each requested row's factory id to exist in the module table, and registration does not run the factory, so the side-effect boundary remains first materialization.
The shared tsdown preset emits `client.js.map` for every plugin and rewrites first-party source paths into the browser-resolvable repository shape `/packages/<group>/<package>/src/...`. The production Client pass consumes `lib/types`; the preset supplies each tsc map to Rolldown and fills `sourcesContent` from the original files, so the final map reaches TypeScript/TSX instead of stopping at emitted JavaScript. Other workspace sources inlined into a bundle likewise resolve to their `packages/` owner, while dependency paths remain unchanged. Batch generation strips each local `sourceMappingURL`, records its generated-line offset, resolves every source against the original per-plugin map URL, and emits one indexed Source Map v3 file whose sections embed the available plugin maps. The Vite shell also emits source maps, letting shell code and batched or individually reloaded plugins map stacks and performance profiles back to TypeScript/TSX.
The shared tsdown preset emits `client.js.map` for every plugin and rewrites first-party source paths into the browser-resolvable repository form `/packages/<group>/<package>/src/...`. The production Client pass consumes `lib/types`; the preset supplies each tsc map to Rolldown and fills `sourcesContent` from the original files, so the final map reaches TypeScript/TSX instead of stopping at emitted JavaScript. Other workspace sources inlined into a bundle likewise resolve to their `packages/` owner, while dependency paths remain unchanged. Combo generation strips each local debug directive, records its generated-line offset, resolves every authored source against the original per-plugin map URL, and emits an Indexed Source Map v3. An authored map supplies its section; otherwise an identity section embeds the generated bundle and uses the packer's `sourceURL` as its source name when present. The absolute map URL mirrors the script resource list by changing every `client.js` suffix to `client.js.map`, so `/plugins/??<package-a>/client.js,<package-b>/client.js&rev=<rev>` points to `/plugins/??<package-a>/client.js.map,<package-b>/client.js.map&rev=<rev>`. One resource follows the same rule and still produces an indexed map with one section. The Vite shell also emits source maps, letting shell code and combo-loaded plugins map stacks and performance profiles back to TypeScript/TSX.
The graph retains each row's revisioned individual URL for HMR and adds content-addressed descriptors for the two startup batches. Initial row revisions are opaque process nonces rather than content hashes; they keep an exceptional initial individual request immutable without hashing every plugin at startup. After the watcher observes one artifact change, `rebuilt(id)` hashes only that bundle and map and publishes the resulting revision. Versioned scripts and maps use immutable caching. The Host serves snapshotted bytes only when the requested revision matches; stale or missing revisions return 404 instead of aliasing newer bytes. An external script's `error` event exposes neither response status nor body, so failure diagnostics name only the URL; the same-origin Host and build-stamped registration id form the identity boundary, while the post-`load` factory-presence check rejects an artifact that did not register the expected id.
The graph retains each row's revisioned one-resource combo URL for HMR and adds a content-addressed descriptor for every startup combo request; several descriptors may carry the same scheduling phase. Initial row revisions are opaque process nonces rather than content hashes; they keep the snapshotted one-resource response immutable without hashing every plugin at startup. After the watcher observes one artifact change, `rebuilt(id)` hashes only that bundle and map and publishes the resulting revision. Startup combo revisions cover the combined script inputs and indexed map. Versioned scripts and maps use immutable caching. The Host serves only exact generated URLs; stale revisions and unadvertised resource lists return 404 instead of aliasing different bytes. An external script's `error` event exposes neither response status nor body, so failure diagnostics name only the URL; the same-origin Host and build-stamped registration id form the identity boundary, while the post-`load` factory-presence check rejects an artifact that did not register the expected id.
### The loading flow, end to end
@ -54,11 +54,11 @@ What happens between `dsh web` starting and the UI appearing? Three stages: the
1. The composing app (`apps/cli`) ships the roster as ordinary rows in its `cordis.yml` config tree — client plugin packages are entry rows like every host plugin, including the always-mounted `client-hmr` row. A roster row that fails to import is caught by `assertEntriesLoaded`; a row whose fiber rejects is reported with its original stack by `assertEntriesActivated` ([host boot decision](2026-07-24-web-config-tree-boot-and-transport-layering.md)).
2. The `dsh-client-modules` node half (the package is dual-face: its browser half is the module table) scans loader entries' package.json `dsh.client` declarations and composes `window.__DSH_BOOT__`: `{ rev, entries: [{ id, url, rev, inject?, immediately?, external? }], batches: [{ phase, url, rev, entries }] }`. The row's three optional fields come from manifests, never hand-copied. Composition orders requested dynamic rows before their consumers, rejects synchronous request cycles, and assigns every row to exactly one initial batch. It refuses declared plugins without built `./client` bundles and groups their package/path rows under one required source-build instruction; malformed declaration fields also fail activation, and the Host audit reports either error from the FAILED fiber.
3. Scanning is incremental per package — there is no full-rescan code path. Each cordis `internal/plugin` emission marks the fiber's entry name dirty (entry-less fibers drop O(1)); a microtask flush reconciles each dirty name against live loader entries, with package metadata (including the negative "not a client package" verdict) cached per name forever and bundle re-hashing reachable only through `rebuilt(id)`. The activation pass seeds the same dirty set from current entries and flushes synchronously, so first scan and steady state share one implementation. Initial rows receive an opaque process nonce plus sequence without hashing their artifacts; batch revisions hash the generated script plus indexed map, and the rows plus batch descriptors hash into `graph.rev`. The graph types are single-sourced in the modules package's `./client` export — the webserver knows nothing about the graph, while modules registers the bundle route and contributes structured index-injection rows.
3. Scanning is incremental per package — there is no full-rescan code path. Each cordis `internal/plugin` emission marks the fiber's entry name dirty (entry-less fibers drop O(1)); a microtask flush reconciles each dirty name against live loader entries, with package metadata (including the negative "not a client package" verdict) cached per name forever and bundle re-hashing reachable only through `rebuilt(id)`. The activation pass seeds the same dirty set from current entries and flushes synchronously, so first scan and steady state share one implementation. Initial rows receive an opaque process nonce plus sequence without hashing their artifacts; startup combo revisions hash the combined script inputs plus indexed map, and the rows plus batch descriptors hash into `graph.rev`. The graph types are single-sourced in the modules package's `./client` export — the webserver knows nothing about the graph, while modules registers the combo route and contributes structured index-injection rows.
Why is the roster yml rows and not a scan? Because which plugins compose into a deployment is a composition decision, not a package property — a package declaring `dsh.client` in the repo does not mean this deployment mounts it, so discovery-by-scan cannot make that call; the node half scans only what the tree actually mounted.
**Phase one — the module face.** The injected HTML installs `window.__ModuleLoader__` in queue mode, starts preloading the application batch, executes the bootstrap batch as one blocking classic script, assigns `window.__DSH_BOOT__`, and then starts the Vite main module. The kernel calls the facade's `create()` with the raw graph and shell seeds. The facade removes and materializes the modules registration with a bootstrap `require` that rejects every external, then calls its `createClientModuleSystem` export. The modules bundle parses the graph, constructs the system, memoizes its own exports, retains the instance in its module closure, and switches the same facade to live registration. The kernel then prefetches every `immediately` row in parallel. Their shared application URL executes once and registers every remaining factory without materializing it. A prefetch failure is swallowed here because phase two's import retries and owns the loud failure. `immediately` remains a registration barrier, not a package identity.
**Phase one — the module face.** The injected HTML installs `window.__ModuleLoader__` in queue mode, starts preloading every application combo URL, executes every bootstrap combo URL as a blocking classic script, assigns `window.__DSH_BOOT__`, and then starts the Vite main module. The kernel calls the facade's `create()` with the raw graph and shell seeds. The facade removes and materializes the modules registration with a bootstrap `require` that rejects every external, then calls its `createClientModuleSystem` export. The modules bundle parses the graph, constructs the system, memoizes its own exports, retains the instance in its module closure, and switches the same facade to live registration. The kernel then prefetches every `immediately` row in parallel. Rows in the same application combo share its execution; separate combos load independently when an immediate row, a requested dependency, or ordinary entry import reaches them. A prefetch failure is swallowed here because phase two's import retries and owns the loud failure. `immediately` remains a registration barrier, not a package identity.
**Phase two — the plugin face.**
@ -76,8 +76,8 @@ How does a rebuilt bundle become a reload signal? The hmr node half observes it
On the browser side, the driver reloads one plugin per frame, serialized:
1. `invalidate` — drop the stale factory and record, and bind the rebuilt frame's revision to that row's individual URL. A live factory would make the next step a no-op.
2. `prefetch` — load the individual external script and register the fresh factory, while the old fiber still serves. The initial batch never executes again.
1. `invalidate` — drop the stale factory and record, and bind the rebuilt frame's revision to that row's one-resource combo URL. A live factory would make the next step a no-op.
2. `prefetch` — load that one-resource external script and register the fresh factory while the old fiber still serves. The initial multi-resource script never executes again.
3. `registry.delete` — before touching the fiber. A bare fiber dispose trips the vendored Loader's self-dispose branch, which would disable the entry permanently.
4. Drain the old fiber's disposers.
5. Remove owned `<style data-plugin>` tags.
@ -96,7 +96,7 @@ The current package inventory and build forms live in the [client shell layering
One governance implementation runs on both sides of the wire; the browser-specific layer is one module system plus one reload plugin. Dynamic packages have one artifact form, so the purity check covers them all. Cordis dependencies, module requests, and the boot tier live with their owners — the manifests — while the composing app holds only the roster. Host graph validation and recursive request arrival keep synchronous factory dependencies explicit. Browser-native script loading preserves the standard mapping among plugin network resources, generated bundles, and TypeScript/TSX sources, while the module system keeps only one replaceable `loadBundle` hook.
Costs accepted: the vendored Loader carries idle machinery in the browser (EntryTree persistence is a no-op, groups/isolation unused); every plugin edit in dev pays a bundle rebuild plus fiber remount; graph `inject` rows guide factory arrival but service availability remains the activation authority, so a mismatch appears at the settled sweep; the static UI libraries keep direct value exports; every bundle gains a source-map artifact; and external-script failures provide only coarse URL diagnostics instead of the HTTP status available to an explicit fetch. The Host retains per-plugin bundle/map snapshots, revision-stamped individual responses, current batches, and one previous batch generation, so memory scales as several copies of the composed client artifacts. This retained state keeps URLs immutable and lets an in-flight request finish across one HMR recomposition.
Costs accepted: the vendored Loader carries idle machinery in the browser (EntryTree persistence is a no-op, groups/isolation unused); every plugin edit in dev pays a bundle rebuild plus fiber remount; graph `inject` rows guide factory arrival but service availability remains the activation authority, so a mismatch appears at the settled sweep; the static UI libraries keep direct value exports; every bundle gains a source-map artifact; and external-script failures provide only coarse URL diagnostics instead of the HTTP status available to an explicit fetch. The Host retains per-plugin bundle/map snapshots, generated one-resource responses, current startup combo responses, and one previous startup generation, so memory scales as several copies of the composed client artifacts. This retained state keeps URLs immutable and lets an in-flight request finish across one HMR recomposition.
Roster: it lives in the web bundle's config tree (`packages/bundle/web-app/cordis.patch.yml`); `mountWebPlugins` and the `CLIENT_PACKAGES` constant are gone, and recomposing a deployment means swapping the yml/overlay. The graph composer lives in the `dsh-client-modules` node half, while the parser-preloaded client face bootstraps the browser module table. The webserver remains a plain route-registration plugin; `/api/*` binding belongs to the connection node half over `api-gateway` (`dsh-host-apiproxy` providing `ctx.apiProxy`), and the dev bundle watch plus SSE channel belongs to the hmr node half.

View file

@ -28,7 +28,7 @@ host 侧,cordis 插件装载站在 Node 的模块机制之上——require cac
[Client 外壳分层 Note](2026-08-15-client-shells-and-dynamic-packages.zh.md)定义当前的静态、动态包集合及其 import 规则。装载机件把每个 `dsh.client` 包视为一个 host graph row,且每个包只有一个普通 `lib/client.js` factory bundle。包声明携带 Cordis `inject` 边、同步模块表 `external` 请求,以及可选的 `immediately` 预取标记;负责组合的 app 只拥有挂载名册。
Web 内核保持不依赖框架,也不 import 任何动态包实体。Modules 本身是动态图 row,但 host parser 会在 Vite 主模块前送达其 factory。内核调用 `create()` 时,由 HTML 安装的 `__ModuleLoader__` facade 使用该 factory 构造模块系统。其他动态图 row 全部经 application 批次到达;React、Cordis 与静态 UI 库的身份由外壳 seed 提供。
Web 内核保持不依赖框架,也不 import 任何动态包实体。Modules 本身是动态图 row,但 host parser 会在 Vite 主模块前送达其 factory。内核调用 `create()` 时,由 HTML 安装的 `__ModuleLoader__` facade 使用该 factory 构造模块系统。其他每个动态图 row 都归属一个 application combo 脚本;React、Cordis 与静态 UI 库的身份由外壳 seed 提供。
### 一套模块系统,一个插件治理器
@ -38,13 +38,13 @@ Web 内核保持不依赖框架,也不 import 任何动态包实体。Modules
vendored Loader 经其 `internal` 约定消费模块系统——唯一调用点是 `tree.import`——并拥有一切 entry 形状的事务:entry 创建、fiber 经 cordis 服务等待的激活(注入的服务未就位即保持 PENDING,服务 provide 时级联激活)、update/refresh、拆除。治理代码按 vendor 政策与 host 侧逐字节相同。浏览器化是壳 vite 配置里的编译期映射:一个 `node:module` stub 别名加若干 `process.*` define,使 `ModuleLoader.fromInternal()` 返回 undefined——这正是留给壳来填的空槽。模块系统挂载为 `ctx.modules`。
### 批量外部脚本到达与源码映射
### Combo 外部脚本到达与源码映射
Host 会快照每个已构建插件产物,并把其 factory registration 拼入两个同源 classic script 之一。阻塞 parser 的 `bootstrap` 批次包含 modules row;HTML 在 bootstrap 执行期间预加载包含其余全部 graph row 的 `application` 批次。模块系统按批次 URL 复用进行中的传输,因此并发 row 到达只执行一次 application 脚本。成功结算仍要求模块表中已经存在被请求 row 的 factory id;登记不会运行 factory,所以副作用边界依然是首次物化。
Host 会快照每个已构建插件产物,并把每个调度阶段的有序 row 划入一个或多个同源 classic script。它在更长的 map 形式请求 URL 保持在 3 KiB 以内时贪心填充每组,既保留 graph 顺序,也以增加请求代替超长 URL。每个脚本都由其中的 package 资源寻址,例如 `/plugins/??<package-a>/client.js,<package-b>/client.js&rev=<rev>`。`bootstrap` 与 `application` 是图中的调度阶段,不是 URL 组成部分:HTML 先预加载所有 application URL,再执行所有阻塞 parser 的 bootstrap URL。模块系统按 combo URL 复用进行中的传输,因此同组 row 的并发到达只执行一个脚本。成功结算仍要求模块表中已经存在被请求 row 的 factory id;登记不会运行 factory,所以副作用边界依然是首次物化。
共享 tsdown 预设为每个插件产出 `client.js.map`,并把第一方源码路径重写成浏览器可识别的仓库形状 `/packages/<group>/<package>/src/...`。生产 Client 构建会消费 `lib/types`;预设把每份 tsc map 交给 Rolldown,并从原文件补齐 `sourcesContent`,使最终 map 回到 TypeScript/TSX,而不是停在编译后的 JavaScript。内联进 bundle 的其他 workspace 源码同样回到其 `packages/` 归属,依赖包路径保持原样。批次生成会移除每个局部 `sourceMappingURL`、记录其生成行偏移、以原插件 map URL 解析每个 source,再产出一份以 section 内嵌现有插件 map 的 indexed Source Map v3 文件。Vite 壳也产出 sourcemap,使壳代码以及批量或独立重载的插件都能从 stack 和性能 profile 回到 TypeScript/TSX。
共享 tsdown 预设为每个插件产出 `client.js.map`,并把第一方源码路径重写成浏览器可识别的仓库形式 `/packages/<group>/<package>/src/...`。生产 Client 构建会消费 `lib/types`;预设把每份 tsc map 交给 Rolldown,并从原文件补齐 `sourcesContent`,使最终 map 回到 TypeScript/TSX,而不是停在编译后的 JavaScript。内联进 bundle 的其他 workspace 源码同样回到其 `packages/` 归属,依赖包路径保持原样。Combo 生成会移除每个局部调试指令、记录其生成行偏移、以原插件 map URL 解析每个自带 source,再产出 Indexed Source Map v3。插件有自带 map 时直接用于对应 section;没有时则生成 identity section,内嵌构建后 bundle,并在存在时把 packer 写入的 `sourceURL` 用作 source 名。绝对 map URL 会平行改写脚本资源列表中的每个 `client.js` 后缀,因此 `/plugins/??<package-a>/client.js,<package-b>/client.js&rev=<rev>` 指向 `/plugins/??<package-a>/client.js.map,<package-b>/client.js.map&rev=<rev>`。单资源也采用相同规则,仍产出只有一个 section 的 indexed map。Vite 壳同样产出 sourcemap,使壳代码与经 combo 加载的插件都能从 stack 和性能 profile 回到 TypeScript/TSX。
图为 HMR 保留每个 row 带 revision 的独立 URL,并为两个启动批次增加按内容寻址的描述。初始 row revision 是进程级不透明 nonce,而不是内容哈希;它无需在启动时哈希每个插件,也能保证异常情况下的初始独立请求不可变。watcher 观察到某个产物变化后,`rebuilt(id)` 只哈希该 bundle 与 map,并发布所得 revision。版本化脚本与 map 使用 immutable 缓存。Host 只在请求 revision 匹配时提供已快照字节;陈旧或缺失 revision 返回 404,不会在旧 URL 下别名到新字节。外部脚本的 `error` 事件不给响应状态与正文,因此失败诊断只报告 URL;同源 Host 与构建期写入的 registration id 是身份边界,`load` 后的 factory 存在性检查负责拒绝未登记预期 id 的产物。
图为 HMR 保留每个 row 带 revision 的单资源 combo URL,并为每个启动 combo 请求增加按内容寻址的描述;多条描述可以使用同一调度阶段。初始 row revision 是进程级不透明 nonce,而不是内容哈希;它无需在启动时哈希每个插件,也能保证已快照的单资源响应不可变。watcher 观察到某个产物变化后,`rebuilt(id)` 只哈希该 bundle 与 map,并发布所得 revision。启动 combo revision 覆盖合并脚本输入与 indexed map。版本化脚本与 map 使用 immutable 缓存。Host 只提供精确生成的 URL;陈旧 revision 与未发布资源列表返回 404,不会别名到其他字节。外部脚本的 `error` 事件不给响应状态与正文,因此失败诊断只报告 URL;同源 Host 与构建期写入的 registration id 是身份边界,`load` 后的 factory 存在性检查负责拒绝未登记预期 id 的产物。
### 装载流程,端到端
@ -54,11 +54,11 @@ Host 会快照每个已构建插件产物,并把其 factory registration 拼
1. 负责组合的 app(`apps/cli`)把名册作为普通行放进它的 `cordis.yml` 配置树——client 插件包与每个 host 插件一样是 entry 行,包括无条件挂载的 `client-hmr` 行。名册行 import 失败由 `assertEntriesLoaded` 捕获;fiber reject 的行则由 `assertEntriesActivated` 报告原始 stack([host boot 决策](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md))。
2. `dsh-client-modules` 的 node 半(该包是双面的:浏览器半就是模块表)扫描 loader entry 的 package.json `dsh.client` 声明,组合出 `window.__DSH_BOOT__`:`{ rev, entries: [{ id, url, rev, inject?, immediately?, external? }], batches: [{ phase, url, rev, entries }] }`。Row 的三个可选字段都来自 manifest,永不人肉抄写。组合会把被请求的动态图 row 排到消费者之前、拒绝同步请求环,并把每个 row 恰好分配给一个初始批次。它会拒绝没有已构建 `./client` bundle 的已声明插件,并把它们的 package/path 行归到一条源码构建要求下;畸形声明字段同样会让激活失败,Host 检查会从 FAILED fiber 报告这两类错误。
3. 扫描是单包增量——不存在全量重扫代码路径。每次 cordis `internal/plugin` 发射把该 fiber 的 entry 名标脏(无 entry 的 fiber O(1) 丢弃);微任务 flush 把每个脏名对账 live loader entries,包元数据(含「非 client 包」的否定结论)按名永久缓存,bundle 重哈希只经 `rebuilt(id)` 可达。激活趟从当前 entries 灌同一脏集合并同步 flush,初扫与稳态共享一条实现。初始 row 使用不透明的进程 nonce 加序号,不对其产物求哈希;批次 revision 对生成的脚本及 indexed map 求哈希,row 与批次描述再共同哈希进 `graph.rev`。图类型单源在 modules 包的 `./client` 出口——webserver 对图一无所知;modules 会注册 bundle 路由并贡献结构化 index 注入行。
3. 扫描是单包增量——不存在全量重扫代码路径。每次 cordis `internal/plugin` 发射把该 fiber 的 entry 名标脏(无 entry 的 fiber O(1) 丢弃);微任务 flush 把每个脏名对账 live loader entries,包元数据(含「非 client 包」的否定结论)按名永久缓存,bundle 重哈希只经 `rebuilt(id)` 可达。激活趟从当前 entries 灌同一脏集合并同步 flush,初扫与稳态共享一条实现。初始 row 使用不透明的进程 nonce 加序号,不对其产物求哈希;启动 combo revision 对合并脚本输入及 indexed map 求哈希,row 与批次描述再共同哈希进 `graph.rev`。图类型单源在 modules 包的 `./client` 出口——webserver 对图一无所知;modules 会注册 combo 路由并贡献结构化 index 注入行。
为什么名册是 yml 行而不是扫描?因为哪些插件组合进一次部署是组合决策,不是包属性——一个在仓库中声明了 dsh.client 的包,不代表这次部署要挂载它,扫描发现无从替人做这个决定;node 半只扫描配置树实际挂载了的东西。
**第一阶段——模块面。**注入的 HTML 以 queue 模式安装 `window.__ModuleLoader__`,开始预加载 application 批次,以一个阻塞式 classic script 执行 bootstrap 批次,赋值 `window.__DSH_BOOT__`,然后启动 Vite 主模块。内核把原始图和外壳 seed 传给 facade 的 `create()`。Facade 移除 modules registration,用拒绝全部 external 的 bootstrap `require` 将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造系统、记忆化自身 exports、在模块闭包中保留该实例,并把同一 facade 切换到 live registration。随后内核并行预取每个 `immediately` row;它们共享的 application URL 只执行一次,并登记其余全部 factory 而不物化。预取失败在这里被吞下,因为第二阶段 import 会重试并拥有那次大声失败。`immediately` 仍是 registration barrier,不是包身份。
**第一阶段——模块面。**注入的 HTML 以 queue 模式安装 `window.__ModuleLoader__`,开始预加载所有 application combo URL,以阻塞式 classic script 依次执行所有 bootstrap combo URL,赋值 `window.__DSH_BOOT__`,然后启动 Vite 主模块。内核把原始图和外壳 seed 传给 facade 的 `create()`。Facade 移除 modules registration,用拒绝全部 external 的 bootstrap `require` 将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造系统、记忆化自身 exports、在模块闭包中保留该实例,并把同一 facade 切换到 live registration。随后内核并行预取每个 `immediately` row;同一 application combo 中的 row 共享一次执行,不同 combo 会在 immediate row、被请求依赖或普通 entry import 首次触及时独立加载。预取失败在这里被吞下,因为第二阶段 import 会重试并拥有那次大声失败。`immediately` 仍是 registration barrier,不是包身份。
**第二阶段——插件面。**
@ -76,8 +76,8 @@ Host 会快照每个已构建插件产物,并把其 factory registration 拼
浏览器侧,驱动插件每帧重载一个插件,串行执行:
1. `invalidate`——丢弃陈旧的 factory 与记录,并把 rebuilt 帧的 revision 绑定到该 row 的独立 URL。Factory 还活着会让下一步变成 no-op。
2. `prefetch`——加载独立外部脚本并登记新 factory,旧 fiber 此刻仍在服役。初始批次不会再次执行。
1. `invalidate`——丢弃陈旧的 factory 与记录,并把 rebuilt 帧的 revision 绑定到该 row 的单资源 combo URL。Factory 还活着会让下一步变成 no-op。
2. `prefetch`——加载该单资源外部脚本并登记新 factory,旧 fiber 此刻仍在服役。初始多资源脚本不会再次执行。
3. `registry.delete`——先于任何 fiber 操作。裸做 fiber dispose 会触发 vendored Loader 的自 dispose 分支,把 entry 永久停用。
4. 排空旧 fiber 的各 disposer。
5. 移除名下的 `<style data-plugin>` 标签。
@ -96,7 +96,7 @@ Host 会快照每个已构建插件产物,并把其 factory registration 拼
Wire 两侧运行同一份治理实现;浏览器特有层只包含一套模块系统和一个重载插件。动态包只有一种产物形态,因此纯度检查覆盖全部动态包。Cordis 依赖、模块请求与启动档位都与其所有者——manifest——同住,负责组合的 app 只握名册。Host graph 校验与递归请求到达使同步 factory 依赖保持显式。浏览器原生 script 装载保留插件网络资源、生成 bundle 与 TypeScript/TSX 源码之间的标准映射,模块系统也只保留一个可替换的 `loadBundle` 钩子。
接受的代价:vendored Loader 在浏览器里背着闲置机件(EntryTree 持久化是 no-op,分组/隔离未用);开发期每次修改插件都要付一次 bundle 重建加 fiber 重挂;graph `inject` row 指导 factory 到达,但服务可用性仍是激活权威,因此不匹配会在 settled 扫描时浮出;静态 UI 库保留直接实体导出;每个 bundle 多出一份 sourcemap 产物,外部 script 失败也只能给出粗粒度 URL 诊断,不能像显式 fetch 那样报告 HTTP 状态。Host 会保留逐插件 bundle/map 快照、带 revision 的独立响应、当前批次及上一代批次,因此内存会随组合出的客户端产物增长为数份副本。这组保留状态使 URL 保持不可变,并让进行中的请求跨越一次 HMR 重组后仍能完成。
接受的代价:vendored Loader 在浏览器里背着闲置机件(EntryTree 持久化是 no-op,分组/隔离未用);开发期每次修改插件都要付一次 bundle 重建加 fiber 重挂;graph `inject` row 指导 factory 到达,但服务可用性仍是激活权威,因此不匹配会在 settled 扫描时浮出;静态 UI 库保留直接实体导出;每个 bundle 多出一份 sourcemap 产物,外部 script 失败也只能给出粗粒度 URL 诊断,不能像显式 fetch 那样报告 HTTP 状态。Host 会保留逐插件 bundle/map 快照、生成的单资源响应、当前启动 combo 响应及上一代启动响应,因此内存会随组合出的客户端产物增长为数份副本。这组保留状态使 URL 保持不可变,并让进行中的请求跨越一次 HMR 重组后仍能完成。
名册位于 web 组合包的配置树(`packages/bundle/web-app/cordis.patch.yml`);`mountWebPlugins` 与 `CLIENT_PACKAGES` 常量已消失,重组一次部署等于替换 yml/overlay。Graph 组合器位于 `dsh-client-modules` node 半,由 parser 预载的 client face 则自举浏览器模块表。Webserver 继续作为朴素路由注册插件;`/api/*` 绑定属于 connection node 半,并经 `api-gateway`(由 `dsh-host-apiproxy` 提供 `ctx.apiProxy`);开发期 bundle 监视与 SSE 通道属于 hmr node 半。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.md
2026-07-24-web-config-tree-boot-and-transport-layering.md: eb30ba84ef293a169931ef6519a9d6d2ea98af7f
2026-07-24-web-config-tree-boot-and-transport-layering.zh.md: a3e310a4a5ab8bc6a40ad8d0336cb94e29c1744f
2026-07-24-web-config-tree-boot-and-transport-layering.md: 3d1ccc2a0f71411d496466288934d9038425e7cf
2026-07-24-web-config-tree-boot-and-transport-layering.zh.md: c2409e128dfdbd8550bb7052a7e0f67a40fe1d1f

View file

@ -18,7 +18,7 @@ English | [中文](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md)
**Config sources have one declaration place each.** Bundle yml values are engineering defaults, Settings sections are writable user preferences, CLI flags address their owning launcher rows, and env values enter through yml `!!js` expressions. Patches replace a row's config wholesale. The resolved frontend `distIndex` uses that patch channel as an assembly fact. The transport-independent provider/model default belongs to `ctx.agentDefaultModel`; the [direct headless entry point](2026-08-09-headless-direct-core-entry-point.md) and the Web gateway consume the same state.
**The transport splits five ways.** `dsh-host-apiproxy` is the gateway plugin (`api-gateway` row): it default-exports `ApiProxyService`, configures only `{nativeOpen?}`, consumes the base layer's entry-point-neutral `ctx.agentDefaultModel`, provides `ctx.apiProxy`, remains transport-agnostic, and registers no routes. `dsh-host-webserver` is a plain route-registration plugin: `WebServer` provides `ctx.webServer` (`register(route) → disposer` with duplicate-pattern throw, `renderIndex` rendering — structured `webserver/index-inject` rows, then raw `tapIndex` transforms in registration order — and `port`), listens on activation, answers per-request failures with 400 and logging, and knows no harness concepts. The connection node half owns the `/api` binding from `ctx.apiProxy` through `toFetchHandler`. The modules node half (`ClientModuleRegistry`, providing `ctx.clientModules`) owns incremental package scanning, the bundle route, the boot injection rows, and `onRebuilt`/`onGraphChanged` notification. The hmr node half owns dev reload through `fs.watchFile` membership and the `/plugins/events` SSE route.
**The transport splits five ways.** `dsh-host-apiproxy` is the gateway plugin (`api-gateway` row): it default-exports `ApiProxyService`, configures only `{nativeOpen?}`, consumes the base layer's entry-point-neutral `ctx.agentDefaultModel`, provides `ctx.apiProxy`, remains transport-agnostic, and registers no routes. `dsh-host-webserver` is a plain route-registration plugin: `WebServer` provides `ctx.webServer` (`register(route) → disposer` with duplicate-pattern throw, `renderIndex` rendering — structured `webserver/index-inject` rows, then raw `tapIndex` transforms in registration order — and `port`), listens on activation, answers per-request failures with 400 and logging, and knows no harness concepts. Its socket-backed Node HTTP entry may apply configured gzip through maintained middleware without adding a response-writing service method or changing route owners; the Web Worker tunnel carries identity bytes. The connection node half owns the `/api` binding from `ctx.apiProxy` through `toFetchHandler`. The modules node half (`ClientModuleRegistry`, providing `ctx.clientModules`) owns incremental package scanning, the bundle route, the boot injection rows, and `onRebuilt`/`onGraphChanged` notification. The hmr node half owns dev reload through `fs.watchFile` membership and the `/plugins/events` SSE route.
**Package export discipline.** The modules package exposes exactly `.` (node half) and `./client` (the complete browser half: `ClientModuleSystem`, `parseBootManifest`, the adoption plugin face) — no bespoke subpaths; wire types re-export through the root for host-side consumers. The adoption handshake: the kernel writes the constructed instance to `window.__DSH_MODULES__` before cordis exists; the `./client` apply reads the slot (missing = loud throw) and provides `ctx.modules`.
@ -40,3 +40,5 @@ English | [中文](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md)
| env vars in the mapping table | The same field would gain env/json double sourcing and need an invented precedence |
| Unbarriered create-after-prefetch (`arrive()` dedup as safety) | Disproved by a 10–25% boot race: in-flight dedup covers same-package double-fetch, not cross-package synchronous require edges |
| json file used directly as loader patches | json keys would couple to yml row structure; profile writers would need cordis knowledge |
| Public response writer plus per-route opt-in | Response coding is Node HTTP policy; exposing it through `ctx.webServer` would make every route owner and test double depend on that policy |
| Hand-written gzip negotiation and stream lifecycle | Maintained middleware already owns negotiation, media-type filtering, header rewriting, backpressure, and threshold behavior |

View file

@ -18,7 +18,7 @@ Status: implemented
**每个配置源有唯一声明位置。** 组合包 yml 值是工程默认,Settings 分节是可写的用户偏好,CLI(命令行界面)flags 面向其归属的启动器配置行,env 值则通过 yml `!!js` 表达式进入。patch 会整体替换一行的 config。解析后的前端 `distIndex` 通过同一条 patch 通道作为组装事实传递。与传输无关的提供方/模型默认值归 `ctx.agentDefaultModel` 所有;[直接 headless 入口](2026-08-09-headless-direct-core-entry-point.zh.md)与 Web 网关消费同一份状态。
**传输五分。** `dsh-host-apiproxy` 是网关插件(`api-gateway` 行):默认导出 `ApiProxyService`,只配置 `{nativeOpen?}`,消费 base 层不偏向特定入口的 `ctx.agentDefaultModel`,provide `ctx.apiProxy`,保持传输无关且不注册路由。`dsh-host-webserver` 是朴素的路由注册插件:`WebServer` provide `ctx.webServer`(`register(route) → disposer`、重复 pattern 即抛、`renderIndex` 渲染——先结构化 `webserver/index-inject` 行、后原始 `tapIndex` 按注册序应用——与 `port`),激活即 listen,单请求失败时答 400 并记日志,且不认识任何 harness 概念。connection node 半拥有从 `ctx.apiProxy` 经 `toFetchHandler` 绑定到 `/api` 的逻辑。modules node 半(`ClientModuleRegistry`,provide `ctx.clientModules`)拥有单包增量扫描、bundle 路由、启动注入行与 `onRebuilt`/`onGraphChanged` 通知。HMR(热模块替换) node 半通过 `fs.watchFile` membership 与 `/plugins/events` SSE 路由拥有开发期重载。
**传输五分。** `dsh-host-apiproxy` 是网关插件(`api-gateway` 行):默认导出 `ApiProxyService`,只配置 `{nativeOpen?}`,消费 base 层不偏向特定入口的 `ctx.agentDefaultModel`,provide `ctx.apiProxy`,保持传输无关且不注册路由。`dsh-host-webserver` 是朴素的路由注册插件:`WebServer` provide `ctx.webServer`(`register(route) → disposer`、重复 pattern 即抛、`renderIndex` 渲染——先结构化 `webserver/index-inject` 行、后原始 `tapIndex` 按注册序应用——与 `port`),激活即 listen,单请求失败时答 400 并记日志,且不认识任何 harness 概念。其基于 socket 的 Node HTTP 入口可以通过受维护的中间件应用已配置的 gzip,无需新增响应写出服务方法或改变 route 所有者;Web Worker 隧道传递 identity 字节。connection node 半拥有从 `ctx.apiProxy` 经 `toFetchHandler` 绑定到 `/api` 的逻辑。modules node 半(`ClientModuleRegistry`,provide `ctx.clientModules`)拥有单包增量扫描、bundle 路由、启动注入行与 `onRebuilt`/`onGraphChanged` 通知。HMR(热模块替换) node 半通过 `fs.watchFile` membership 与 `/plugins/events` SSE 路由拥有开发期重载。
**包出口纪律。** modules 包只暴露 `.`(node 半)与 `./client`(完整浏览器半:`ClientModuleSystem`、`parseBootManifest`、收编插件面)——不设专用子路径;wire 类型经根出口 re-export 给 host 侧消费方。收编握手:内核在 cordis 之前把建好的实例写入 `window.__DSH_MODULES__`;`./client` 的 apply 读取该槽位(缺少时显式抛错)并 provide `ctx.modules`。
@ -40,3 +40,5 @@ Status: implemented
| env 进映射表 | 同一字段将出现 env/json 双源,需再发明优先级 |
| create 不等预取(以 `arrive()` 去重为安全依据) | 被 10–25% boot 竞态证伪:在途去重只覆盖同包双拉,不覆盖跨包同步 require 边 |
| json 直接当 loader patches 文件 | json 键名将耦合 yml 行结构,profile 编写者要懂 cordis |
| 公开响应写出方法并让每条 route 选择接入 | 响应编码属于 Node HTTP 策略;经 `ctx.webServer` 暴露会让每个 route 所有者与测试替身依赖这项策略 |
| 手写 gzip 协商与流生命周期 | 受维护的中间件已经处理协商、媒体类型筛选、响应头改写、背压与阈值行为 |

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-28-api-browser-trust-boundary.md
2026-07-28-api-browser-trust-boundary.md: 92b76c109aa9b55bc72bb76f8b208ea2d814d8ce
2026-07-28-api-browser-trust-boundary.zh.md: 653ff32f2e62bf0e2982517f82649ff2d06e40d4
2026-07-28-api-browser-trust-boundary.md: 4401dc8e361281b1239eb84319fc5353948bc217
2026-07-28-api-browser-trust-boundary.zh.md: c1db2632a6019585ab6a948bcb0f4f2ef853b8a7

View file

@ -6,7 +6,7 @@ English | [中文](2026-07-28-api-browser-trust-boundary.zh.md)
## Problem
The web GUI host serves `/api` over plain HTTP (default `127.0.0.1:3080`, `--host 0.0.0.0` supported), and the surface includes remote-code-execution-grade methods — `session.prompt` drives an agent that runs bash. A browser turns the operator into a confused deputy against such a local API in two classic ways: a malicious page fires a "simple" cross-site POST (`text/plain` — sent without a CORS preflight) whose side effects execute even though the response stays unreadable, and a DNS-rebound origin talks to the socket as if same-origin, making CORS inapplicable entirely, with only the `Host` header betraying the attacker's domain. Before this decision the system's only browser-trust check (`isTrustedNativeDialogRequest`: loopback socket + same-origin + loopback Host) guarded exactly one cosmetic route — `host.pickDirectory`, whose native dialog pops on the host's screen — while every consequential method was unguarded. Guarding per-RPC also could not survive the in-app directory browser, whose whole point is serving legitimately remote clients that a loopback rule would refuse.
The web GUI host serves `/api` over plain loopback HTTP (default `127.0.0.1:3080`; the CLI rejects `--host 0.0.0.0`), and the surface includes remote-code-execution-grade methods — `session.prompt` drives an agent that runs bash. A browser turns the operator into a confused deputy against such a local API in two classic ways: a malicious page fires a "simple" cross-site POST (`text/plain` — sent without a CORS preflight) whose side effects execute even though the response stays unreadable, and a DNS-rebound origin talks to the socket as if same-origin, making CORS inapplicable entirely, with only the `Host` header betraying the attacker's domain. Before this decision the system's only browser-trust check (`isTrustedNativeDialogRequest`: loopback socket + same-origin + loopback Host) guarded exactly one cosmetic route — `host.pickDirectory`, whose native dialog pops on the host's screen — while every consequential method was unguarded. Guarding per-RPC also could not survive the in-app directory browser, whose whole point is serving legitimately remote clients that a loopback rule would refuse.
## Decision
@ -15,17 +15,17 @@ Enforce browser trust once, at the carrier, for the entire `/api` prefix — two
- **Media-type fence (dsh-host-apiproxy)**: every `/api` POST must declare `application/json`, else 415 before parsing. Cross-site "simple" requests thereby stop existing: any cross-site attempt is forced into a CORS preflight this server never answers.
- **Authority fence (dsh-client-connection, `src/api-request-trust.ts`)**: every request must present a `Host` that is loopback or matches a `trustedHosts` entry (exact on `host:port`, any port on port-less entries, WHATWG-normalized; rebinding defense). Deliberately no shortcut for unmarked requests: over plain HTTP a browser attaches neither `Origin` nor Fetch-Metadata to reads (EventSource, images, navigations — those headers go only to trustworthy destinations), so an unmarked request may be a rebound browser read whose response the page can read, and Host is the one header rebinding cannot forge; non-browser clients pass via loopback, the derived LAN IP literals, or a declared authority. An attached `Origin` must equal the Host authority; `sec-fetch-site: cross-site` is refused outright. A `trustedHosts` entry that is not a bare, canonical authority fails the plugin load — WHATWG parsing would otherwise quietly authorize the hostname inside a typo or broaden an exact-port grant. `host.pickDirectory` loses its bespoke guard and rides the same fence.
Two boundaries stay deliberately out of scope: reachability is the webserver binding's policy (`host: 127.0.0.1 | 0.0.0.0`), and authentication for genuinely remote deployments is deferred work recorded in the connection README — the fence is a confused-deputy defense, not an auth layer. The old guard's loopback-socket check was dropped rather than generalized: with binding expressing reachability and `trustedHosts` naming remote authorities, the socket address adds nothing a header fence does not already cover.
Reachability is the webserver binding's policy (`host: 127.0.0.1 | 0.0.0.0`), and this fence is a confused-deputy defense rather than identity. Connection applies the separate [browser token authentication](2026-08-24-browser-token-authentication.md) after the fence. The fence does not inspect peer socket addresses: binding expresses reachability, `trustedHosts` names accepted authorities, and the socket address adds nothing the Host/Origin checks need.
## Alternatives considered
- **Per-RPC guards (status quo extended).** Rejected: the guard list trails the method list forever, the highest-value methods were already unguarded, and a loopback rule on browse RPCs would break the remote deployments they exist for.
- **CORS headers + credential omission.** Rejected: we never want cross-origin reads at all, so answering preflights only widens the surface; refusing them is strictly stronger and simpler.
- **Authentication tokens.** Rejected for this change: token minting/storage/rotation is real product surface; the fence closes the browser-deputy holes without pre-deciding the auth design.
- **Authentication tokens.** Rejected for this change: token minting, storage, and rotation are separate product decisions. The later [browser token authentication](2026-08-24-browser-token-authentication.md) owns them without changing this fence.
## Consequences
- Any future `/api` method is covered by construction; there is no per-route trust decision left to forget.
- Non-loopback deployments must have their serving authorities trusted or requests are refused. The dsh CLI keeps its advertised `--host 0.0.0.0` LAN URL working by deriving the machine's LAN IP literals into the connection row (port-less entries — an IP-literal Host cannot be a rebound name, and the bound port may be OS-assigned) and offers `dsh web --trusted-host` for named authorities; compositions the CLI does not boot declare `trustedHosts` themselves. Non-browser automation rides the same fence: loopback, a derived LAN IP, or a declared authority passes; an undeclared DNS alias is refused.
- A custom non-loopback composition must trust its serving authorities or requests are refused, then satisfy browser authentication like every loopback request. The shipped CLI rejects `--host 0.0.0.0`; `--trusted-host` only extends the Host/Origin fence and grants no identity.
- Clients must label POST bodies `application/json` (ours always did; raw-fetch tests gained the header).
- The trusted-network assumption of an unauthenticated `0.0.0.0` deployment is now documented instead of implicit.
- Host and Origin remain request-routing evidence only. The process token and signed cookie establish the browser identity used by every Host method.

View file

@ -6,7 +6,7 @@ Status: implemented
## 问题
Web GUI 宿主以纯 HTTP 提供 `/api`(默认 `127.0.0.1:3080`,支持 `--host 0.0.0.0`),而这个面上有远程代码执行级别的方法——`session.prompt` 驱动的 agent(智能体)可以运行 bash。浏览器会用两种经典方式把操作者变成攻击此类本地 API 的「混淆代理人」:恶意页面发出跨站「简单请求」 POST(`text/plain`——不经 CORS 预检即发出),其副作用照常执行、只是响应不可读;以及 DNS rebinding 后的源以「同源」身份直连 socket,CORS 整体失效,只有 `Host` 头会暴露攻击者的域名。在本决策之前,系统里唯一的浏览器信任检查(`isTrustedNativeDialogRequest`:回环 socket、同源、回环 Host)只守着一个装饰性的路由——`host.pickDirectory`,其原生对话框弹在宿主屏幕上——而所有真正具有严重后果的方法都没有防护。按 RPC 逐个设防也活不过应用内目录浏览器:它存在的意义就是服务合法的远程客户端,回环规则恰恰会拒绝它们。
Web GUI 宿主以纯 loopback HTTP 提供 `/api`(默认 `127.0.0.1:3080`;CLI 拒绝 `--host 0.0.0.0`),而这个面上有远程代码执行级别的方法——`session.prompt` 驱动的 agent(智能体)可以运行 bash。浏览器会用两种经典方式把操作者变成攻击此类本地 API 的「混淆代理人」:恶意页面发出跨站「简单请求」 POST(`text/plain`——不经 CORS 预检即发出),其副作用照常执行、只是响应不可读;以及 DNS rebinding 后的源以「同源」身份直连 socket,CORS 整体失效,只有 `Host` 头会暴露攻击者的域名。在本决策之前,系统里唯一的浏览器信任检查(`isTrustedNativeDialogRequest`:回环 socket、同源、回环 Host)只守着一个装饰性的路由——`host.pickDirectory`,其原生对话框弹在宿主屏幕上——而所有真正具有严重后果的方法都没有防护。按 RPC 逐个设防也活不过应用内目录浏览器:它存在的意义就是服务合法的远程客户端,回环规则恰恰会拒绝它们。
## 决策
@ -15,17 +15,17 @@ Web GUI 宿主以纯 HTTP 提供 `/api`(默认 `127.0.0.1:3080`,支持 `--ho
- **媒体类型栅栏(dsh-host-apiproxy)**:每个 `/api` POST 必须声明 `application/json`,否则在解析前以 415 拒绝。跨站「简单请求」由此不复存在:任何跨站尝试都被逼进一次本服务器从不应答的 CORS 预检。
- **权威栅栏(dsh-client-connection,`src/api-request-trust.ts`)**:每个请求的 `Host` 都必须是回环地址,或与某个 `trustedHosts` 条目匹配(带端口的 `host:port` 条目精确匹配,不带端口的条目匹配任意端口,均经 WHATWG 归一化;rebinding 防御)。刻意不为无标记请求开捷径:明文 HTTP 下浏览器的读取(EventSource、图片、导航——这些头只发给可信目标)既不带 `Origin` 也不带 Fetch-Metadata,因此无标记请求可能是被重绑页面发起且响应可被读走的读取,而 Host 是重绑唯一伪造不了的请求头;非浏览器客户端经由回环地址、推导的 LAN IP 字面量或已声明的权威通过。若带 `Origin` 则必须与 Host 权威完全一致;`sec-fetch-site: cross-site` 一律拒绝。不是单纯规范化 authority 的 `trustedHosts` 条目会导致插件加载失败——否则 WHATWG 解析会悄悄授权笔误里的 hostname,或放大精确端口授权。`host.pickDirectory` 失去专属守卫,与其他请求同栅而行。
两条边界刻意留在范围之外:可达性由 webserver 的绑定配置(`host: 127.0.0.1 | 0.0.0.0`)控制;真正远程部署的认证是延期工作,记录在 connection README——这道栅栏是混淆代理人防御,不是认证层。旧守卫的回环 socket 检查被放弃而非泛化:绑定表达可达性、`trustedHosts` 点名远程权威之后,socket 地址提供不了头部栅栏覆盖不到的任何东西。
可达性由 webserver 的绑定配置(`host: 127.0.0.1 | 0.0.0.0`)控制,这道栅栏是混淆代理人防御,而不是身份。Connection 在栅栏之后应用独立的[浏览器令牌认证](2026-08-24-browser-token-authentication.zh.md)。栅栏不检查对端 socket 地址:绑定表达可达性,`trustedHosts` 点名接受的 authority,socket 地址提供不了 Host/Origin 校验需要的额外信息。
## 曾考虑的替代方案
- **按 RPC 设防(延续现状)。** 否决:守卫清单永远追着方法清单跑,价值最高的方法本来就没被守住,而 browse RPC 上的回环规则会破坏它们为之存在的远程部署。
- **CORS 头与省略凭据。** 否决:我们根本不想要任何跨源读取,应答预检只会扩大暴露面;拒绝预检严格更强也更简单。
- **认证令牌。** 在本变更中否决:令牌的签发、存储、轮换是真实的产品面;栅栏能够封死浏览器混淆代理人漏洞,无需预先决定认证设计。
- **认证令牌。** 在本变更中否决:令牌签发、存储与轮换属于独立产品决策。后续[浏览器令牌认证](2026-08-24-browser-token-authentication.zh.md)持有这些机制,不改变本栅栏。
## 后果
- 未来任何 `/api` 方法天然在覆盖范围内;不存在会被遗忘的按路由信任决定。
- 非回环部署的对外服务 authority 必须列入信任范围,否则请求会被拒绝。dsh CLI 通过把本机 LAN IP 字面量推导进 connection 行(不带端口的条目——IP 字面量 Host 不可能是被重绑的域名,且绑定端口可能由操作系统分配)来保住它公布的 `--host 0.0.0.0` LAN URL,并提供 `dsh web --trusted-host` 声明具名权威;并非由 CLI 启动的组合自行声明 `trustedHosts`。非浏览器自动化走同一道栅栏:回环地址、推导的 LAN IP 或已声明的权威可通过;未声明的 DNS 别名会被拒绝。
- 自定义非 loopback 组合必须信任其服务 authority,否则请求会被拒绝;随后仍像每个 loopback 请求一样满足浏览器认证。随附 CLI 拒绝 `--host 0.0.0.0`;`--trusted-host` 只扩展 Host/Origin 栅栏,绝不授予身份。
- 客户端必须给 POST 体标注 `application/json`(我们自己的客户端一向如此;裸 fetch 测试补上了该头)。
- 无认证 `0.0.0.0` 部署的「可信网络」假设从隐含变为成文。
- Host 与 Origin 仍只是请求路由证据。进程令牌与签名 cookie 建立每个 Host 方法使用的浏览器身份。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-29-package-regrouping.md
2026-07-29-package-regrouping.md: 52a1fa28e4826daa7b0bb84a37be3c094fddb68c
2026-07-29-package-regrouping.zh.md: bfba4c6bfc6190956b40b295b650d0bed78eb3f4
2026-07-29-package-regrouping.md: 8d7c84434bf45568a7a78006759e002edbfc02d6
2026-07-29-package-regrouping.zh.md: fce454181061a1af51e46a6a2395a47cf6cfdec1

View file

@ -29,7 +29,7 @@ Five regrouping decisions remain current; every other group keeps its prior boun
- **`session/`** is the durable session data plane: the persistence seam with its backends and checkpoint policy, the projection fold that serves whole values from that log, log-backed titles, and OTel reporting. The title fold is itself load-bearing for the read side (`session-query` peer-depends on `dsh-session-title`), so titles belong with the data plane, not in a derived-services annex. The plain name is deliberate (prefer names a human would say); the nearby `core/session` package remains the live in-memory service, while this group is the durable family around it. `session-query/` stays a standalone group — the read/tool surface has its own model tools and SQLite FTS backend and is consumed independently of persistence internals.
- **`interaction/`** is the human-collaboration plane plus the terminal channel that answers it: the question/approval seams, the permission preset, the model-facing `ask_user_question` tool, the human-command registry (`plan-mode` and `command-goal` already consume `commands` together with the interaction seams), and `tui` — the interactive channel is the plane's richest provider and consumer (peer edges to `commands` and `user-questions`), and a one-package `tui/` group would spend a top-level name on one plugin.
- **`boot/`** is a role-complete single-package group: the shared bin boot glue that belongs to no channel and no assembly (consumed by `apps/cli` and the `examples/` demo bins).
- **`boot/`** is a role-complete single-package group: the shared boot glue that belongs to no channel and no assembly (consumed by `apps/cli` and test-only Loader drivers).
- **`guard/`** keeps its documented role, loop-hygiene guards, and gains the tool-call timeout enforcer, dissolving the one-package `timeout/` group whose name collided with `util/timeout`.
- **`extensions/`** names the role `cordis/` obscured: the toolset with which the agent inspects and mounts plugins in its own live runtime, and the landing zone for future self-modification packages.

View file

@ -29,7 +29,7 @@ Status: implemented
- **`session/`** 是持久会话数据平面:持久化 seam 连同其各后端与检查点策略、从该日志折叠(fold)出全量值并对外提供的投影、基于日志的标题,以及 OTel 上报。标题折叠本身就是读取侧的承重构件(`session-query` 对 `dsh-session-title` 声明对等依赖),所以标题属于数据平面,而非某个「派生服务」附属区。用这个朴素的名字是有意为之(名字要像人起的);旁边的 `core/session` 包仍是常驻内存的实时服务,本组则是围绕它的持久家族。`session-query/` 保持独立成组:这个读取/工具面自带模型工具和 SQLite FTS 后端,其消费不依赖持久化内部实现。
- **`interaction/`** 是人机协作平面加上应答它的终端通道:提问/批准 seam、权限预设、面向模型的 `ask_user_question` 工具、人类命令注册表(`plan-mode` 与 `command-goal` 已经把 `commands` 和各交互 seam 放在一起消费),以及 `tui`——这个交互通道是该平面功能最丰富的提供方与消费方(对 `commands` 与 `user-questions` 均有对等依赖边),而一个单包 `tui/` 组会把一个顶层名字花在一个插件上。
- **`boot/`** 是角色完备的单包组:不归属任何通道也不归属任何组装的共享 bin boot 胶水(被 `apps/cli` 与 `examples/` 各演示 bin 消费)。
- **`boot/`** 是角色完备的单包组:不归属任何通道也不归属任何组装的共享 boot 胶水(被 `apps/cli` 与仅限测试的 Loader driver 消费)。
- **`guard/`** 保留其文档记载的角色(循环卫生守卫),并新纳入强制执行工具调用超时的包;那个与 `util/timeout` 撞名的单包组 `timeout/` 随之解散。
- **`extensions/`** 把 `cordis/` 遮蔽掉的角色说了出来:它是供 agent(智能体)在自身当前运行时中检查和挂载插件的工具集,也是未来自我修改类包的落点。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-30-command-row-copy-contract.md
2026-07-30-command-row-copy-contract.md: f6d5199389b3907780c501894e2861e6add85e77
2026-07-30-command-row-copy-contract.zh.md: 86f0c044e841a6a0148b1f785582868bba5bb3d0
2026-07-30-command-row-copy-contract.md: d410adf62dbbc4445936c70309d5902bc184b16f
2026-07-30-command-row-copy-contract.zh.md: 0f680f9beb8b335dd6311cadc6c7e022809fb8f2

View file

@ -32,4 +32,4 @@ The log is unchanged: `command/run` keeps the structured `name`/`args` split, so
## Consequences
Every command row gets shorter, and the rule scales: a new command's author writes its outcome without knowing which surface renders it, and no surface has to de-duplicate. The cost is that the dispatched arguments leave the collapsed row — while a command is still executing the row shows only its name and `执行中…` — and that the no-caption rule is a convention the reviewer enforces, not a gate. The `/permission` texts are pinned by the permission package's command tests, and the assembled row copy by the [seeded-history](../../../../apps/web/tests/snapshots/seeded-history/command-row.expected.md) web golden, which reaches a real settled command row keylessly because `/permission` runs entirely on the host.
Every command row gets shorter, and the rule scales: a new command's author writes its outcome without knowing which surface renders it, and no surface has to de-duplicate. The cost is that the dispatched arguments leave the collapsed row — while a command is still executing the row shows only its name and `执行中…` — and that the no-caption rule is a convention the reviewer enforces, not a gate. The `/permission` texts are pinned by the permission package's command tests, and the assembled row copy by the [seeded-history](../../../../snapshots/web/seeded-history/command-row.expected.md) web golden, which reaches a real settled command row keylessly because `/permission` runs entirely on the host.

View file

@ -32,4 +32,4 @@ Web 命令条目由一对落库的[命令生命周期事件](../../proposed/arch
## 后果
每一条命令条目都变短了,而且这条规则可扩展:新命令的作者写结果时无需知道由哪个界面渲染,任何界面也都不必再去重。代价是分派参数离开了折叠行——命令仍在执行时,行上只有名字和 `执行中…`——以及「不加题头」这条规则是靠评审执行的约定,而非门禁。`/permission` 的文案由 permission 包的命令测试钉住,装配后的行文案由 [seeded-history](../../../../apps/web/tests/snapshots/seeded-history/command-row.expected.md) web 预期输出钉住:由于 `/permission` 完全在 host 上执行,该预期输出无需密钥即可覆盖一条真实的已落定命令条目。
每一条命令条目都变短了,而且这条规则可扩展:新命令的作者写结果时无需知道由哪个界面渲染,任何界面也都不必再去重。代价是分派参数离开了折叠行——命令仍在执行时,行上只有名字和 `执行中…`——以及「不加题头」这条规则是靠评审执行的约定,而非门禁。`/permission` 的文案由 permission 包的命令测试钉住,装配后的行文案由 [seeded-history](../../../../snapshots/web/seeded-history/command-row.expected.md) web 预期输出钉住:由于 `/permission` 完全在 host 上执行,该预期输出无需密钥即可覆盖一条真实的已落定命令条目。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-30-config-plane-boundaries.md
2026-07-30-config-plane-boundaries.md: 7b234ec64669cc1e6f0d5a67437005747edcea98
2026-07-30-config-plane-boundaries.zh.md: f543c511d9374a50955331911b2eb2d62dab0675
2026-07-30-config-plane-boundaries.md: d7ae13f08ca5c3957a8a5c1871b0022453e9b562
2026-07-30-config-plane-boundaries.zh.md: e53a68f91f4d961af5a01fd7bbfb4472b4b17dfe

View file

@ -20,7 +20,7 @@ Three smaller defects sat beside them. `llm/adapters-updated` documented contain
## Decision
**Reading configuration is as privileged as writing it.** `settings.describe` and `credentials.describe` join the loopback-only set, so the whole configuration plane stays same-origin until real authentication exists. The model catalog (`llm.providers`, `llm.models`) deliberately does not: it carries provider ids, display names, and model lists — no endpoints, no key state — and a LAN client's model picker needs it. The boundary is asserted over a real HTTP server rather than a hand-assembled request, because the `Host` header a browser actually sends is what decides it.
**Reading configuration is as privileged as writing it.** `settings.describe`, `credentials.describe`, the model catalog, and every other Host operation require one browser session. The configuration plane still redacts secrets independently of authentication. The boundary is asserted over a real HTTP server rather than a hand-assembled request, proving that a forged loopback `Host` value never establishes identity.
**The plane serves exactly the namespaces a registered model provider addresses.** `ctx.llm.listConfigurableProviders()` is the allow-list, so the product boundary is enforced rather than inferred from the installed plugin set, and a future namespace becomes web-configurable only by joining that directory. An unregistered namespace and an unexposed one answer identically (`settings-not-exposed`), so probing cannot enumerate the registry.

View file

@ -20,7 +20,7 @@ Status: implemented
## 决策
**读取配置与写入配置同样属于特权操作。**`settings.describe` 与 `credentials.describe` 加入仅限回环的集合,因此在真正的认证层出现之前,整个配置面都保持同源。模型目录(`llm.providers`、`llm.models`)刻意不在其中:它携带的是提供方 id、显示名与模型列表——没有端点、没有密钥状态——而 LAN 客户端的模型选择器正需要它。这条边界由一台真实 HTTP 服务器来断言,而不是手工拼装的请求,因为真正决定它的,是浏览器实际发出的那个 `Host` 头。
**读取配置与写入配置同样属于特权操作。**`settings.describe`、`credentials.describe`、模型目录及其他所有 Host 操作都要求同一个浏览器会话。配置面仍独立于认证对 secret 脱敏。这条边界由真实 HTTP 服务器而非手工请求来断言,证明伪造 loopback `Host` 值绝不建立身份。
**这个面恰好服务于已注册模型提供方所指向的那些 namespace。**`ctx.llm.listConfigurableProviders()` 就是允许列表,于是产品边界是被执行的,而不是从已安装的插件集合里推断出来的;将来的 namespace 只有加入该目录才会变得可在 Web 上配置。未注册的 namespace 与未暴露的 namespace 得到完全相同的答复(`settings-not-exposed`),因此探测无法枚举注册表。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-30-web-config-plane.md
2026-07-30-web-config-plane.md: c8397d03cd7b8eb4ea127cdc26124f5e721e822d
2026-07-30-web-config-plane.zh.md: b0724c3a8cb160cbac5fc88fe07d35e79accfc49
2026-07-30-web-config-plane.md: d5bd9c05e8352536c5c6f8b265db7dbd56a4fb84
2026-07-30-web-config-plane.zh.md: 3c7f801766b1a6c197cc208a3c4a030b8aaac771

View file

@ -12,11 +12,11 @@ The request-level configuration seam made LLM adapter configuration restart-free
## Decision
**Wire domains on the compiled RPC map, rejections as codes, owner events forwarded verbatim.** `settings.describe/openDocument/update/replace/mutate`, `credentials.describe/set/unset`, `llm.providers`, and `llm.models` join `RpcMethodMap`, so the compiler-locked wiring sites keep schema, handler, and client in lockstep. Seam rejections fold into `settings-rejected {ns}` / `credential-rejected {ref}` business errors, while clients subscribe to forwarded settings, credentials, and LLM owner events and converge without polling ([forwarded Remote events](2026-08-10-remote-event-delivery.md)). Settings reads, native actions, and writes join `pickDirectory`/`openPath` in the connection guard's privileged set: loopback + same-origin or 403, because a LAN-exposed dsh web must not accept configuration access from another origin.
**Wire domains on the compiled RPC map, rejections as codes, owner events forwarded verbatim.** `settings.describe/openDocument/update/replace/mutate`, `credentials.describe/set/unset`, `llm.providers`, and `llm.models` join `RpcMethodMap`, so the compiler-locked wiring sites keep schema, handler, and client in lockstep. Seam rejections fold into `settings-rejected {ns}` / `credential-rejected {ref}` business errors, while clients subscribe to forwarded settings, credentials, and LLM owner events and converge without polling ([forwarded Remote events](2026-08-10-remote-event-delivery.md)). Connection authenticates settings reads, native actions, writes, `pickDirectory`, `openPath`, and every other Host operation with one browser session; Host/Origin failures still return 403 before identity is checked.
**`describe()` grows layers and structural secret redaction.** `SettingsDescriptor` carries `base`/`user` beside the effective value, so the form marks "overridden" by presence in the user layer, not value inequality (an override *equal* to the base is still an override). `describe({ redactSecrets: true })` — mandatory at every wire face — strips `role('secret')` subtrees from all three layers via a pure structural walk of the schema (object/dict/array containers; a secret-role subtree is one opaque leaf) and enumerates the stripped slots as `{path, set}`, so a page can render write-only inputs without ever receiving a value.
**The Host identifies and opens the local settings document.** The settings seam exposes optional `documentPath` provider metadata and a `prepareDocument()` operation; `settings-file` returns its fully resolved custom or `$DSH_HOME/settings.yaml` filename and exclusively creates an absent empty document with owner-only permissions, while non-file providers retain the base `undefined`. The loopback-only `settings.describe` response carries only the boolean `hasDocument` capability beside the redacted namespace views. `ui-settings-general` registers a `settings.action` entry only on loopback pages, shows it only after the metadata confirms that a provider-owned local document can be prepared, and invokes pathless `settings.openDocument`; the Host resolves the provider path again before a text-document handoff (`open -t` on macOS so an arbitrary YAML file association cannot redirect the gesture, `xdg-open` on desktop Linux, `Invoke-Item` on Windows, and `wslpath -w` followed by that Windows handoff on WSL). Generic workspace paths retain the default intent, including its browser preference for browser-renderable documents. The browser neither derives `$DSH_HOME` nor receives a filesystem target; remote pages make no privileged settings read for this action.
**The Host identifies and opens the local settings document.** The settings seam exposes optional `documentPath` provider metadata and a `prepareDocument()` operation; `settings-file` returns its fully resolved custom or `$DSH_HOME/settings.yaml` filename and exclusively creates an absent empty document with owner-only permissions, while non-file providers retain the base `undefined`. The browser-authenticated `settings.describe` response carries only the boolean `hasDocument` capability beside the redacted namespace views. `ui-settings-general` registers a `settings.action` entry only on loopback pages, shows it only after the metadata confirms that a provider-owned local document can be prepared, and invokes pathless `settings.openDocument`; the Host resolves the provider path again before a text-document handoff (`open -t` on macOS so an arbitrary YAML file association cannot redirect the gesture, `xdg-open` on desktop Linux, `Invoke-Item` on Windows, and `wslpath -w` followed by that Windows handoff on WSL). Generic workspace paths retain the default intent, including its browser preference for browser-renderable documents. The browser neither derives `$DSH_HOME` nor receives a filesystem target; non-loopback pages retain the Client policy that makes no Host settings read for this action.
**The llm seam declares configurability and announces topology.** `registerConfigurableProviders()` is an all-or-nothing, fiber-scoped directory of `{provider, displayName, settingsNs, settingsPath}` — the addressing a config page needs to open the right settings subtree for a route that may not exist yet; `listConfigurableProviders()` merges with live routes in the wire handler so undeclared live routes still report active. The zero-payload `'llm/adapters-updated'` event fires from all four registration/unregistration commit points with contained listener dispatch (INVARIANT rethrow), following the settings/commands precedent. `llm-deepseek`'s route renamed to `deepseek-official` because the pi-ai catalog legitimately owns `deepseek` as an aggregator entry; pre-release stance, no alias.

View file

@ -12,11 +12,11 @@ Status: implemented
## 决策
**wire 领域挂上编译期 RPC 映射,拒绝落为错误码,owner 事件原样转发。**`settings.describe/openDocument/update/replace/mutate`、`credentials.describe/set/unset`、`llm.providers` 与 `llm.models` 一同加入 `RpcMethodMap`,由编译器锁定的接线位点让 schema、处理器与客户端保持步调一致。seam 侧拒绝折叠为业务错误,客户端则订阅转发的 settings、credentials 与 LLM owner 事件,无需轮询即可收敛(见[转发的 Remote 事件](2026-08-10-remote-event-delivery.zh.md))。settings 读取、原生操作与写入和 `pickDirectory`/`openPath` 一起进入连接守卫的特权集合:回环 + 同源,否则 403,因为暴露在局域网上的 dsh web 绝不能接受来自其他源的配置访问。
**wire 领域挂上编译期 RPC 映射,拒绝落为错误码,owner 事件原样转发。**`settings.describe/openDocument/update/replace/mutate`、`credentials.describe/set/unset`、`llm.providers` 与 `llm.models` 一同加入 `RpcMethodMap`,由编译器锁定的接线位点让 schema、处理器与客户端保持步调一致。seam 侧拒绝折叠为业务错误,客户端则订阅转发的 settings、credentials 与 LLM owner 事件,无需轮询即可收敛(见[转发的 Remote 事件](2026-08-10-remote-event-delivery.zh.md))。Connection 用一个浏览器会话认证 settings 读取、原生操作、写入、`pickDirectory`、`openPath` 与其他所有 Host 操作;Host/Origin 失败仍会在身份校验前返回 403。
**`describe()` 增加分层与结构化 secret 脱敏。**`SettingsDescriptor` 在生效值之外携带 `base`/`user`,表单据此按「字段是否出现在用户层」来标记「已覆盖」,而非按值是否不等(与 base *相等*的覆盖仍然是覆盖)。`describe({ redactSecrets: true })`——在每个 wire 面都强制启用——经由对 schema 的纯结构遍历(object/dict/array 容器;secret 角色子树整体是一个不透明叶节点)从全部三层剥除 `role('secret')` 子树,并把剥除的槽位枚举为 `{path, set}`,页面因此不必收到任何值就能渲染只写输入框。
**Host 识别并打开本地设置文档。** settings seam 暴露可选的 `documentPath` 提供方元数据和 `prepareDocument()` 操作;`settings-file` 返回已完全解析的自定义文件名或 `$DSH_HOME/settings.yaml` 文件名,并在文档缺失时以仅属主可访问的权限独占创建空文档,非文件提供方则保留基类的 `undefined`。仅限回环访问的 `settings.describe` 响应会在脱敏 namespace 视图旁只携带布尔型 `hasDocument` 能力。`ui-settings-general` 只在回环页面注册一条 `settings.action` 条目,只有元数据确认可准备好一份由提供方持有的本地文档后才显示,并调用无路径参数的 `settings.openDocument`;Host 会在文本文档交接前再次解析提供方路径(macOS 上使用 `open -t`,使任意 YAML 文件关联无法重定向这次操作;桌面 Linux 上使用 `xdg-open`;Windows 上使用 `Invoke-Item`;WSL 上先执行 `wslpath -w`,再使用同一 Windows 交接)。通用 Workspace 路径仍保留默认意图,包括针对浏览器可渲染文档的浏览器偏好。浏览器既不推导 `$DSH_HOME`,也不会收到文件系统目标;远程页面不会为这项操作发起特权 settings 读取。
**Host 识别并打开本地设置文档。** settings seam 暴露可选的 `documentPath` 提供方元数据和 `prepareDocument()` 操作;`settings-file` 返回已完全解析的自定义文件名或 `$DSH_HOME/settings.yaml` 文件名,并在文档缺失时以仅属主可访问的权限独占创建空文档,非文件提供方则保留基类的 `undefined`。经浏览器认证的 `settings.describe` 响应会在脱敏 namespace 视图旁只携带布尔型 `hasDocument` 能力。`ui-settings-general` 只在回环页面注册一条 `settings.action` 条目,只有元数据确认可准备好一份由提供方持有的本地文档后才显示,并调用无路径参数的 `settings.openDocument`;Host 会在文本文档交接前再次解析提供方路径(macOS 上使用 `open -t`,使任意 YAML 文件关联无法重定向这次操作;桌面 Linux 上使用 `xdg-open`;Windows 上使用 `Invoke-Item`;WSL 上先执行 `wslpath -w`,再使用同一 Windows 交接)。通用 Workspace 路径仍保留默认意图,包括针对浏览器可渲染文档的浏览器偏好。浏览器既不推导 `$DSH_HOME`,也不会收到文件系统目标;非 loopback 页面保留 Client 策略,不为这项操作发起 Host settings 读取。
**llm seam 声明可配置性并公布拓扑。**`registerConfigurableProviders()` 是一个全有或全无、以 fiber 为作用域的目录,条目为 `{provider, displayName, settingsNs, settingsPath}`——这正是配置页要为一条可能尚不存在的路由打开正确设置子树时所需要的寻址;`listConfigurableProviders()` 在 wire 处理器里与存活路由合并,未声明的存活路由因此仍报告为激活。零负载的 `'llm/adapters-updated'` 事件从全部四个注册/注销提交点触发,listener 派发带异常隔离(INVARIANT 重抛),沿用 settings/commands 的先例。`llm-deepseek` 的路由重命名为 `deepseek-official`,因为 pi-ai catalog 名正言顺地拥有 `deepseek` 这个聚合器条目;依预发布立场,不设别名。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-03-per-session-agent-presets.md
2026-08-03-per-session-agent-presets.md: b2195004580e1f4bf4be5527c61a8ee9b816c964
2026-08-03-per-session-agent-presets.zh.md: 9f689c74dd570d6ec9aaa7cd1274618326a836dd
2026-08-03-per-session-agent-presets.md: 97352a0e3376ce1fe26bac62b88c2c674202adc7
2026-08-03-per-session-agent-presets.zh.md: b62baf3e3097ba99247c2e86cc3fb5b7e43e2fab

View file

@ -53,7 +53,7 @@ Which preset an unnamed session gets is a user setting (`agent-presets.default`)
**Switching is allowed only while a session is blank.** Once a turn has run, that history was produced under the preset's tools and swapping them would strand logged tool calls, so `agentPreset.select` answers `agent-preset-locked`. A blank switch keeps the agent and the session and replaces only the subtree, because the host discards the `AgentHandle` it creates and there is no delete RPC — and keeping them is the better outcome anyway, since the session id, its workspace attachment, and its projections all stay put. The swap is unmount-then-mount (two compositions would register the same tool names into one layer), so it resolves the new preset before tearing anything down and restores the previous one when the new mount fails.
**Authoring a preset is an RPC, and a privileged one.** A composition is a file, but "edit it on the filesystem" is not a browser affordance, so the roster gained `read`/`write`/`remove` beside `select`. Those three are loopback-pinned: a composition names the plugins a session runs, so reading one is reconnaissance and writing one is arbitrary capability. `list` and `select` deliberately stay ordinary. The roster carries ids and trust only, and a LAN client's picker needs it; and choosing a preset looked like escalation — one of them mounts the toolset that edits the live runtime — but `session.create` already takes an `agentPreset`, so pinning only the switch would have left the same capability one method over. The capability is not the preset's to grant either: the deployment's own default already carries `bash` and the filesystem tools, so any caller that may start a session at all can already run commands as this process. Containment is a property of the id (`[a-z0-9][a-z0-9-]*`), checked before it becomes a directory name rather than by inspecting the joined path afterwards; the text is parsed with the loader's own schema and dialect, so a save cannot leave a file no session could load. Shipped presets are refused for writes and deletes, because the deployment's copy is what a broken local preset is compared against — which also makes "duplicate, then edit" the authoring path rather than an afterthought.
**Authoring a preset is an RPC, and a privileged one.** A composition is a file, but "edit it on the filesystem" is not a browser affordance, so the roster gained `read`/`write`/`remove` beside `select`. Connection authenticates authoring, `list`, `select`, and the complete Host API with one browser session: a composition names the plugins a session runs, so reading one is reconnaissance and writing one is arbitrary capability, while choosing a preset grants nothing `session.create` with `agentPreset` did not already grant. The capability is not the preset's to grant either: the deployment's own default already carries `bash` and the filesystem tools, so any caller that may start a session at all can already run commands as this process. Containment is a property of the id (`[a-z0-9][a-z0-9-]*`), checked before it becomes a directory name rather than by inspecting the joined path afterwards; the text is parsed with the loader's own schema and dialect, so a save cannot leave a file no session could load. Shipped presets are refused for writes and deletes, because the deployment's copy is what a broken local preset is compared against — which also makes "duplicate, then edit" the authoring path rather than an afterthought.
**A service with a consumer outside the agent plane cannot move into a preset.** The aggressive split moved the `subagents` registry and its spawn/fork backends into the delegation group's entry-local realm, and `dsh web` then failed to boot: `dsh-host-apiproxy` is a HOST row that injects `subagents` to answer the browser's cross-session queries (`listChildren`, `followup`), so it waited forever for a service only sessions now provided. A per-session copy is wrong twice over — a provider name registers once, so the second session would have collided anyway. The registry and every shared backend, including the [fixed Codex and Claude Code product providers](2026-08-10-product-subagent-providers-in-shared-host.md), are host-plane; a preset contributes whichever delegation TOOLS its agent should see, and those tools resolve the host registry. `workflows` stays entry-local because nothing outside an agent reads it. Grepping injectors is what should have caught this and did not: the search has to include the host packages, not just the agent-plane ones.

View file

@ -54,7 +54,7 @@ Status: implemented
**只有空白会话才允许切换。** 一旦跑过任何轮次,那段历史就是在该 preset 的工具下产生的,替换会留下无法执行的已记录 tool call,因此 `agentPreset.select` 返回 `agent-preset-locked`。空白期的切换保留 agent 与 session,只替换子树——因为宿主丢弃了它创建的 `AgentHandle`,也没有 delete RPC;而保留它们本身就是更好的结果,会话 id、workspace 挂接与 projections 都原地不动。该替换是"先卸后装"(两份组装会把同名工具注册进同一分层),因此它在拆除任何东西之前先解析新 preset,并在新组装装载失败时恢复原来的那一份。
**创作 preset 是一次 RPC,而且是特权 RPC。** 组装是一个文件,但“去文件系统里改它”并不是浏览器能提供的操作,因此名单在 `select` 之外新增了 `read`/`write`/`remove`。这三者被固定在环回地址:组装指明了一个会话所运行的插件,因此读取它是侦察,写入它是任意能力。`list` 与 `select` 刻意保持为普通方法。名单只携带 id 与信任级别,而局域网客户端的选择器需要它;至于选择本身,它看起来像提权——其中一个 preset 会挂载可编辑活动运行时的工具集——但 `session.create` 本就接受 `agentPreset`,只固定切换会把同一能力留在隔壁一个方法上。这份能力也不由 preset 授予:部署自带的默认 preset 本就带着 `bash` 与文件系统工具,因此任何被允许开启会话的调用方,早已能以本进程的身份执行命令。约束是 id 自身的性质(`[a-z0-9][a-z0-9-]*`),在它成为目录名之前就检查,而不是事后再去审视拼接出的路径;文本使用 loader 自身的 schema 与方言解析,因此保存不会留下任何会话都无法加载的文件。随部署提供的 preset 拒绝写入与删除,因为部署自带的那一份正是用来对照有问题的本地 preset 的——这也让“先复制、再编辑”成为创作路径本身,而非事后补充。
**创作 preset 是一次 RPC,而且是特权 RPC。** 组装是一个文件,但“去文件系统里改它”并不是浏览器能提供的操作,因此名单在 `select` 之外新增了 `read`/`write`/`remove`。Connection 用一个浏览器会话认证创作操作、`list`、`select` 与完整 Host API:组装指明一个会话所运行的插件,因此读取它是侦察,写入它是任意能力;选择 preset 则没有授予 `session.create` 携带 `agentPreset` 时尚未拥有的能力。这份能力也不由 preset 授予:部署自带的默认 preset 本就带着 `bash` 与文件系统工具,因此任何被允许开启会话的调用方,早已能以本进程的身份执行命令。约束是 id 自身的性质(`[a-z0-9][a-z0-9-]*`),在它成为目录名之前就检查,而不是事后再去审视拼接出的路径;文本使用 loader 自身的 schema 与方言解析,因此保存不会留下任何会话都无法加载的文件。随部署提供的 preset 拒绝写入与删除,因为部署自带的那一份正是用来对照有问题的本地 preset 的——这也让“先复制、再编辑”成为创作路径本身,而非事后补充。
**在 agent 平面之外还有消费方的服务,不能搬进 preset。** 激进拆分把 `subagents` 注册表连同 spawn/fork 后端一起搬进了 delegation 组的 entry-local realm,于是 `dsh web` 直接起不来:`dsh-host-apiproxy` 是宿主行,它注入 `subagents` 来回答浏览器的跨会话查询(`listChildren`、`followup`),因而永远等待一个此刻只有会话才提供的服务。按会话各一份在两个层面上都是错的——provider 名只能注册一次,第二个会话本来也会相撞。注册表与所有共享后端,包括[固定的 Codex 与 Claude Code 产品 provider](2026-08-10-product-subagent-providers-in-shared-host.zh.md),都属于宿主平面;preset 只贡献自己的 agent 应看见的委派**工具**,这些工具解析宿主注册表。`workflows` 保持 entry-local,因为 agent 之外没有任何东西读它。本该拦下它的是「检索注入方」这一步,而它没拦住:检索必须覆盖宿主包,而不只是 agent 平面的包。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-04-draft-provider-endpoint-interrogation.md
2026-08-04-draft-provider-endpoint-interrogation.md: 6a545ea5c97ef7e7f0e916c1676c1361f7a9bf3b
2026-08-04-draft-provider-endpoint-interrogation.zh.md: eb8133d0f2ec4a1999770468782db9c47ba170c8
2026-08-04-draft-provider-endpoint-interrogation.md: 0132dba5888faa1b9e6a4e59b45ee56a259eeef7
2026-08-04-draft-provider-endpoint-interrogation.zh.md: 8cea5d2809611696606113b2e38578f1b54b4e09

View file

@ -19,7 +19,7 @@ Interrogation is keyed by **settings namespace**, not by provider route:
- `ctx.llm.registerModelDiscovery(settingsNs, discover)` lets an adapter plugin offer to interrogate endpoints for the namespace it owns, and `ctx.llm.discoverModels(settingsNs, request)` asks. There is no way to enumerate which namespaces registered: a surface that cannot interrogate learns it from the refusal, and a list nothing consumed would be a required wire field doing nothing. The namespace is the right key because a configuration surface already holds it from the configurable-provider directory, and because a provider being added has no route to name.
- `LlmModelDiscoveryRequest` carries the draft — an optional `provider`, an optional `baseURL`, an optional `api`, an optional `apiKey`, and a signal — and needs at least one of `provider` or `baseURL` to have anything to answer about. `provider` exists because a route the adapter already describes is answered from its own registry with no network call at all; only a route it does not describe reaches an endpoint. Nothing in this path writes settings or credentials. The one read is the credential of a route the request names: a configuration surface holds a redacted descriptor rather than the stored secret, so the draft's `apiKey` is present only while the user is typing one, and without that read an already-configured route would be interrogated unauthenticated and answer 401. The typed key wins, being the one under test.
- `LlmDiscoveredModel` makes every field but `id` optional, because most listings disclose an id and nothing else. The reply is candidates, not a catalog: a surface adopting one still owes the capacities the adapter requires.
- `llm.discoverModels` carries the same draft over the wire. Its `apiKey` is the third and last payload on which a secret may ride, alongside `settings.update`/`mutate` and `credentials.set`, and it is never stored or echoed back. It does ride the client's outgoing envelope like every other secret-bearing payload, where a `subscribeEnvelopes()` observer can see it; redacting that tap is a configuration-plane-wide change, not this method's to make alone. The method is loopback-only for a second reason besides the key: it makes the host issue a GET to a caller-chosen URL and reports the outcome, which is a probe an anonymous LAN caller must not have. Every refusal folds into `model-discovery-failed`, whose message is the adapter's own text and whose details name the endpoint asked but never the credential offered.
- `llm.discoverModels` carries the same draft over the wire. Its `apiKey` is the third and last payload on which a secret may ride, alongside `settings.update`/`mutate` and `credentials.set`, and it is never stored or echoed back. It does ride the client's outgoing envelope like every other secret-bearing payload, where a `subscribeEnvelopes()` observer can see it; redacting that tap is a configuration-plane-wide change, not this method's to make alone. Connection authenticates the method with the complete Host API: it makes the host issue a GET to a caller-chosen URL and reports the outcome, which an anonymous caller must not receive. Every refusal folds into `model-discovery-failed`, whose message is the adapter's own text and whose details name the endpoint asked but never the credential offered.
`dsh-llm-pi-ai` implements the wire path as a plain `GET {baseURL}/models`, reading `openai-completions` and `openai-responses`: their `GET /models` shape with bearer auth is the one a gateway, a self-hosted server, and the official endpoints all agree on. Azure is excluded despite its OpenAI lineage — it authenticates with an `api-key` header and requires an `api-version` query — and Codex uses OAuth; both would have reported an authentication failure as a provider with no models. Every other protocol answers `DISCOVERY_UNSUPPORTED`, so the surface falls back to hand-entry rather than reporting a guessed response shape as an empty provider. `baseURL` is treated as a prefix rather than a URL to resolve against, so a deployment path such as `https://gateway.example/openai/v1` keeps its segments. The reply is read under a four-megabyte ceiling enforced on the bytes actually received — the endpoint is a URL the user typed, so a declared `content-length` is checked first as a courtesy but never trusted as the bound, matching `dsh-web-fetch`'s two-stage shape for its own caller-supplied URLs.

View file

@ -19,7 +19,7 @@ Status: implemented
- `ctx.llm.registerModelDiscovery(settingsNs, discover)` 让适配器插件为自己拥有的 namespace 提供「询问端点」的能力,`ctx.llm.discoverModels(settingsNs, request)` 发起询问。没有任何办法枚举哪些 namespace 注册过:询问不了的界面会从那句拒绝里知道,而一份无人消费的列表只会变成一个什么都不做的必填协议字段。以 namespace 为键是对的,因为配置界面已经从可配置提供方目录里拿到了它,也因为正在新增的提供方没有路由可点名。
- `LlmModelDiscoveryRequest` 携带草稿——可选的 `provider`、可选的 `baseURL`、可选的 `api`、可选的 `apiKey`,以及一个 signal——且 `provider` 与 `baseURL` 至少要有一个,才有东西可答。`provider` 之所以存在,是因为适配器已经描述过的路由直接由它自己的注册表作答、完全不联网;只有它未描述的路由才会抵达某个端点。这条路径不写 settings 与 credentials。唯一的读取是请求所点名路由的凭据:配置界面拿到的是脱敏描述符而非已存的机密,因此草稿里的 `apiKey` 只在用户正键入时才存在;没有这次读取,已配置好的路由就会被不带认证地询问,只换回一个 401。键入的密钥优先,因为那正是被测试的那一把。
- `LlmDiscoveredModel` 除 `id` 外每个字段都可选,因为大多数列表只公布 id。回复是候选而非 catalog:采纳其中一条的界面仍要补上适配器所需的容量。
- `llm.discoverModels` 把同一份草稿送过协议层。它的 `apiKey` 是可承载机密的第三个、也是最后一个载荷(另两个是 `settings.update`/`mutate` 与 `credentials.set`),且绝不被存储或回显。它确实会像其他承载机密的载荷一样随客户端外发信封同行,`subscribeEnvelopes()` 观察者看得到;把那个抽头脱敏是整个配置面的改动,不该由这一个方法独自决定。除密钥之外,将它限制为仅可通过回环访问还有第二个理由:它让宿主向调用方选定的 URL 发起 GET 并回报结果,这是匿名 LAN 调用者不该拥有的探测能力。每一种拒绝都折叠为 `model-discovery-failed`,其消息是适配器自己的文本,details 点名被询问的端点,绝不点名所提供的凭据。
- `llm.discoverModels` 把同一份草稿送过协议层。它的 `apiKey` 是可承载机密的第三个、也是最后一个载荷(另两个是 `settings.update`/`mutate` 与 `credentials.set`),且绝不被存储或回显。它确实会像其他承载机密的载荷一样随客户端外发信封同行,`subscribeEnvelopes()` 观察者看得到;把那个抽头脱敏是整个配置面的改动,不该由这一个方法独自决定。Connection 用与完整 Host API 相同的会话认证该方法:它让宿主向调用方选定的 URL 发起 GET 并回报结果,匿名调用者绝不能获得这类探测能力。每一种拒绝都折叠为 `model-discovery-failed`,其消息是适配器自己的文本,details 点名被询问的端点,绝不点名所提供的凭据。
`dsh-llm-pi-ai` 的实现只是一次朴素的 `GET {baseURL}/models`,且仅限 OpenAI 兼容协议。它们的列表形状是网关、自建服务与官方端点三方一致认可的那一种,而这正是该动作存在的场景。其余协议一律以 `DISCOVERY_UNSUPPORTED` 回答,让界面回退到手工填写,而不是把猜错的响应形状报成一个空提供方。`baseURL` 按前缀而非待解析 URL 处理,因此 `https://gateway.example/openai/v1` 这类部署路径会保留其路径段。回复在四兆字节上限下读取,且上限落在实际收到的字节上——端点是用户自己填的 URL,因此会先看声明的 `content-length` 作为善意提示,但绝不把它当作边界;这与 `dsh-web-fetch` 面对自己的调用方提供 URL 时所用的两段式形状一致。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-04-websocket-downlink-carrier.md
2026-08-04-websocket-downlink-carrier.md: f757487c7b0880cfff25b93644341dc85b5e0e38
2026-08-04-websocket-downlink-carrier.zh.md: 5fa603b9a8dc66146c777f54dfbdc25cb131debd
2026-08-04-websocket-downlink-carrier.md: 420a0d30f31cca58448adc0afd46a5d6d5e9107f
2026-08-04-websocket-downlink-carrier.zh.md: 5f277a4ed33ffe97b51587fef960746b93eff1c1

View file

@ -16,7 +16,7 @@ WebSocket carries only the host→browser downlink. All client→host unary call
## Upgrade and lifecycle boundaries
`dsh-host-webserver` provides an exact upgrade-route registration point alongside ordinary routes, dispatches Node upgrade sockets by pathname only, contains raw-socket errors, and waits for surviving upgraded connections to close during server teardown; it knows nothing about Harness frames or WebSocket messages. `dsh-client-connection` owns the WebSocket handshake, frame output, and stream cancellation, and reuses the `/api` Host/Origin trust fence before upgrade. An untrusted authority or cross-origin Origin is rejected before `ctx.apiProxy.events.*` starts.
`dsh-host-webserver` provides an exact upgrade-route registration point alongside ordinary routes, dispatches Node upgrade sockets by pathname only, contains raw-socket errors, and waits for surviving upgraded connections to close during server teardown; it knows nothing about Harness frames or WebSocket messages. `dsh-client-connection` owns the WebSocket handshake, frame output, and stream cancellation. Before upgrade it applies the `/api` Host/Origin checks followed by the same signed browser-cookie authentication as unary HTTP. An untrusted authority or cross-origin Origin receives 403; a trusted but unauthenticated request receives 401; neither starts a Remote stream.
A browser abort or socket close cancels the corresponding host stream; plugin teardown also waits for that source iterator's cleanup. If a host stream throws midway, the carrier sends one existing `stream/error` frame and then closes the socket; the client treats that frame as connection loss rather than delivering it to a business sink. Each WebSocket reports open independently, and the existing readiness handshake still waits until mux and host are both open and the `host.describe` HTTP call has succeeded before publishing connected.

View file

@ -16,7 +16,7 @@ WebSocket 只承担 host→browser 下行。所有 client→host unary 调用和
## Upgrade 与生命周期边界
`dsh-host-webserver` 提供与普通 route 并列的精确 upgrade-route 注册点,只按 pathname 分发 Node upgrade socket,隔离原始 socket 错误,并在 server teardown 期间等待仍存活的升级连接关闭;它不认识 Harness 帧或 WebSocket 消息。`dsh-client-connection` 拥有 WebSocket handshake、frame 写出和流取消,并在 upgrade 前复用 `/api` 的 Host/Origin 信任栅栏。未受信任的 authority 或跨来源 Origin 在 `ctx.apiProxy.events.*` 启动前即被拒绝。
`dsh-host-webserver` 提供与普通 route 并列的精确 upgrade-route 注册点,只按 pathname 分发 Node upgrade socket,隔离原始 socket 错误,并在 server teardown 期间等待仍存活的升级连接关闭;它不认识 Harness 帧或 WebSocket 消息。`dsh-client-connection` 拥有 WebSocket handshake、frame 写出和流取消。upgrade 前先执行 `/api` Host/Origin 校验,再执行与一元 HTTP 相同的签名浏览器 cookie 认证。未受信任的 authority 或跨来源 Origin 得到 403;Host 可信但未认证的请求得到 401;两者都不会启动 Remote stream。
浏览器 abort 或 socket close 会取消对应的 host 流;插件 teardown 还会等待该 source iterator 完成清理。host 流中途抛错时,载体发送一个现有的 `stream/error` frame 后关闭 socket;客户端把该 frame 收敛为连接丢失,不投递给业务 sink。每条 WebSocket 独立报告 open,既有 readiness handshake 仍等待 mux、host 都 open 且 `host.describe` HTTP 调用成功后才发布 connected。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-06-subagent-list-identity-projection.md
2026-08-06-subagent-list-identity-projection.md: 2e88639eade2b5a83ef491bcb93693d432bd676b
2026-08-06-subagent-list-identity-projection.zh.md: 56fa5bad853a5c089bb19e58cbc180c3e830c1b7
2026-08-06-subagent-list-identity-projection.md: d48d6f869b49861ac54d19c9efc9bda42cf8e575
2026-08-06-subagent-list-identity-projection.zh.md: a5765a98bbd6aa5aa8552fa80a0e253d488b790e

View file

@ -165,7 +165,7 @@ Consuming surfaces: diagnostic handling across wire, tool, and GUI **stays entir
## Verification
`packages/subagent/subagent/tests/list-children.spec.ts` is rewritten to this contract: live-only listing without persistence, query services, or the continuation runtime; with the registry absent, even zero children loudly report `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`; a live child incurs zero `inspect` throughout while a cold child incurs exactly one per listing; multiple descriptors resolve last-wins to the final one; corrupt payloads and unknown versions fold to `corrupt`; a cold-read failure maps to `unavailable` and retries on the next listing; the ancestor descriptor in a fork seed forms a row under that identity (pinning deviation one); ordinary forks and descendants without a subagent origin neither enter the list nor count toward `hasChildren`; `createdAt`-then-id ordering; an unmounted provider does not affect listing; compacted and uncompacted twins list identically; the three cases of pre-abort, persistence listing, and cold-read cancellation all normalize to `CANCELLED`; the empty list and stable error codes. A hostile-unit dual-path probe (`apply` lazily poisons, `view` detonates) proves that any registered unit's fold/schema throw on this child's log is contained as that child's `corrupt` row on both the live and the cold retrieval paths, with siblings and the listing itself unaffected. Second-rung cases: an own-seq identity used directly with zero `inspect`, a fork seed's ancestor identity (seq inside the seed range) rejected by the gate and falling through, an in-row identity absence (null sentinel or absent key) falling through, an absent cache service falling through, and a poisoned cache row silently falling through to the refold; cold-path lifecycle tampering degrades to `corrupt` field by witness field (`it.each` over the seven). The `tool-subagent-control` list-agents tests are updated for the narrowed load requirement; `optional-session-query.spec.ts` is deleted with the dependency it guarded; the existing keyless snapshots (`subagent-list-agents` among others) are unchanged, pinning that the healthy path's wire and model-visible surfaces did not move; a new keyless snapshot, `subagent-diagnostic` (examples/headless-agent), pins the four-state mapping's diagnostic classification — the model-visible changes such as descriptor-less settled debris becoming a `corrupt` row.
`packages/subagent/subagent/tests/list-children.spec.ts` is rewritten to this contract: live-only listing without persistence, query services, or the continuation runtime; with the registry absent, even zero children loudly report `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`; a live child incurs zero `inspect` throughout while a cold child incurs exactly one per listing; multiple descriptors resolve last-wins to the final one; corrupt payloads and unknown versions fold to `corrupt`; a cold-read failure maps to `unavailable` and retries on the next listing; the ancestor descriptor in a fork seed forms a row under that identity (pinning deviation one); ordinary forks and descendants without a subagent origin neither enter the list nor count toward `hasChildren`; `createdAt`-then-id ordering; an unmounted provider does not affect listing; compacted and uncompacted twins list identically; the three cases of pre-abort, persistence listing, and cold-read cancellation all normalize to `CANCELLED`; the empty list and stable error codes. A hostile-unit dual-path probe (`apply` lazily poisons, `view` detonates) proves that any registered unit's fold/schema throw on this child's log is contained as that child's `corrupt` row on both the live and the cold retrieval paths, with siblings and the listing itself unaffected. Second-rung cases: an own-seq identity used directly with zero `inspect`, a fork seed's ancestor identity (seq inside the seed range) rejected by the gate and falling through, an in-row identity absence (null sentinel or absent key) falling through, an absent cache service falling through, and a poisoned cache row silently falling through to the refold; cold-path lifecycle tampering degrades to `corrupt` field by witness field (`it.each` over the seven). The `tool-subagent-control` list-agents tests are updated for the narrowed load requirement; `optional-session-query.spec.ts` is deleted with the dependency it guarded; the existing keyless snapshots (`subagent-list-agents` among others) are unchanged, pinning that the healthy path's wire and model-visible surfaces did not move; the owner-local `apps/cli/tests/profiles/headless/tests/subagent-diagnostic.expected.e2e.ts` pins the four-state mapping's diagnostic classification — the model-visible changes such as descriptor-less settled debris becoming a `corrupt` row.
## Consequences

View file

@ -165,7 +165,7 @@ export type SubagentListEntry =
## 验证
`packages/subagent/subagent/tests/list-children.spec.ts` 重写为本约定:无 persistence、query 服务与继续运行时的 live-only 列表;registry 缺席时零 children 也响亮报 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`;live child 全程零 `inspect`、cold child 每次列表恰一次;多描述符 last-wins 取末者;损坏载荷与未知版本折为 `corrupt`;冷读失败映射 `unavailable` 且下次列表重试;fork seed 里的祖先描述符按该身份成行(偏差一钉住);普通 fork 与无 subagent origin 的后代不入列也不计入 `hasChildren`;`createdAt`→id 排序;提供方未挂载不影响列表;压缩与未压缩孪生一致;预中止、持久化列表与冷读取消三例归一 `CANCELLED`;空列表与稳定错误码。敌意 unit 双路探针(`apply` 惰性置毒、`view` 引爆)证明任一注册 unit 在该 child 日志上的 fold/schema 抛错,在 live 与 cold 两条取值路径上都收纳为该 child 的 `corrupt` 行,sibling 与列表本身不受影响。第二级例:own-seq 身份直用零 `inspect`、fork 种子祖先身份(seq 落在 seed 区间)被门拒绝落底、行内无身份(null 哨兵或 key 缺席)落底、cache 服务缺席落底、缓存行中毒静默落底重折;冷路径 lifecycle 篡改按见证七字段逐一(`it.each`)降级为 `corrupt`。`tool-subagent-control` 的 list-agents 测试随加载要求收窄更新;`optional-session-query.spec.ts` 随依赖消失删除;既有无密钥快照(`subagent-list-agents` 等)零变化,钉住健康路径的 wire 与 model-visible 面不变;新增无密钥快照 `subagent-diagnostic`(examples/headless-agent)钉住四态映射的诊断分类——descriptor-less 定局残骸成 `corrupt` 行等模型可见变化。
`packages/subagent/subagent/tests/list-children.spec.ts` 重写为本约定:无 persistence、query 服务与继续运行时的 live-only 列表;registry 缺席时零 children 也响亮报 `SUBAGENT_CONTROL_PROJECTIONS_UNAVAILABLE`;live child 全程零 `inspect`、cold child 每次列表恰一次;多描述符 last-wins 取末者;损坏载荷与未知版本折为 `corrupt`;冷读失败映射 `unavailable` 且下次列表重试;fork seed 里的祖先描述符按该身份成行(偏差一钉住);普通 fork 与无 subagent origin 的后代不入列也不计入 `hasChildren`;`createdAt`→id 排序;提供方未挂载不影响列表;压缩与未压缩孪生一致;预中止、持久化列表与冷读取消三例归一 `CANCELLED`;空列表与稳定错误码。敌意 unit 双路探针(`apply` 惰性置毒、`view` 引爆)证明任一注册 unit 在该 child 日志上的 fold/schema 抛错,在 live 与 cold 两条取值路径上都收纳为该 child 的 `corrupt` 行,sibling 与列表本身不受影响。第二级例:own-seq 身份直用零 `inspect`、fork 种子祖先身份(seq 落在 seed 区间)被门拒绝落底、行内无身份(null 哨兵或 key 缺席)落底、cache 服务缺席落底、缓存行中毒静默落底重折;冷路径 lifecycle 篡改按见证七字段逐一(`it.each`)降级为 `corrupt`。`tool-subagent-control` 的 list-agents 测试随加载要求收窄更新;`optional-session-query.spec.ts` 随依赖消失删除;既有无密钥快照(`subagent-list-agents` 等)零变化,钉住健康路径的 wire 与 model-visible 面不变;归属方本地的 `apps/cli/tests/profiles/headless/tests/subagent-diagnostic.expected.e2e.ts` 钉住四态映射的诊断分类——descriptor-less 定局残骸成 `corrupt` 行等模型可见变化。
## 后果

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-08-bounded-session-persistence-write-batching.md
2026-08-08-bounded-session-persistence-write-batching.md: 6d77ef90276dcf143bb63019f801379b702b43b2
2026-08-08-bounded-session-persistence-write-batching.zh.md: e773f20ade706e97a5f84d4a4cc15cc9d3b2d836
2026-08-08-bounded-session-persistence-write-batching.md: fc22a10537ebb9009ab1ae1ec21ca625c9025651
2026-08-08-bounded-session-persistence-write-batching.zh.md: c2763e157fcfc2e004b14116694054c604fbfb9c

View file

@ -12,7 +12,7 @@ Dropping chunk events or replacing them with assembled messages would reduce log
### Quantified baseline
Repository fixtures make the logical volume concrete. Decoding the current packed rows in [`goal-multi-turn-actions`](../../../../apps/web/tests/snapshots/goal-multi-turn-actions/session.jsonl) yields 2,098 events: 2,017 chunks (96.1%). Their unpacked JSONL lines occupy 332,647 of 379,225 event bytes (87.7%), while chunk packing reduces the committed file to 89,176 bytes and 182 storage rows, including 23 packed chunk rows. [`permission-policy-context`](../../../../apps/web/tests/snapshots/permission-policy-context/session.jsonl) yields 813 events: 746 chunks (91.8%) and 118,935 of 184,821 unpacked event bytes (64.4%); its packed file is 84,917 bytes and 123 storage rows, including 14 packed rows. These are tracked deterministic fixtures, not a production workload distribution, but they demonstrate why deleting chunks would reduce logical volume and why the existing packed-row layout already removes much of their JSON envelope cost.
Repository fixtures make the logical volume concrete. Decoding the current packed rows in [`goal-multi-turn-actions`](../../../../snapshots/web/goal-multi-turn-actions/session.jsonl) yields 2,098 events: 2,017 chunks (96.1%). Their unpacked JSONL lines occupy 332,647 of 379,225 event bytes (87.7%), while chunk packing reduces the committed file to 89,176 bytes and 182 storage rows, including 23 packed chunk rows. [`permission-policy-context`](../../../../snapshots/web/permission-policy-context/session.jsonl) yields 813 events: 746 chunks (91.8%) and 118,935 of 184,821 unpacked event bytes (64.4%); its packed file is 84,917 bytes and 123 storage rows, including 14 packed rows. These are tracked deterministic fixtures, not a production workload distribution, but they demonstrate why deleting chunks would reduce logical volume and why the existing packed-row layout already removes much of their JSON envelope cost.
SQLite stores one row per logical event, so those same logical logs would retain 2,098 and 813 event rows respectively; batching does not change those counts. JSONL writes one Zstandard frame and fsync per durable append batch, while SQLite performs one transaction and one session-revision increment per batch. Runtime files do not record former append boundaries, so fixture row counts cannot honestly be presented as fsync or transaction counts.

View file

@ -12,7 +12,7 @@ Status: implemented
### 量化基线
仓库 fixture(测试前置数据)让逻辑数据量有了具体依据。对当前 [`goal-multi-turn-actions`](../../../../apps/web/tests/snapshots/goal-multi-turn-actions/session.jsonl) 中的打包行进行解码,可得到 2,098 个事件,其中 2,017 个是分片(96.1%)。这些分片解包后的 JSONL 行共 332,647 字节,占全部事件 379,225 字节的 87.7%;分片打包则把仓库中的已提交文件缩小到 89,176 字节和 182 个存储行,其中包括 23 个打包分片行。[`permission-policy-context`](../../../../apps/web/tests/snapshots/permission-policy-context/session.jsonl) 可得到 813 个事件,其中 746 个是分片(91.8%);这些分片解包后的 JSONL 行共 118,935 字节,占全部事件 184,821 字节的 64.4%。其打包文件为 84,917 字节,共 123 个存储行,其中包括 14 个打包行。这些是纳入版本控制的确定性 fixture,不代表生产工作负载分布;但它们说明了删除分片为何会降低逻辑数据量,也说明现有打包行布局已经消除了大量 JSON 包装开销。
仓库 fixture(测试前置数据)让逻辑数据量有了具体依据。对当前 [`goal-multi-turn-actions`](../../../../snapshots/web/goal-multi-turn-actions/session.jsonl) 中的打包行进行解码,可得到 2,098 个事件,其中 2,017 个是分片(96.1%)。这些分片解包后的 JSONL 行共 332,647 字节,占全部事件 379,225 字节的 87.7%;分片打包则把仓库中的已提交文件缩小到 89,176 字节和 182 个存储行,其中包括 23 个打包分片行。[`permission-policy-context`](../../../../snapshots/web/permission-policy-context/session.jsonl) 可得到 813 个事件,其中 746 个是分片(91.8%);这些分片解包后的 JSONL 行共 118,935 字节,占全部事件 184,821 字节的 64.4%。其打包文件为 84,917 字节,共 123 个存储行,其中包括 14 个打包行。这些是纳入版本控制的确定性 fixture,不代表生产工作负载分布;但它们说明了删除分片为何会降低逻辑数据量,也说明现有打包行布局已经消除了大量 JSON 包装开销。
SQLite 每个逻辑事件存储一行,因此同样的逻辑日志会分别保留 2,098 和 813 个事件行;批处理不会改变这些数量。JSONL 每个持久化追加批次会写入一个 Zstandard 帧并执行一次 fsync,SQLite 每个批次会执行一次事务并递增一次会话修订版本。运行时文件不记录原有追加边界,因此不能把 fixture 的存储行数当作 fsync 或事务次数。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-09-client-conversation-node-assembly.md
2026-08-09-client-conversation-node-assembly.md: e6c0e790a361265870a04ee63301b9f11940c648
2026-08-09-client-conversation-node-assembly.zh.md: 702ddba0019e125d3976727f841db775276b3b77
2026-08-09-client-conversation-node-assembly.md: ea2505d4a72f483a9df6fcd78d7e5c9a96b02f5c
2026-08-09-client-conversation-node-assembly.zh.md: b87f127d753cadf2805ed5cd948fc58ad01830aa

View file

@ -260,6 +260,7 @@ Page size, the number of history loads, and RAF coalescing affect only when evid
| Next-turn Inbox / `inbox-next-turn` | Splice Event seq | Each `agent/inbox/spliced` targeting next-turn | None | Apply the current splice to the pending/claimed instantaneous state from `reader.previous(ownKind)` |
| Next-step Inbox / `inbox-next-step` | Splice Event seq | Each `agent/inbox/spliced` targeting next-step | None | Build the same per-instruction instantaneous state; Message reads its claimed set |
| Message / `input-message` | Message ID | Append-surface `user/message` | None | Use source for a context message, or read the nearest next-step Inbox to distinguish user from steering |
| Request Prompt / `request-prompt` | Header Event seq | Each `request/header` | None | Read the preceding Request Prompt through Reader, retain the full prompt state, and classify system/tool changes |
| Assistant / `assistant-step` | `turn:step` | `step/start` | `assistant/chunk`, final `assistant/message`, and same-step Retry | Aggregate blocks, usage, first-token time, final evidence, and retry-hidden state, then publish same-key Step data |
| Tool / `tool-call` | Root call ID | Root `tool/call` | Root result and Code Dispatch start/result | Aggregate the root, children, and parent Map; Dispatch Events route exactly through `rootCallId` |
| Command / `command` | Command ID | `command/run` | `command/done` and compact lifecycle/checkpoint Events carrying a source command ID | Aggregate command outcome and manual-compaction evidence |
@ -276,6 +277,7 @@ Page size, the number of history loads, and RAF coalescing affect only when evid
|---|---|---|---|
| Inbox | `none` | No Node | Recompute instantaneous states along the Reader chain when prepend supplies earlier splices |
| Message | Immediate by default | `user`, `steering`, or `context` | Window-gap repair can reclassify the same message key |
| Request Prompt | Immediate by default | One `system-prompt` for every header carrying a non-empty system field | A step's first header anchors before its request messages; a later same-step series anchors after its surface rewrite; prepend of the preceding header can correct a partial-window anchor |
| Assistant | RAF for chunks, immediate for final, none for pure usage/finish | Same-key `assistant-step` with running/settled/interrupted status | Matches support fallback without `step/start`; Location close produces interruption presentation |
| Tool | Immediate by default | One recursive `tool-call` root containing all `subCalls` | A result-only history window supports fallback; running→settled retains its key |
| Command | Immediate by default | Ordinary `command` or integrated `manual-compaction` | Checkpoint arrival may change the anchor without changing the Context key |
@ -288,6 +290,8 @@ Page size, the number of history loads, and RAF coalescing affect only when evid
Inbox demonstrates that every Event can be a start-only instantaneous-state Context; not every business requires a start/update pair. Reader links each state to the prior same-kind Context instead of inventing a lifecycle ID for the entire Inbox.
Request Prompt demonstrates shared pure interpretation without shared target State: Chat and Trajectory call `inspectRequestPrompt()` from their own Definitions. The function canonicalizes the full header and classifies model-visible system/tool differences; each target then chooses its own output. Chat materializes every header carrying a non-empty system field, including `series` snapshots that repeat an unchanged header for an explicitly declared series or a post-replacement request, while Trajectory retains the complete request fact and its change classification. Ordinary append-only later Turns do not write another unchanged header. The first header in a Step follows the provider envelope rather than the header Event position: step one uses the owning Turn start and later steps use their Step start, placing the system field before the request's user-role messages; a later header in the same Step stays at its own Event after the surface rewrite that began the new series. When the preceding header is outside a partial window, a non-`initial` header stays at its own Event until prepend supplies that predecessor. Every header is a full snapshot, so a first loaded `resume`, `change`, or `series` header can render its system field without fabricating a comparison to unloaded history.
Retry, Assistant, and Turn Tail demonstrate independent claims on one Event. Each Definition updates only its own State and produces its own atomic Chat Node.
Assistant, Turn Tail, and Deliverables demonstrate layered Location data composition. Assistant writes `assistant-step` data for each Step; Turn Tail derives `turn-tail` data from those Step values; Deliverables independently maintains `deliverables` data for the same Turn. Consumers read only declaration-merged keys, do not scan another business's Nodes, and cannot obtain the provider's Context State.

View file

@ -260,6 +260,7 @@ Chat `order` 的结构性变化仍可能重排当前可见 key;纯 data 更新
| Next-turn Inbox / `inbox-next-turn` | splice Event seq | 每条目标为 next-turn 的 `agent/inbox/spliced` | 无 | 从 `reader.previous(ownKind)` 的 pending/claimed 瞬间态应用当前 splice |
| Next-step Inbox / `inbox-next-step` | splice Event seq | 每条目标为 next-step 的 `agent/inbox/spliced` | 无 | 同样形成逐指令瞬间态,claimed 集合供 Message 读取 |
| Message / `input-message` | message ID | append-surface `user/message` | 无 | 根据 source 生成 context message,或读取最近 next-step Inbox 判断 user/steering |
| Request Prompt / `request-prompt` | header Event seq | 每条 `request/header` | 无 | 通过 Reader 读取前一条 Request Prompt,保留完整 prompt 状态,并判定 system/tool 变化 |
| Assistant / `assistant-step` | `turn:step` | `step/start` | `assistant/chunk`、final `assistant/message`、同 step Retry | 聚合 blocks、usage、首 token 时间、final 和 retry 隐藏状态,并发布同 key Step data |
| Tool / `tool-call` | root call ID | root `tool/call` | root result、Code Dispatch start/result | 聚合 root、children 和 parent Map;Dispatch Event 用 `rootCallId` 精确路由 |
| Command / `command` | command ID | `command/run` | `command/done`、带 source command ID 的 compact lifecycle/checkpoint | 聚合 command outcome 和手动压缩证据 |
@ -276,6 +277,7 @@ Chat `order` 的结构性变化仍可能重排当前可见 key;纯 data 更新
|---|---|---|---|
| Inbox | `none` | 不生成 Node | prepend 补前序 splice 时沿 Reader 链重算瞬间态 |
| Message | 默认 immediate | `user`、`steering` 或 `context` | window gap 修复可让同一 message key 重新分类 |
| Request Prompt | 默认 immediate | 每条带非空 system 字段的 header 都生成一个 `system-prompt` | Step 首条 header 锚定在请求消息之前;同 step 后续序列锚定在表层改写之后;prepend 补入前序 header 后可纠正部分窗口的锚点 |
| Assistant | chunk 为 RAF,final immediate,纯 usage/finish 为 none | 同 key `assistant-step`,状态为 running/settled/interrupted | 缺 `step/start` 可先用 Matches fallback;Location close 生成中断表现 |
| Tool | 默认 immediate | 一个递归 `tool-call` root,包含全部 `subCalls` | result-only 历史窗口可 fallback;running→settled 保持 key |
| Command | 默认 immediate | 普通 `command` 或集成 `manual-compaction` | checkpoint 到达可改变 anchor,但不改变 Context key |
@ -288,6 +290,8 @@ Chat `order` 的结构性变化仍可能重排当前可见 key;纯 data 更新
Inbox 展示了“每条 Event 都是一个 start-only 瞬间态 Context”,不是所有业务都需要 start/update 配对。它通过 Reader 与前一个同 kind Context 形成连续 fold,而非给整个 Inbox 人工制造生命周期 ID。
Request Prompt 展示了如何在不共享 target State 的前提下共用纯解释逻辑:Chat 与 Trajectory 各自在自己的 Definition 中调用 `inspectRequestPrompt()`。该函数规范化完整 header,并判定面向模型的 system/tool 差异;随后每个 target 自行选择产物。Chat 会物化每条带非空 system 字段的 header,包括为显式声明的序列或表层替换后的请求重复未变 header 的 `series` 快照;Trajectory 则保留完整请求事实及其变化分类。普通的仅追加后续 Turn 不会再次写入未变 header。一个 Step 中的首条 header 遵循提供方信封,而不是 header Event 位置:step one 使用所属 Turn start,后续 step 使用各自的 Step start,把 system 字段放到该请求的 user-role 消息之前;同一 Step 的后续 header 保留在开启新序列的表层改写之后。部分窗口未包含前序 header 时,非 `initial` header 会保留在自身 Event,直到 prepend 补入该前序 header。每条 header 都是完整快照,因此已加载窗口中的首条 `resume`、`change` 或 `series` header 无需凭空构造与未加载历史的比较,也能渲染其 system 字段。
Retry、Assistant 和 Turn Tail 展示了同一 Event 被多个 Definition 独立认领。每个 Definition 只更新自己的 State,最终分别生成原子 Chat Node。
Assistant、Turn Tail 和 Deliverables 展示了 Location data 的分层组合。Assistant 负责写好每个 Step 的 `assistant-step` data;Turn Tail 从这些 Step values 计算 `turn-tail` data;Deliverables 独立维护同一 Turn 的 `deliverables` data。消费者只读取声明合并后的 key,不扫描其他业务 Node,也不取得提供方的 Context State。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-10-fork-children-stay-one-shot.md
2026-08-10-fork-children-stay-one-shot.md: b2d9a77d7cc9969e517a4f5d6973aa2aee1134f5
2026-08-10-fork-children-stay-one-shot.zh.md: b5dc1a7fda4e2b4152baf63e9c49b89bcebdeef0
2026-08-10-fork-children-stay-one-shot.md: ba1e99c4d78d14230a2199cd7fb3c3eb9d3ad754
2026-08-10-fork-children-stay-one-shot.zh.md: 0d4e214be952f5a4d96c296f38a5b0dbfbd85c86

View file

@ -1,4 +1,4 @@
# Agent Note: Cache-preserving forked children stay one-shot
# Agent Note: Forked children stay one-shot
Status: implemented
@ -12,7 +12,7 @@ The child-scoped `report` return channel is now the largest such addition, and s
## Decision
The cache-preserving compositions bind the fork delegation tool to `backgroundMode: one-shot`: [the base bundle](../../../../packages/bundle/base/cordis.patch.yml), [the ACP example](../../../../examples/acp-agent/cordis.yml), and [the headless example](../../../../examples/headless-agent/cordis.yml). The base bundle leaves `run_in_background` available, because it mounts a task service; the two examples set `enableRunInBackground: false`, because they mount none and a one-shot background start would otherwise fail at call time on a missing `tasks` service. The standard, code, and Cordis CLI presets instead bind fork to `continuable`; their child-scoped `report` additions invalidate the inherited prefix and accept the recomputation cost described here.
Every shipped composition inherits the fork delegation tool's `backgroundMode: one-shot` from the [base bundle](../../../../packages/bundle/base/cordis.patch.yml). The base bundle leaves `run_in_background` available because it also mounts the task service needed to settle background work.
One-shot children — foreground and background alike — are created through `SubagentRuntime.start()`, which never enters the continuable activation-setup registry, so neither `report` nor its prompt section is installed. A forked one-shot child's system prompt and tool schemas therefore equal its parent's, apart from the `persona` and `toolFilter` deltas a deployment opts into per delegation tool.
@ -20,9 +20,9 @@ One-shot children — foreground and background alike — are created through `S
### The restriction is composition, not code
`ForkInProcessProvider.prepareContinuable` stays implemented and `ctx.subagents.startContinuable()` accepts `fork`; composition chooses whether the fork tool is one-shot or continuable. `tool-subagent` knows both the provider's `inheritsParentContext` and its own `backgroundMode` at mount, so a load-time rejection of the pair is available and deliberately absent: the pair is not wrong in general. It is costly only while a child-scope delta precedes inherited history, and the package that creates that delta — [`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md) — is separately installable and, by its own design, invisible to `tool-subagent`. A deployment that omits the report package can run continuable forked children with the prefix intact. Encoding one roster's consequence as a delegation-tool invariant would make the tool assert something it cannot observe.
`ForkInProcessProvider.prepareContinuable` stays implemented and `ctx.subagents.startContinuable()` still accepts `fork`; only the shipped `cordis.yml` rows changed. `tool-subagent` knows both the provider's `inheritsParentContext` and its own `backgroundMode` at mount, so a load-time rejection of the pair was available and is deliberately not added: the pair is not wrong in general. It is wrong only while a child-scope delta precedes inherited history, and the package that creates that delta — [`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.md) — is separately installable and, by its own design, invisible to `tool-subagent`. A deployment that omits the report package can run continuable forked children with the prefix intact. Encoding one roster's consequence as a delegation-tool invariant would make the tool assert something it cannot observe.
The cache-preserving condition is recorded as a `TODO(fork-continuable-prefix-reuse)` marker on `prepareContinuable` and tracked as issue #2124: continuable fork preserves its inherited prefix when the child's system prompt and tool schemas can match the parent's byte for byte.
The reintroduction condition is recorded as a `TODO(fork-continuable-prefix-reuse)` marker on `prepareContinuable` itself, the one method the shipped compositions do not call, and tracked as issue #2124: continuable fork reopens when a child's system prompt and tool schemas can match its parent's byte for byte.
## Alternatives considered
@ -30,7 +30,7 @@ The cache-preserving condition is recorded as a `TODO(fork-continuable-prefix-re
**Stop mounting the fork provider at all.** This was the broader form of the restriction. Rejected because foreground fork *is* the prefix-reusing case and is untouched by the report channel, so a full ban gives up the capability without buying anything the one-shot binding does not already buy — and would leave no shipped composition exercising session seeding.
**Use continuable forked children in cache-preserving compositions and accept the loss.** Rejected for the base bundle and ACP/headless examples because the loss is total rather than marginal: reuse breaks ahead of the inherited history, so the child pays full prefill on a transcript it duplicated for the sole purpose of not paying it. The CLI presets make the other tradeoff and retain continuable fork. A deployment that wants a long-lived child with no inherited context already has `spawn`.
**Ship continuable forked children and accept the loss.** Rejected because the loss is total rather than marginal: reuse breaks ahead of the inherited history, so the child pays full prefill on a transcript it duplicated for the sole purpose of not paying it. A deployment that wants a long-lived child with no inherited context already has `spawn`.
**Make `report` visible to every Agent.** A global registration would restore byte-identical prefixes by giving parent and child the same schema and section. Rejected because roots, one-shot children, remote children, and agentless callers would advertise a tool with no derivable recipient, and execution-time rejection would make schema visibility disagree with authority — the scope-local decision the [report tool Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.md) already settled.
@ -38,12 +38,12 @@ The cache-preserving condition is recorded as a `TODO(fork-continuable-prefix-re
## Consequences
- The base bundle and ACP/headless examples create only one-shot forked children; their `subagent_fork` returns a result to the caller's turn, and `send_message` addresses only spawned children there. The three CLI presets create continuable forked children.
- A one-shot forked child's request prefix stays byte-identical to its parent's unless the deployment configures `persona`, `toolFilter`, or a different LLM route on the fork delegation tool, so the token cost of seeding can buy provider-side reuse. Continuable fork adds `report` before the inherited history and forfeits that reuse.
- The fork provider's continuable path has CLI production callers and package-level tests. The same seam accepts one-shot composition, so a bundle or `--patch` overlay can choose either lifecycle without a code change or warning.
- No shipped composition creates a continuable forked child; `subagent_fork` returns a result to its caller's turn, and `send_message` addresses only spawned children.
- A forked child's request prefix stays byte-identical to its parent's unless the deployment configures `persona` or `toolFilter` on the fork delegation tool, so the token cost of seeding buys provider-side reuse again.
- The fork provider's continuable path has no production caller and no assembled-composition coverage. It keeps its package-level tests, and the seam still accepts it, so a bundle or `--patch` overlay can reintroduce it with no code change and no warning.
- `subagent_fork`'s model-visible schema changes: the continuable background wording is replaced by the one-shot task wording in the base bundle, and disappears entirely from the two examples. The affected keyless snapshot tool-schema sidecars are re-recorded in the same change.
- The report obligation reaches spawned children in every continuable composition and forked children in the CLI presets. Its default `next-step` scheduling, authority model, and coverage remain independent of fork composition.
- The report obligation's reach narrows to spawned children in shipped deployments. Its default `next-step` scheduling, authority model, and coverage remain independent of fork composition.
### Accepted risks
The one-shot constraint lives in three configuration files and a code comment, not in a gate; the CLI preset rows already choose `backgroundMode: continuable` and incur the prefix loss. Any bundle or profile patch can make either choice without a warning. That is the accepted cost of not encoding one roster's consequence into `tool-subagent`.
The constraint lives in three configuration files and a code comment, not in a gate. A future bundle row or profile patch can set `backgroundMode: continuable` on a fork tool and silently reintroduce the prefix loss; nothing fails loud. That is the accepted cost of not encoding one roster's consequence into `tool-subagent`.

View file

@ -1,4 +1,4 @@
# Agent Note: 保留缓存的 fork child 保持 one-shot
# Agent Note: fork 出的 child 保持 one-shot
Status: implemented
@ -12,7 +12,7 @@ fork 与 spawn 的唯一区别是 child 的 Session 会以 parent 已完成轮
## 决策
保留缓存的组合会把 fork 委派工具绑定为 `backgroundMode: one-shot`:[base 组合包](../../../../packages/bundle/base/cordis.patch.yml)、[ACP 示例](../../../../examples/acp-agent/cordis.yml)与[headless 示例](../../../../examples/headless-agent/cordis.yml)。base 组合包保留 `run_in_background`,因为它挂载了 task 服务;两个示例设置 `enableRunInBackground: false`,因为它们都不挂载 task 服务,否则一次 one-shot 后台启动会在调用时因缺少 `tasks` 服务而失败。standard、code 与 Cordis CLI preset 则把 fork 绑定为 `continuable`;其子级作用域的 `report` 增量会使继承前缀失效,并接受这里说明的重算成本。
所有交付组合都从 [base bundle](../../../../packages/bundle/base/cordis.patch.yml)继承 fork 委派工具的 `backgroundMode: one-shot`。base bundle 保留 `run_in_background`,因为它也挂载了结算后台工作所需的 task 服务。
one-shot child——前台与后台皆然——经由 `SubagentRuntime.start()` 创建,该路径从不进入可继续的 activation setup 注册表,因此 `report` 与它的提示词 section 都不会被安装。于是一个 fork 出的 one-shot child 的系统提示词与工具 schema 与其 parent 相同,只差部署逐个委派工具主动选择的 `persona` 与 `toolFilter` 增量。
@ -20,9 +20,9 @@ one-shot child——前台与后台皆然——经由 `SubagentRuntime.start()`
### 该限制在于组合,不在于代码
`ForkInProcessProvider.prepareContinuable` 仍然实现完好,`ctx.subagents.startContinuable()` 也接受 `fork`;组合会选择 fork 工具采用 one-shot 还是 continuable。`tool-subagent` 在挂载时同时知道提供方的 `inheritsParentContext` 与自身的 `backgroundMode`,因此一个加载期拒绝该组合的检查是可行的,而这里刻意不加:该组合并非普遍错误。只有在某个 child 作用域增量位于继承历史之前时,它才会产生高昂成本,而产生该增量的包——[`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.zh.md)——是独立安装的,并且按其自身设计对 `tool-subagent` 不可见。一个不安装 report 包的部署可以在前缀完好的前提下运行可继续的 fork child。把某一份插件清单的后果写成委派工具的不变量,会让该工具断言它无法观察到的事实。
`ForkInProcessProvider.prepareContinuable` 仍然实现完好,`ctx.subagents.startContinuable()` 也仍接受 `fork`;改动的只有随附的 `cordis.yml` 行。`tool-subagent` 在挂载时同时知道提供方的 `inheritsParentContext` 与自身的 `backgroundMode`,因此一个加载期拒绝该组合的检查是可行的,而这里刻意不加:该组合并非普遍错误。它只在某个 child 作用域增量位于继承历史之前时才是错的,而产生该增量的包——[`dsh-tool-subagent-report`](../../../../packages/subagent/tool-subagent-report/README.zh.md)——是独立安装的,并且按其自身设计对 `tool-subagent` 不可见。一个不安装 report 包的部署可以在前缀完好的前提下运行可继续的 fork child。把某一份插件清单的后果写成委派工具的不变量,会让该工具断言它无法观察到的事实。
保留缓存的条件记录为 `prepareContinuable` 方法上的 `TODO(fork-continuable-prefix-reuse)` 标记,并由 issue #2124 跟踪:当 child 的系统提示词与工具 schema 能与其 parent 逐字节一致时,可继续 fork 就能保留继承前缀。
重新开放的条件记录为 `prepareContinuable` 方法上的 `TODO(fork-continuable-prefix-reuse)` 标记——随附组合不调用这个方法——并由 issue #2124 跟踪:当 child 的系统提示词与工具 schema 能与其 parent 逐字节一致时,可继续 fork 即可重新开放。
## 备选方案
@ -30,7 +30,7 @@ one-shot child——前台与后台皆然——经由 `SubagentRuntime.start()`
**干脆不挂载 fork 提供方。** 这是该限制更彻底的形式。否决的原因是前台 fork *正是*复用前缀的那种情形,且不受 report 通道影响,因此全面禁用会在不换来任何 one-shot 绑定尚未换来的东西的同时放弃该能力——并且随附组合将没有任何一个演练 session 初始内容。
**在保留缓存的组合中随附可继续的 fork child 并接受这份损失。** base 组合包与 ACP/headless 示例不采用,因为这份损失是全额而非边际的:复用在继承历史之前就已中断,于是 child 为一份自己复制过来、目的恰恰是不必付费的 transcript 付了全额预填充。CLI preset 选择了另一项取舍并保留可继续 fork。想要一个没有继承上下文的长期 child 的部署,本来就有 `spawn`。
**照常随附可继续的 fork child 并接受这份损失。** 否决的原因是这份损失是全额而非边际的:复用在继承历史之前就已中断,于是 child 为一份自己复制过来、目的恰恰是不必付费的 transcript 付了全额预填充。想要一个没有继承上下文的长期 child 的部署,本来就有 `spawn`。
**让 `report` 对每个 Agent 可见。** 全局注册会通过让 parent 与 child 拥有相同的 schema 与 section 来恢复逐字节相同的前缀。否决的原因是根 agent、one-shot child、远端 child 与无 agent 调用方都会宣告一件推导不出收件方的工具,而执行期拒绝会让 schema 可见性与权限彼此矛盾——这正是[report 工具 Agent Note](../feature/2026-07-30-continuable-subagent-report-tool.zh.md)已经定下的作用域局部决策。
@ -38,12 +38,12 @@ one-shot child——前台与后台皆然——经由 `SubagentRuntime.start()`
## 后果
- base 组合包与 ACP/headless 示例只创建 one-shot fork child;其中的 `subagent_fork` 会把结果返回给调用方的轮次,`send_message` 也只寻址 spawn 出的 child。三个 CLI preset 会创建可继续的 fork child。
- 除非部署在 fork 委派工具上配置了 `persona`、`toolFilter` 或不同的 LLM 路由,one-shot fork child 的请求前缀会与其 parent 逐字节相同,因此初始内容的 token 成本可以换来提供方侧的复用。可继续 fork 会在继承历史之前增加 `report`,从而失去该复用。
- fork 提供方的可继续路径有 CLI 生产调用方与包内测试。同一条 seam 也接受 one-shot 组合,因此某个组合包或 `--patch` 覆盖层可以无需改动代码、也不会有任何警告地选择任一生命周期。
- 没有任何随附组合会创建可继续的 fork child;`subagent_fork` 把结果返回给调用方的轮次,而 `send_message` 只寻址 spawn 出的 child。
- 除非部署在 fork 委派工具上配置了 `persona` 或 `toolFilter`,fork child 的请求前缀与其 parent 逐字节相同,因此初始内容的 token 成本重新换来了提供方侧的复用。
- fork 提供方的可继续路径没有生产调用方,也没有整体组装层面的覆盖。它保留自己的包内测试,seam 也仍然接受它,因此某个组合包或 `--patch` 覆盖层可以无需改动代码、也不会有任何警告地把它重新引入。
- `subagent_fork` 面向模型的 schema 发生变化:base 组合包中可继续的后台措辞被 one-shot 的 task 措辞取代,在两个示例中则完全消失。受影响的无密钥快照工具 schema 伴随文件在同一次改动中重新记录。
- 在每个可继续组合中,report 义务都会覆盖 spawn 出的 child;在 CLI preset 中,它也覆盖 fork 出的 child。它的 `next-step` 默认调度、权限模型与覆盖仍独立于 fork 组合。
- 在随附部署中,report 义务的覆盖范围收窄到 spawn 出的 child。它的 `next-step` 默认调度、权限模型与覆盖仍独立于 fork 组合。
### 已接受的风险
one-shot 限制存在于三个配置文件与一处代码注释中,而不在门禁里;CLI preset 行已经选择 `backgroundMode: continuable` 并承担前缀损失。任何组合包或 profile 补丁都能选择任一方式,且不会收到警告。这就是不把某一份插件清单的后果写入 `tool-subagent` 所接受的代价。
该限制存在于三个配置文件与一处代码注释中,而不在门禁里。未来某个组合包行或 profile 补丁可以在 fork 工具上设置 `backgroundMode: continuable`,从而悄然重新引入前缀损失;没有任何东西会失败得很响亮。这就是不把某一份插件清单的后果写入 `tool-subagent` 所接受的代价。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-11-repository-naming-contract-and-rename-ledger.md
2026-08-11-repository-naming-contract-and-rename-ledger.md: 403a7288bb3af0eff09b229e7d399b94492ae6c0
2026-08-11-repository-naming-contract-and-rename-ledger.zh.md: 644e1c0bebfc43414e9e67def8d6c42acddf4c72
2026-08-11-repository-naming-contract-and-rename-ledger.md: 269f18c494a17a1af5ab4252ad812a5ded03d672
2026-08-11-repository-naming-contract-and-rename-ledger.zh.md: 614332795d46d810bdef32c0d0f7e1010fc7f8a0

View file

@ -276,7 +276,7 @@ Keep MCP, Todo, and the Plan Mode package, key, events, and tool names. This dec
| `PluginInventoryService` | `PluginInventoryGateway` | The class is a Remote-only adapter from the live Loader tree to the `pluginInventory/list` RPC. It owns no same-process service, cache, history, or mutation path. `Gateway` states the role that exists. |
| `@deepseek-ai/dsh-jsonrpc-demo`, `@deepseek-ai/dsh-sdk-jsonrpc-demo`, `@deepseek-ai/dsh-sdk-python-runtime` | removed | The Python runtime packages the existing `@deepseek-ai/dsh` CLI and its `sdk` profile; a private application package would recreate a second launcher. |
| `packages/examples/jsonrpc-demo/`, `packages/sdk/python-runtime/` | removed | The Python runtime wheel's closure manifest owns packaging without a separate application package. |
| `examples/jsonrpc-agent/` | `examples/python-sdk-agent/` | The example demonstrates Python use of the `sdk` profile and ordered patches. |
| `examples/jsonrpc-agent/` | `python/sdk/examples/` | The example demonstrates Python use of the `sdk` profile and ordered patches. |
| `@deepseek-ai/dsh-acp-demo` | `@deepseek-ai/dsh-acp-app` | The package is the ACP profile's application bundle, not a standalone demo bin. |
| Deploy-root manifests `dsh-jsonrpc-agent-pkg`, `dsh-sdk-python-runtime-closure` | `dsh-python-runtime-closure` | The zero-code manifest defines the Python runtime wheel's complete `dsh` dependency closure without naming a separate SDK application. |
| `@deepseek-ai/dsh-frontend` | `@deepseek-ai/dsh-web-frontend` | The application is the web frontend. Keep its physical `apps/web/` folder. |

View file

@ -276,7 +276,7 @@ PascalCase 标识符中的首字母缩略词使用首字母大写格式:`Ui`
| `PluginInventoryService` | `PluginInventoryGateway` | 该类只负责把实时 Loader 树适配到 `pluginInventory/list` RPC。它不拥有同进程服务、缓存、历史或修改路径。`Gateway` 准确说明现有角色。 |
| `@deepseek-ai/dsh-jsonrpc-demo`、`@deepseek-ai/dsh-sdk-jsonrpc-demo`、`@deepseek-ai/dsh-sdk-python-runtime` | 已删除 | Python 运行时打包现有 `@deepseek-ai/dsh` CLI 与其 `sdk` profile;私有应用包会重新产生第二个启动器。 |
| `packages/examples/jsonrpc-demo/`、`packages/sdk/python-runtime/` | 已删除 | Python 运行时 wheel 的闭包 manifest 负责打包,无需独立应用包。 |
| `examples/jsonrpc-agent/` | `examples/python-sdk-agent/` | 该示例演示 Python 使用 `sdk` profile 与有序 patch。 |
| `examples/jsonrpc-agent/` | `python/sdk/examples/` | 该示例演示 Python 使用 `sdk` profile 与有序 patch。 |
| `@deepseek-ai/dsh-acp-demo` | `@deepseek-ai/dsh-acp-app` | 该包是 ACP profile 的应用组合包,不是独立 demo bin。 |
| 部署根 manifest `dsh-jsonrpc-agent-pkg`、`dsh-sdk-python-runtime-closure` | `dsh-python-runtime-closure` | 该零代码 manifest 定义 Python 运行时 wheel 的完整 `dsh` 依赖闭包,不再命名独立 SDK 应用。 |
| `@deepseek-ai/dsh-frontend` | `@deepseek-ai/dsh-web-frontend` | 该应用是 Web 前端。保留其物理目录 `apps/web/`。 |

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-12-pi-ai-route-default-input-modalities.md
2026-08-12-pi-ai-route-default-input-modalities.md: eb03d5330a1283d439e16262325965cf2e7e8087
2026-08-12-pi-ai-route-default-input-modalities.zh.md: dfbbd2ae6db7a78e82955db66fe506d3506209fd
2026-08-12-pi-ai-route-default-input-modalities.md: 83158d686b028a085fc8801f83a7cea1a369471b
2026-08-12-pi-ai-route-default-input-modalities.zh.md: 18ee56db7bfe5cba63c125341b5ecc5fe3905c5b

View file

@ -50,4 +50,4 @@ A model that declares image input its endpoint does not serve is not caught loca
`config.spec.ts` holds the schema boundary: an unknown modality refused at both levels, the empty route list accepted by the schema and refused by the namespace validator that the settings seam actually runs, and the `[]` materialization for an absent array that the inheritance rule depends on.
No keyless snapshot lane exercises a pi-ai route: the snapshot examples drive `dsh-llm-replay`, which declares modalities directly in its configuration, and a pi-ai route needs a live endpoint whose port a static `cordis.yml` cannot name. The admission points this change feeds are already covered there through that provider (`examples/acp-agent/image.cordis.snapshot.yml` and `image-text-route.cordis.snapshot.yml`) and are unaffected — what changed is what one adapter reports, not how a gate reads it.
No keyless snapshot lane exercises a pi-ai route: the recorded-session corpus drives `dsh-llm-replay`, which declares modalities directly in its configuration, and a pi-ai route needs a live endpoint whose port a static `cordis.yml` cannot name. The admission points this change feeds are covered by [`read-image`](../../../../snapshots/session/read-image/) and [`read-image-text-route`](../../../../snapshots/session/read-image-text-route/) and are unaffected — what changed is what one adapter reports, not how a gate reads it.

View file

@ -50,4 +50,4 @@ DeepSeek 直接适配器拥有独立的精确模型目录。支持视觉的条
`config.spec.ts` 负责 schema 边界:两个层级上的未知模态拒绝、路由空列表被 schema 接受而由 settings seam 真正运行的命名空间校验器拒绝,以及继承规则所倚赖的「缺省数组物化为 `[]`」这一事实。
没有任何无密钥 snapshot 通道会跑 pi-ai 路由:snapshot 示例驱动的是 `dsh-llm-replay`,它在自己的配置里直接声明模态,而 pi-ai 路由需要一个真实端点,其端口是静态 `cordis.yml` 无法写出的。本次变更所供给的那些准入点已经通过该提供方在那里得到覆盖(`examples/acp-agent/image.cordis.snapshot.yml` 与 `image-text-route.cordis.snapshot.yml`)且不受影响——改变的是某个适配器报告什么,而非门禁如何读取它。
没有任何无密钥 snapshot 通道会跑 pi-ai 路由:录制会话语料驱动的是 `dsh-llm-replay`,它在自己的配置里直接声明模态,而 pi-ai 路由需要一个真实端点,其端口是静态 `cordis.yml` 无法写出的。本次变更所供给的准入点由 [`read-image`](../../../../snapshots/session/read-image/) 与 [`read-image-text-route`](../../../../snapshots/session/read-image-text-route/)覆盖且不受影响——改变的是某个适配器报告什么,而非门禁如何读取它。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-12-plugin-owned-settings-surface.md
2026-08-12-plugin-owned-settings-surface.md: daed91b8ac4ce0acb81e19908ec08c9f3f7ac21e
2026-08-12-plugin-owned-settings-surface.zh.md: ba39b84faaf90be413ad8d7b8327c67b2fa27cea
2026-08-12-plugin-owned-settings-surface.md: f1c9b48abd3459ed58e3e10fbdd95558884c9a22
2026-08-12-plugin-owned-settings-surface.zh.md: ff9f37290a8dd8f1ab9ef1bfef3790a3ef20086b

View file

@ -32,7 +32,7 @@ Keying makes absence the signal, and that is what removes the bookkeeping the pr
The gate did keep one thing off the wire, and this note states it plainly because the decision has to survive the accurate version: a registered namespace the list did not name never had its resolved, `base`, or `user` values reach the browser at all. The plugin inventory page is not a substitute — `PluginInventoryEntry` carries `entryId`, `moduleName`, `enabled`, and `fiberPhase`, and its "configuration" row renders an enabled/disabled tag, never a stored value.
What the gate was not is the boundary its position suggested. Every `settings.*` method sits in `PRIVILEGED_METHODS` (`packages/client/connection`), so a non-loopback or cross-origin request is refused with 403 before reaching this code; `role('secret')` fields are structurally stripped from every layer of every response; and the document the plane edits is the user's own `settings.yaml`, which the same settings page offers to open. The writes it did not block were also the consequential ones: `permission` (which can widen the approval preset) and `agent-presets` (which decides what a session mounts) were both already served.
What the gate was not is the security boundary its position suggested. Connection authenticates every Host API request before reaching this code; `role('secret')` fields are structurally stripped from every layer of every response; and the document the plane edits is the user's own `settings.yaml`, which the same settings page offers to open. The writes it did not block were also the consequential ones: `permission` (which can widen the approval preset) and `agent-presets` (which decides what a session mounts) were both already served.
So the exposure this change actually adds, in this repository, is one namespace: `agent-default-model`, whose two fields name a provider and a model and which no browser half renders. A future namespace whose values genuinely must not cross the wire is answered per field by `role('secret')` — finer than a namespace switch, and already enforced.

View file

@ -32,7 +32,7 @@ Status: implemented
这道门确实挡住了一样东西,本 note 如实写出,因为这个决策必须在准确版本下也站得住:不在名单上的已注册命名空间,其 resolved、`base` 与 `user` 值根本不会抵达浏览器。插件清单页不能替代它——`PluginInventoryEntry` 携带的是 `entryId`、`moduleName`、`enabled` 与 `fiberPhase`,它那一行「configuration」渲染的是启用/停用标签,从不是任何已存值。
这道门不是的,是它所处位置暗示的那种边界。每个 `settings.*` 方法都在 `PRIVILEGED_METHODS` 里(`packages/client/connection`),非回环或跨源请求在到达这段代码之前就以 403 被拒;`role('secret')` 字段在每种响应的每一层都被结构性剥离;而这个面所编辑的文档,本就是用户自己的 `settings.yaml`,同一个设置页还提供了打开它的入口。它没有挡住的写入,恰恰是有分量的那些:`permission`(能放宽审批预设)与 `agent-presets`(决定一个会话挂载什么)本来就已被服务。
这道门不是的,是它所处位置暗示的那种安全边界。Connection 在到达这段代码前认证每个 Host API 请求;`role('secret')` 字段在每种响应的每一层都被结构性剥离;而这个面所编辑的文档,本就是用户自己的 `settings.yaml`,同一个设置页还提供了打开它的入口。它没有挡住的写入,恰恰是有分量的那些:`permission`(能放宽审批预设)与 `agent-presets`(决定一个会话挂载什么)本来就已被服务。
因此本次改动在本仓库实际新增的暴露面是一个命名空间:`agent-default-model`——它的两个字段指明一个提供方与一个模型,且没有任何浏览器半侧渲染它。将来若某个命名空间的值确实不该跨越协议,由 `role('secret')` 逐字段作答:比整命名空间开关更精细,而且已经在执行。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-15-client-shells-and-dynamic-packages.md
2026-08-15-client-shells-and-dynamic-packages.md: 8713234735ddb53db27e17a2ea9087de665d86f7
2026-08-15-client-shells-and-dynamic-packages.zh.md: ec8a2fb358e86f4fe74db13d555b8be8432ad62e
2026-08-15-client-shells-and-dynamic-packages.md: 016314d10f55e0b590e98944ca417bae658ab56a
2026-08-15-client-shells-and-dynamic-packages.zh.md: 4e0277d1becab8467521dc21d0e5b7509d1ee993

View file

@ -46,12 +46,12 @@ There is no general `dsh.client.provide` alias mechanism. Dynamic rows and stati
The modules Node half injects the startup protocol into the served HTML in this order:
1. Install `window.__ModuleLoader__` in queue mode with `pendingQueue`, `load()`, and `create()`.
2. Start preloading the content-addressed application batch containing every row except modules.
3. Execute one blocking bootstrap batch containing the ordinary modules factory registration.
4. Assign `window.__DSH_BOOT__`, including both batch descriptors and every row's individual HMR URL.
2. Start preloading every content-addressed application combo URL containing the rows other than modules.
3. Execute every blocking bootstrap combo URL; these currently contain the ordinary modules factory registration.
4. Assign `window.__DSH_BOOT__`, including all scheduling descriptors and every row's one-resource HMR combo URL.
5. Execute the Vite main module.
The bootstrap batch only registers the modules factory. The startup kernel passes the raw graph and shell seeds to `__ModuleLoader__.create()`. The facade removes the modules registration, materializes it with a `require` function that rejects every external, and invokes its `createClientModuleSystem` export. The modules bundle parses the graph, constructs `ClientModuleSystem`, caches its own exports as the modules row, retains the system in a module closure, and switches the same facade to live mode. The modules client face consequently has a zero-external bootstrap requirement.
The bootstrap combo currently registers only the modules factory. The startup kernel passes the raw graph and shell seeds to `__ModuleLoader__.create()`. The facade removes the modules registration, materializes it with a `require` function that rejects every external, and invokes its `createClientModuleSystem` export. The modules bundle parses the graph, constructs `ClientModuleSystem`, caches its own exports as the modules row, retains the system in a module closure, and switches the same facade to live mode. The modules client face consequently has a zero-external bootstrap requirement.
After the `immediately` tier has registered its factories, the kernel creates all Loader entries, awaits Cordis quiescence, and requires every fiber to be ACTIVE. It then calls `ctx.uiRenderer.mount(container)`. The dynamic `ui-renderer` package owns React, slot rendering, hydration of the existing boot DOM, and the React root lifecycle; the startup kernel and failure page remain React-free.
@ -79,7 +79,7 @@ Ordinary installed libraries remain `dependencies`: a dynamic build may bundle a
Bundle contents stay stable when an npm dependency moves between peer and development sections, because each build face declares externality directly. Static libraries remain host-assembled, while dynamic packages retain uniform artifacts and lifecycle governance.
The startup protocol depends on the modules package id, and modules must remain self-contained at runtime. Batch generation preserves its ordinary package artifact and gives every other row one shared initial transport; HMR still uses each row's revisioned individual artifact. A missing bootstrap registration fails before Cordis starts; later plugin import, apply, and service-wait failures remain visible through the boot page's ACTIVE scan.
The startup protocol depends on the modules package id, and modules must remain self-contained at runtime. Combo generation preserves its ordinary package artifact and gives every other row one shared initial transport; HMR uses the same route with that row as its sole resource. A missing bootstrap registration fails before Cordis starts; later plugin import, apply, and service-wait failures remain visible through the boot page's ACTIVE scan.
The shell consumes built `lib/` products, so source and browser artifacts can drift until the relevant build or watcher runs. Typechecking source alone does not prove the served application uses the same code.

View file

@ -46,12 +46,12 @@ Client npm 依赖区段描述安装和开发关系,但不能可靠描述 bundl
Modules Node 半按以下顺序向实际返回的 HTML 注入启动协议:
1. 以 queue 模式安装 `window.__ModuleLoader__`,包含 `pendingQueue`、`load()` 与 `create()`。
2. 开始预加载按内容寻址的 application 批次,其中包含 modules 之外的全部 row。
3. 执行一个阻塞式 bootstrap 批次,其中包含普通的 modules factory registration。
4. 赋值 `window.__DSH_BOOT__`,其中包含两个批次描述及每个 row 的独立 HMR URL。
2. 开始预加载所有按内容寻址的 application combo URL,其中包含 modules 之外的 row。
3. 执行所有阻塞式 bootstrap combo URL;当前其中包含普通的 modules factory registration。
4. 赋值 `window.__DSH_BOOT__`,其中包含全部调度描述及每个 row 的单资源 HMR combo URL。
5. 执行 Vite 主模块。
Bootstrap 批次只登记 modules factory。启动内核把原始图与外壳 seed 传给 `__ModuleLoader__.create()`。Facade 移除 modules registration,用拒绝全部 external 的 `require` 函数将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造 `ClientModuleSystem`、把自身 exports 缓存为 modules row、在模块闭包中保留该系统,并把同一 facade 切换到 live 模式。因此 modules client face 必须满足零 external 的自举要求。
Bootstrap combo 当前只登记 modules factory。启动内核把原始图与外壳 seed 传给 `__ModuleLoader__.create()`。Facade 移除 modules registration,用拒绝全部 external 的 `require` 函数将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造 `ClientModuleSystem`、把自身 exports 缓存为 modules row、在模块闭包中保留该系统,并把同一 facade 切换到 live 模式。因此 modules client face 必须满足零 external 的自举要求。
`immediately` 层级完成 factory 注册后,内核创建全部 Loader entry,等待 Cordis 静止,并要求每个 fiber 都进入 ACTIVE。随后调用 `ctx.uiRenderer.mount(container)`。动态 `ui-renderer` 包拥有 React、slot 渲染、已有启动 DOM 的 hydrate 和 React root 生命周期;启动内核与失败页保持 React-free。
@ -79,7 +79,7 @@ Bootstrap 批次只登记 modules factory。启动内核把原始图与外壳 se
Npm 依赖在 peer 与开发区段间移动时,bundle 内容保持稳定,因为每个构建 face 都直接声明 external。静态库继续由宿主装配,动态包则保留统一产物与生命周期治理。
启动协议依赖 modules 的 package id,modules 还必须保持运行期自包含。批次生成保留其普通 package 产物,并为其他全部 row 提供一条共享初始传输;HMR 仍使用每个 row 带 revision 的独立产物。缺少 bootstrap registration 会在 Cordis 启动前失败;后续插件 import、apply 与 service 等待失败仍由启动页的 ACTIVE 扫描呈现。
启动协议依赖 modules 的 package id,modules 还必须保持运行期自包含。Combo 生成保留其普通 package 产物,并为其他全部 row 提供一条共享初始传输;HMR 使用同一条路由,并只把该 row 作为资源。缺少 bootstrap registration 会在 Cordis 启动前失败;后续插件 import、apply 与 service 等待失败仍由启动页的 ACTIVE 扫描呈现。
外壳消费已构建 `lib/` 产品,因此在相关 build 或 watcher 运行前,源码与浏览器产物可能漂移。仅源码 typecheck 通过不能证明实际服务的应用使用同一份代码。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-18-session-history-and-event-transport.md
2026-08-18-session-history-and-event-transport.md: 206d3d13d1f183b97b02b89644b022367be2ebfd
2026-08-18-session-history-and-event-transport.zh.md: 3d0b5d4ab768ecd3877bfde86822245d0b63ac49
2026-08-18-session-history-and-event-transport.md: 808565ff7df60b8aa6aa3f18820c1139b8bf5362
2026-08-18-session-history-and-event-transport.zh.md: 8bd00def4531afa9cdf77ae7f689f2e77908545e

View file

@ -134,7 +134,7 @@ If a page request is canceled with its physical carrier generation, the journal
`packages/api/session-controller` provides Host `ctx.sessionController` and the generated `ctx.remote.session` namespace.
It owns Session list, search, create, models, selectModel, rename, fork, prompt, attachment, updateQueue, cancel, page, follow, and control.
It owns Session list, search, create, selectModel, rename, fork, prompt, attachment, updateQueue, cancel, page, follow, and control. The Host-generation model catalog is exposed separately through `llm.models` because it is not Session-specific.
The package separates agent, commands, control, history, and list controllers internally, but Session identity resolution, activation policy, subagent ownership, and Remote error projection have one public owner.
@ -148,9 +148,9 @@ Each method explicitly selects a cold inspection, live-only lookup, or resume-ca
| Operation | Source or result without a live Agent | Activation rule |
|---|---|---|
| `session.list`, `search` | persistence, projection cache, or cold log | Never resumes an Agent |
| `session.list`, `search` | headers and projection cache; a bounded small-log read can resolve uncertain blankness | Never resumes an Agent |
| `session.page(address)` | attached Session or persistence log | Never resumes an Agent |
| `session.follow(address)` | cold-read current cursor, then wait for future appends | Neither opening nor waiting resumes an Agent |
| `session.follow(address)` | one live or prepared observation carrying the opening page and projections | Publishes the snapshot first, then promotes an ordinary cold Session once in the background |
| `session.control()` | current attached Agents, pending registry, and process-local registries | Baseline and reconnect do not resume an Agent |
| `session.attachment`, fork source read | authorized durable Session data | A read does not resume an Agent |
| `session.updateQueue`, `cancel` | only the current live Agent | Does not resume vanished state |
@ -159,6 +159,12 @@ Each method explicitly selects a cold inspection, live-only lookup, or resume-ca
Reading titles, lists, and projections does not require an Agent. An observation operation cannot inherit resume authority merely because another Remote endpoint uses Agent lookup.
`SessionQuery.observeSession()` chooses an attached Session or borrows one prepared source from `SessionPersistence.borrowSession()`. The persistence preparation cache shares concurrent cold reads and pins the exact unpublished Session until every observation lease is released. An observation computes either all registered projections or none; callers may expose a subset, but no caller creates a partial projection state.
`session.list` never performs an unbounded cold-log scan. It uses cached projection hints when available and may fully observe only an individually stored artifact within the configured small-log byte limit to distinguish an abandoned blank Session. Missing or unreadable hints keep the row visible with unknown metadata.
`model/selection` is a required-on-read durable event because it changes the model route used by the next request. Its projection records both the last request selection and a later pending selection; prompt assembly consumes the pending value when the matching `request/header` is committed.
#### Session journal
`session.page` returns a history window clipped on message boundaries with contiguous internal sequence numbers. Every request must carry an explicit `throughSeq`; this value comes from the corresponding `session.follow` generation's opening cursor and fixes the read at the same log cut. A tail page without `beforeSeq` must end exactly at `throughSeq`, where `-1` denotes an empty log. `beforeSeq` only selects an older page before that cut and cannot replace the synchronization cursor. `maxMessages` limits user/assistant message count without dropping chunks, tools, or state events between those messages.
@ -167,20 +173,20 @@ The tail page also carries a projection baseline no later than `throughSeq`; old
Ordinary Sessions and direct subagents use one `SessionAddress` protocol. A direct-subagent address carries parent Session, child Session, and mode; a cold Host read verifies durable ownership and descriptor rather than authorizing access from the child id alone.
`session.follow` installs `session/event` and `session/created` listeners before checking an attached Session or persistence, then reads the current cursor.
`session.follow` installs `session/event` and `session/created` listeners before observing an attached or prepared Session.
The first follow response is `{ type: 'opened', cursor }`. A generation with `afterSeq` first replays the missing suffix from the authoritative log, then emits commits buffered during the read in sequence order.
The first follow response is a complete `{ type: 'snapshot', header, cursor, events, hasMore, projections }` frame. Every reconnect sends another complete snapshot replacement; the protocol has no `afterSeq`. Events committed during observation remain buffered and are emitted after the snapshot in sequence order.
A cold Session can open history immediately and keep follow waiting. Future events appear only after another explicit command resumes the Agent.
A cold ordinary Session can publish its prepared snapshot immediately. After that first frame, the Controller transfers a retained observation to one background promotion; follow does not wait for activation. Direct-subagent addresses never use this promotion path.
Client `SessionEventStream` extends `RemoteJournalStream` and supplies only `session.follow`, `session.page`, the Session sequence algorithm, and repair requests. The general layer first obtains opening cursor `C`, then calls `session.page({ throughSeq: C })`; entries `C + 1...` received during the read remain in the follow queue, and the page must cover exactly through `C` before the layer merges and publishes a continuous sequence.
Client `SessionEventStream` extends `RemoteJournalStream` and supplies only `session.follow`, `session.page`, the Session sequence algorithm, and repair requests. The general layer validates and publishes the opening snapshot directly. It calls `session.page({ throughSeq })` only for older history or when a later event reveals a sequence gap.
```text
ctx.remote.session.follow(address, afterSeq?) --------|
|[]> SessionEventStream
ctx.remote.session.page(address, throughSeq, pageArgs) -| |-- replace(window)
|-- prepend(history)
`-- append(live entry)
ctx.remote.session.follow(address, pageArgs) ----------------|
snapshot(header, cursor, page, projections), event* |[]> SessionEventStream
ctx.remote.session.page(address, throughSeq, pageArgs) -------| |-- replace(window)
|-- prepend(history)
`-- append(live entry)
```
Each Client Session owns only one current `events: SessionEventStream | undefined`. The read-only `SessionEventSource` gives the materialized event window to Conversation consumers.
@ -328,7 +334,7 @@ Connection tests pin missing, duplicate, and withdrawn generation sources; the r
`RemoteSnapshotStream` tests pin exactly one opening snapshot per generation, rejection of an update before a snapshot, rejection of duplicate snapshots, and reconnect replacement.
`RemoteJournalStream` tests pin follow-before-page, opening-overlap removal, contiguous append, historical prepend, reconnect catch-up, gap repair, and one atomic replacement.
`RemoteJournalStream` tests pin snapshot-first opening, contiguous append, historical prepend, reconnect replacement, gap repair, and one atomic replacement.
Session Host tests pin cold page/follow without increasing attached Agents, contiguous events reaching a cold follow after an explicit prompt, direct-subagent ownership, message-aligned pagination, and terminal-error projection.
@ -352,7 +358,7 @@ Static checks pin that API Proxy exports no Session/Workspace Host-frame carrier
## Consequences
The browser can read and follow a durable Session while its Agent is stopped. Observation does not implicitly resume execution; only explicitly authorized Session commands create or resume Agents according to their own rules.
The browser can read a durable Session while its Agent is stopped. Opening an ordinary Session publishes the prepared snapshot before one background promotion begins; list, search, page, and other observation-only reads never activate it.
Durable logs repair a missing suffix by sequence number and page; Session control and Workspace state converge through opening snapshots; ordinary Remote Events promise no replay. Recovery semantics follow the data kind instead of imitating one another.

View file

@ -134,7 +134,7 @@ repair 期间旧 window 保持可读;page 与期间积累的 live entries 拼
`packages/api/session-controller` 提供 Host `ctx.sessionController` 与生成的 `ctx.remote.session` namespace。
它拥有 Session list、search、create、models、selectModel、rename、fork、prompt、attachment、updateQueue、cancel、page、follow 与 control。
它拥有 Session list、search、create、selectModel、rename、fork、prompt、attachment、updateQueue、cancel、page、follow 与 control。Host generation 的 model catalog 通过独立的 `llm.models` 公开,因为它不属于特定 Session。
包内的 agent、commands、control、history 与 list controller 分开实现,但 Session 身份解析、激活策略、subagent ownership 和 Remote 错误投影只有一个公开 owner。
@ -148,9 +148,9 @@ Session Remote 方法传递 `SessionId` 或 `SessionAddress`,不靠参数类
| 操作 | 无 live Agent 时的数据来源或结果 | 激活规则 |
|---|---|---|
| `session.list`、`search` | persistence、投影缓存或冷日志 | 永不恢复 Agent |
| `session.list`、`search` | header 与投影缓存;可通过有界的小日志读取判断不确定的 blank 状态 | 永不恢复 Agent |
| `session.page(address)` | attached Session 或 persistence 日志 | 永不恢复 Agent |
| `session.follow(address)` | 冷读当前 cursor,等待将来的 append | 建联和等待都不恢复 Agent |
| `session.follow(address)` | 一份携带 opening page 与 projection 的 live 或 prepared observation | 先发布 snapshot,再在后台把普通冷 Session 提升一次 |
| `session.control()` | 当前 attached Agent、pending registry 与进程内 registry | baseline 与重连不恢复 Agent |
| `session.attachment`、fork 源读取 | 已授权的持久 Session 数据 | 读取不恢复 Agent |
| `session.updateQueue`、`cancel` | 仅命中当前 live Agent | 不为已消失状态恢复 Agent |
@ -159,6 +159,12 @@ Session Remote 方法传递 `SessionId` 或 `SessionAddress`,不靠参数类
读取 title、列表和投影不要求 Agent。观察操作不能因为另一个 Remote endpoint 使用了 Agent lookup 而继承其恢复权限。
`SessionQuery.observeSession()` 选择 attached Session,或从 `SessionPersistence.borrowSession()` 借用 prepared source。Persistence preparation cache 共享并发冷读取,并在所有 observation lease 释放前固定同一个未发布 Session。一次 observation 要么计算所有已注册 projection,要么完全不计算;调用方可以只公开其中一部分,但不会建立只计算部分 projection 的中间状态。
`session.list` 不会无界扫描冷日志。它优先使用缓存的 projection hint,仅在独立存储 artifact 不超过配置的小日志字节上限时,才可能完整观察日志以判断不确定的 blank 状态。hint 缺失或不可读时,列表仍保留该行,并把 metadata 视为未知。
`model/selection` 是 required-on-read 的持久 event,因为它改变下一次请求使用的 model route。对应 projection 同时记录最近一次 request selection 与之后的 pending selection;prompt assembly 在提交匹配的 `request/header` 时消费 pending value。
#### Session 日志
`session.page` 返回一段按消息边界裁剪、内部 seq 连续的历史窗口。每个请求必须显式携带 `throughSeq`;该值来自对应 `session.follow` generation 的 opening cursor,并把本次读取固定在同一个日志切点。无 `beforeSeq` 的 tail page 必须精确结束于 `throughSeq`,其中 `-1` 表示空日志;`beforeSeq` 只选择该切点之前的更早页面,不能替代同步 cursor。`maxMessages` 限制 user/assistant 消息数,不丢弃这些消息之间的 chunk、tool 或状态事件。
@ -167,20 +173,20 @@ tail page 同时携带不晚于 `throughSeq` 的 projection baseline;旧页只
普通 Session 与 direct subagent 使用同一个 `SessionAddress` 协议。direct subagent 地址同时携带父 Session、子 Session 与 mode,Host 冷读时验证持久 ownership 和 descriptor,不能只凭 child id 越权读取。
`session.follow` 在检查 attached Session 或 persistence 前先安装 `session/event` 与 `session/created` listener,再读取当前 cursor。
`session.follow` 在观察 attached 或 prepared Session 前先安装 `session/event` 与 `session/created` listener。
首次 follow 返回 `{ type: 'opened', cursor }`。带 `afterSeq` 的 generation 先从权威日志重放缺失后缀,再按 seq 排出读取期间缓存的 commit。
首次 follow 返回完整的 `{ type: 'snapshot', header, cursor, events, hasMore, projections }` frame。每次重连都发送另一份完整 snapshot replacement;协议不含 `afterSeq`。观察期间提交的 event 会保留在缓冲区,并在 snapshot 之后按 seq 发出。
冷 Session 可以立即打开历史并保持 follow 等待。只有另一条显式命令恢复 Agent 后,后续事件才会出现。
普通冷 Session 可以立即发布 prepared snapshot。首帧之后,Controller 把 retained observation 交给一次后台 promotion;follow 不等待激活。Direct-subagent 地址不会进入该 promotion 路径。
Client 的 `SessionEventStream` 继承 `RemoteJournalStream`,只提供 `session.follow`、`session.page`、Session seq 算法与 repair request。通用层先取得 opening cursor `C`,再调用 `session.page({ throughSeq: C })`;读取期间收到的 `C + 1...` entries 留在 follow 队列中,page 精确覆盖至 `C` 后才按连续 seq 合并并发布。
Client 的 `SessionEventStream` 继承 `RemoteJournalStream`,只提供 `session.follow`、`session.page`、Session seq 算法与 repair request。通用层直接校验并发布 opening snapshot;仅在读取更早历史或后续 event 暴露 seq gap 时调用 `session.page({ throughSeq })`。
```text
ctx.remote.session.follow(address, afterSeq?) --------|
|[]> SessionEventStream
ctx.remote.session.page(address, throughSeq, pageArgs) -| |-- replace(window)
|-- prepend(history)
`-- append(live entry)
ctx.remote.session.follow(address, pageArgs) ----------------|
snapshot(header, cursor, page, projections), event* |[]> SessionEventStream
ctx.remote.session.page(address, throughSeq, pageArgs) -------| |-- replace(window)
|-- prepend(history)
`-- append(live entry)
```
每个 Client Session 只持有一个当前 `events: SessionEventStream | undefined`。只读 `SessionEventSource` 把已物化 event window 交给 Conversation consumer。
@ -328,7 +334,7 @@ Connection 测试固定 generation source 缺失、重复注册、撤回、`$eve
`RemoteSnapshotStream` 测试固定每 generation 恰好一份 opening snapshot、update-before-snapshot 拒绝、重复 snapshot 拒绝和重连 replacement。
`RemoteJournalStream` 测试固定 follow-before-page、opening overlap 去重、连续 append、历史 prepend、重连 catch-up、gap repair 与一次性 replacement。
`RemoteJournalStream` 测试固定 snapshot-first opening、连续 append、历史 prepend、重连 replacement、gap repair 与一次性 replacement。
Session Host 测试固定 cold page/follow 不增加 attached Agent、显式 prompt 后 cold follow 收到连续事件、direct subagent ownership、message-aligned pagination 和终止错误投影。
@ -352,7 +358,7 @@ Remote Event Client 测试固定实例私有 key、Cordis 注册顺序、Agent C
## 后果
浏览器可以在 Agent 停止时读取并跟随持久 Session。观察不隐式恢复执行,只有明确获得授权的 Session 命令按各自约定创建或恢复 Agent。
浏览器可以在 Agent 停止时读取持久 Session。打开普通 Session 时先发布 prepared snapshot,再开始一次后台 promotion;list、search、page 及其他只读 observation 不会激活 Agent。
持久日志用 seq 与 page 修复缺失后缀;Session control 和 Workspace state 用 opening snapshot 收敛;普通 Remote Event 不承诺重放。恢复语义由数据类型决定,不再互相模拟。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-20-client-session-conversation-ownership.md
2026-08-20-client-session-conversation-ownership.md: 8e5521ff1981d83ab72db00dea556b4b2acc97fa
2026-08-20-client-session-conversation-ownership.zh.md: a007a42b3d10ceeced8a2a64696521a96382f4e5
2026-08-20-client-session-conversation-ownership.md: 12e137209bb43d21c3437d2ce5e9d2f9180bbc77
2026-08-20-client-session-conversation-ownership.zh.md: f0d9861eeafc07481c536b85f5ba9667573a66ef

View file

@ -82,7 +82,7 @@ Adding a target does not add a branch to the renderer or Session Controller. The
| `client/ui-session` | Session scope, standard sources, `SessionProvider`, and pending-interaction aggregation | Session transport, Conversation assembly, Approval/Question results |
| `client/ui-workspace` | Workspace hook, browser UI, and cross-Controller navigation policy | Workspace transport, copies of Session data |
| `client/ui-conversation` | Conversation core, registries, bindings, shell, input, composer, queue, and View navigation | Session transport, Chat/Trajectory snapshots |
| `client/ui-chat` | Chat target, Node definitions, renderers, selection, details, locale, and historical images | Session lifecycle, generic View navigation, Trajectory |
| `client/ui-chat` | Chat target, Node definitions, renderers, selection, details, and locale | Session lifecycle, generic View navigation, Trajectory, historical-image cache |
| `client/ui-trajectory` | Trajectory target, event-record projection, and inspection view | Session snapshots, Chat snapshots |
| `client/ui-approval` | Pending Approval, Remote listener, composer, and approval UI | Session control, generic composer election |
| `client/ui-user-questions` | Pending Question, Remote listener, composer, and question UI | Session control, generic composer election |
@ -296,13 +296,13 @@ Draft and input state belong to Conversation UI and do not enter the Session sna
### Chat owner
`client/ui-chat` registers target id `chat` and owns the Chat snapshot builder, Conversation Node definitions, keyed node renderers, selection, details, statistics, locale, Tool-inspection collaboration, and historical-image cache.
`client/ui-chat` registers target id `chat` and owns the Chat snapshot builder, Conversation Node definitions, keyed node renderers, selection, details, statistics, locale, and Tool-inspection collaboration.
It registers the `chat` target source through `ctx.uiSession.provide()`. `ChatNodeSeat` and internal Chat consumers use `useChat` instead of passing `useConversation(snapshot => snapshot.views.get('chat'))`.
Only visible non-command Chat Nodes activate Chat. Ordinary command-only history keeps the Hero visible; the `/goal` `command-input` Node activates a fresh Conversation.
The historical-image cache's Session key, pending promise, generation guard, blob URL, and disposer all belong to `ui-chat`; draft images remain part of Conversation input.
The historical-image cache moved to `ui-conversation` (`ctx.uiConversation.imageUrl`), so Chat and Trajectory share one authorized read and one browser URL per session attachment ([Trajectory durable image attachments](../feature/2026-08-24-trajectory-image-attachments.md)); draft images remain part of Conversation input.
### Trajectory owner

View file

@ -82,7 +82,7 @@ UI 层可以同时读取多个 Controller 做一次导航决定,但不得把
| `client/ui-session` | Session scope、标准 source、`SessionProvider`、pending interaction 聚合 | Session transport、Conversation 组装、Approval/Question 结果 |
| `client/ui-workspace` | Workspace hook、浏览器 UI 和跨 Controller 导航策略 | Workspace transport、Session 数据副本 |
| `client/ui-conversation` | Conversation core、registry、binding、shell、input、composer、queue 和 View 导航 | Session transport、Chat/Trajectory snapshot |
| `client/ui-chat` | Chat target、Node definitions、renderer、selection、details、locale 和历史图片 | Session 生命周期、通用 View 导航、Trajectory |
| `client/ui-chat` | Chat target、Node definitions、renderer、selection、details 和 locale | Session 生命周期、通用 View 导航、Trajectory、历史图片 cache |
| `client/ui-trajectory` | Trajectory target、事件记录投影和检查视图 | Session snapshot、Chat snapshot |
| `client/ui-approval` | Pending Approval、Remote listener、composer 和审批 UI | Session control、通用 composer election |
| `client/ui-user-questions` | Pending Question、Remote listener、composer 和问题 UI | Session control、通用 composer election |
@ -296,13 +296,13 @@ Draft 与输入状态属于 Conversation UI,不进入 Session snapshot。Queue
### Chat owner
`client/ui-chat` 注册 target id `chat`,并拥有 Chat snapshot builder、Conversation Node definitions、keyed node renderers、selection、details、stats、locale、tool inspection 协作和历史图片 cache。
`client/ui-chat` 注册 target id `chat`,并拥有 Chat snapshot builder、Conversation Node definitions、keyed node renderers、selection、details、stats、locale 和 tool inspection 协作。
它通过 `ctx.uiSession.provide()` 注册 `chat` target source。`ChatNodeSeat` 和 Chat 内部消费者使用 `useChat`,不再传递 `useConversation(snapshot => snapshot.views.get('chat'))`。
Chat activity 只由可见且非 command 的 Chat Node 激活。普通 command-only history 保持 Hero,`/goal` 的 `command-input` Node 激活 fresh Conversation。
历史图片 cache 的 Session key、pending promise、generation guard、blob URL 和 disposer 同属 `ui-chat`;Draft 图片仍属于 Conversation input。
历史图片 cache 已移入 `ui-conversation`(`ctx.uiConversation.imageUrl`),Chat 与 Trajectory 对同一会话附件共享一次授权读取和一个浏览器 URL([Trajectory 持久化图片附件](../feature/2026-08-24-trajectory-image-attachments.zh.md));Draft 图片仍属于 Conversation input。
### Trajectory owner

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-20-webworker-pack-lowering-and-preview.md
2026-08-20-webworker-pack-lowering-and-preview.md: 72a6ccf856c95f835e103bd355223bf3cf42f692
2026-08-20-webworker-pack-lowering-and-preview.zh.md: 074d44833c34a2999a5c47da809533065201b580
2026-08-20-webworker-pack-lowering-and-preview.md: 1ec8fb050445b0a90d8fbf0d97c9ef10cb28b287
2026-08-20-webworker-pack-lowering-and-preview.zh.md: 86da66560b509b38ba3ffa49d58035f3e5a173f1

View file

@ -10,9 +10,9 @@ The browser worker can neither compile modules at load nor be served by the prod
## Decision
**Lowering happens at pack time only.** `@deepseek-ai/dsh-experimental-webworker-packer` composes the profile, materializes the closure, and lowers every JavaScript body; `LOWERING_VERSION` and `WRAPPER_PARAMS` are the pack↔worker contract and live in `src/image-layout.ts` beside the rest of the image layout. The loader wraps bodies exactly as the image holds them: a body still carrying module syntax is a refusal naming the image, and `startWorkerHost` requires the manifest's `lowered` to equal this build's contract before it mounts a single module. `lowerModuleSource` is the transform's only face and the packer its only caller; inside the worker graph, imports name the module that owns the value — never the package barrel, which is the edge that smuggled the parser in.
**Lowering happens at pack time only.** `@deepseek-ai/dsh-experimental-webworker-packer` composes the profile, materializes the closure, and lowers every JavaScript body; `LOWERING_VERSION` and `WRAPPER_PARAMS` are the pack↔worker contract and live in `src/image-layout.ts` beside the rest of the image layout. The loader wraps bodies exactly as the image holds them: a body still carrying module syntax is a refusal naming the image, and `startWorkerHost` requires the manifest's `lowered` to equal this build's contract before it mounts a single module. `lowerModuleSource` is the transform's only face and the packer its only caller; inside the worker graph, imports name the module that owns the value — never the package barrel, which is the edge that smuggled the parser in. Source-directory exclusion applies only to workspace and vendored packages whose runtime plane is built `lib/`; installed third-party packages retain JavaScript under `src/` and `dist/` because their published entrypoints may resolve there.
**The preview is the served page plus one tag.** One Vite build emits `dist/index.html` and `dist/preview.html` sharing every chunk; the only difference is a prepended bootstrap entry whose module connects the worker host. Startup then converges on one protocol: whichever side applies the injection table settles the `__DSH_BOOT_READY__` deferred — the served renderer resolves it in a tail script after the rendered rows, the worker bootstrap installs it before its first await and settles it after the last row — and the client entry awaits it before reading any injected state, so the chain from the stock entry onward is the served chain verbatim. The build uses a relative base so the output mounts under any static directory; the served form anchors deep SPA-fallback paths by rendering `<base href="/">` at serve time, keeping the on-disk pages byte-shared.
**The preview is the served page plus one tag.** One Vite build emits `dist/index.html` and `dist/preview.html` sharing every chunk; the only difference is a prepended bootstrap entry whose module connects the worker host. Startup then converges on one protocol: whichever side applies the injection table settles the `__DSH_BOOT_READY__` deferred — the served renderer resolves it in a tail script after the rendered rows, the worker bootstrap installs it before its first await and settles it after the last row — and the client entry awaits it before reading any injected state, so the chain from the stock entry onward is the served chain verbatim. Plugin combo scripts and maps travel through the tunnel; the page-side loader embeds each tunnel-only map as a Base64 data URL before executing its script Blob, preserving indexed-map component names in DevTools without another object-URL lifetime. The build uses a relative base so the output mounts under any static directory; the served form anchors deep SPA-fallback paths by rendering `<base href="/">` at serve time, keeping the on-disk pages byte-shared.
**The repository preview carries selectable filesystem sources.** The packer emits one base image and a small overlay archive for each named built-in fixture. Without a source query, `preview.html` waits at a chooser for an empty filesystem, the built-in fixtures, or the separately owned WebFS provider. A valid `preview-fixture=none|<built-in-id>` query selects directly and skips the chooser for deterministic browser runs; its distinct name avoids the Client's existing `fixture` transport switch. The Worker mounts the base and then applies the selected overlays in order, restricted to `home/` and `workspace/`, before it validates the base manifest or boots Cordis. `packages/experimental/webworker-runtime/tests/fixtures/vfs-example/` supplies one built-in overlay without giving the packer Session or Workspace knowledge. Its plaintext JSONL logs use the persistence backend's real project/session directory layout, so Session Persistence reads them cold and Workspace Registry derives the Workspace from their `/dsh/workspace` headers. The main Session exceeds the Client's 50-message page and keeps representative tool results at its tail; persisted one-shot and continuable children exercise the subagent catalog. WebFS authorization and user data remain a separate provider and never share this fixture tree.

View file

@ -10,9 +10,9 @@
## 决定
**Lowering 只发生在 pack 期。** `@deepseek-ai/dsh-experimental-webworker-packer` 组合 profile、物化闭包、lower 每个 JavaScript 模块体;`LOWERING_VERSION` 与 `WRAPPER_PARAMS` 是 pack↔worker 的契约,与镜像布局的其余部分一起放在 `src/image-layout.ts`。装载器完全按镜像持有的形态包装模块体:仍带模块语法的模块体是一次点名镜像的拒绝,且 `startWorkerHost` 在挂载任何模块之前要求 manifest 的 `lowered` 等于本构建的契约。`lowerModuleSource` 是转换器唯一的面、packer 是它唯一的调用方;worker 图内部的 import 一律指向拥有该值的模块——绝不指向包 barrel,那正是把解析器偷运进来的那条边。
**Lowering 只发生在 pack 期。** `@deepseek-ai/dsh-experimental-webworker-packer` 组合 profile、物化闭包、lower 每个 JavaScript 模块体;`LOWERING_VERSION` 与 `WRAPPER_PARAMS` 是 pack↔worker 的契约,与镜像布局的其余部分一起放在 `src/image-layout.ts`。装载器完全按镜像持有的形态包装模块体:仍带模块语法的模块体是一次点名镜像的拒绝,且 `startWorkerHost` 在挂载任何模块之前要求 manifest 的 `lowered` 等于本构建的契约。`lowerModuleSource` 是转换器唯一的面、packer 是它唯一的调用方;worker 图内部的 import 一律指向拥有该值的模块——绝不指向包 barrel,那正是把解析器偷运进来的那条边。源码目录排除只用于运行期使用已构建 `lib/` 的 workspace 与 vendored 包;已安装第三方包会保留 `src/` 和 `dist/` 下的 JavaScript,因为其发布入口可能解析到这些位置。
**preview 就是服务页面加一个标签。** 一次 Vite 构建产出共享全部 chunk 的 `dist/index.html` 与 `dist/preview.html`;唯一差异是前插的一个引导入口,其模块负责连接 worker host。启动随之汇于一个协议:应用注入表的一方 settle `__DSH_BOOT_READY__` deferred——served 渲染器在渲染完的行之后用尾部脚本 resolve,worker 引导段在首个 await 之前安装、末行生效后 settle——client 入口在读取任何注入状态前 await 它,因此从标准入口起的链路逐字就是 served 链路。构建使用相对 base,产物可挂载于任意静态目录;served 形态在 serve 期渲染 `<base href="/">` 锚定深层 SPA fallback 路径,磁盘上的两个页面保持字节共享。
**preview 就是服务页面加一个标签。** 一次 Vite 构建产出共享全部 chunk 的 `dist/index.html` 与 `dist/preview.html`;唯一差异是前插的一个引导入口,其模块负责连接 worker host。启动随之汇于一个协议:应用注入表的一方 settle `__DSH_BOOT_READY__` deferred——served 渲染器在渲染完的行之后用尾部脚本 resolve,worker 引导段在首个 await 之前安装、末行生效后 settle——client 入口在读取任何注入状态前 await 它,因此从标准入口起的链路逐字就是 served 链路。插件 combo 脚本与 map 都通过 tunnel;页面侧 loader 会在执行脚本 Blob 前,把每个仅 tunnel 可达的 map 内嵌为 Base64 data URL,从而不依赖另一条 object URL 的生命周期,并在 DevTools 中保留 indexed map 的组件名称。构建使用相对 base,产物可挂载于任意静态目录;served 形态在 serve 期渲染 `<base href="/">` 锚定深层 SPA fallback 路径,磁盘上的两个页面保持字节共享。
**仓库 preview 携带可选择的文件系统来源。** Packer 产出一份基础镜像,并为每套具名内置 fixture 产出一份小型 overlay 归档。没有来源 query 时,`preview.html` 会停在选择面板,可选择空文件系统、内置 fixtures,或归另一实现所有的 WebFS provider。合法的 `preview-fixture=none|<built-in-id>` query 会直接选择并跳过面板,供确定性的浏览器流程使用;该独立名称避开 Client 既有的 `fixture` transport 开关。Worker 先挂载基础镜像,再按顺序把所选 overlays 应用到仅限 `home/` 和 `workspace/` 的路径,随后才校验基础 manifest 并启动 Cordis。`packages/experimental/webworker-runtime/tests/fixtures/vfs-example/` 提供其中一套内置 overlay,Packer 无需理解 Session 或 Workspace。明文 JSONL 日志使用 persistence backend 的真实 project/session 目录布局,因此 Session Persistence 会冷读取它们,Workspace Registry 则根据其 `/dsh/workspace` header 派生 Workspace。主 Session 超过 Client 的 50-message page,并把代表性工具结果留在尾页;持久化的 one-shot 与 continuable child 用于验证 subagent catalog。WebFS 授权与用户数据仍属于独立 provider,绝不与该 fixture 共用目录。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-22-single-dsh-application-launcher.md
2026-08-22-single-dsh-application-launcher.md: 4640173068998d518f5cbdf537526ea479242b50
2026-08-22-single-dsh-application-launcher.zh.md: 4f42652497b84a431a88442fe812045550bcb4ff
2026-08-22-single-dsh-application-launcher.md: feac31b3eafced8158a8d79a0e5967a8de5e2f87
2026-08-22-single-dsh-application-launcher.zh.md: 88fe4ef4d4e0436ecb450e3f5319be7acee6882b

View file

@ -46,7 +46,7 @@ Direct SDK use follows normal Harness-home resolution: explicit `dshHome`, inher
### Python runtime
The Python runtime wheel packages the ordinary `@deepseek-ai/dsh` CLI from `node_modules/@deepseek-ai/dsh/lib/bin.js` through the private `dsh-python-runtime-closure` deploy manifest. The Python client selects `dsh --profile sdk` by default, ordered patch files, and an explicit Harness home; the runnable Python example selects `sdk-minimal`. The installed `dsh` console command exposes the same profile grammar and the separately packaged `web` application.
The Python runtime wheel packages the ordinary `@deepseek-ai/dsh` CLI from `node_modules/@deepseek-ai/dsh/lib/bin.js` through the private `dsh-python-runtime-closure` deploy manifest. The Python client selects `dsh --profile sdk` by default, ordered patch files, and an explicit Harness home; the runnable example under `python/sdk/examples` selects `sdk-minimal`. The installed `dsh` console command exposes the same profile grammar and the separately packaged `web` application.
The executable family is `deepseek-harness-sdk-runtime-<platform>-<arch>`. The SDK wire, wheel and import distribution names, sidecar names, and wire identity `deepseek-harness-sdk-runtime` remain stable. The SDK package family is `@deepseek-ai/dsh-sdk-client`, `@deepseek-ai/dsh-sdk-protocol`, and `@deepseek-ai/dsh-sdk-jsonrpc-server`; `@deepseek-ai/dsh-acp` remains the ACP protocol plugin. There is no Python-specific Node application, checked-in complete config, compatibility package, forwarding executable, fallback parser, or SDK/ACP launcher alias. The [Python profile-runtime decision](2026-08-23-python-sdk-dsh-profile-runtime.md) owns this launch, and the [Windows x64 runtime decision](2026-08-23-python-sdk-windows-x64-runtime.md) owns the fourth carrier.

View file

@ -46,7 +46,7 @@ SDK 用户通过 profile 自定义插件。`dsh plugin --profile <name> ...` 管
### Python 运行时
Python 运行时 wheel 通过私有 `dsh-python-runtime-closure` 部署 manifest,打包来自 `node_modules/@deepseek-ai/dsh/lib/bin.js` 的普通 `@deepseek-ai/dsh` CLI。Python 客户端默认选择 `dsh --profile sdk`、有序 patch 文件与显式 Harness home;可运行 Python 示例选择 `sdk-minimal`。安装的 `dsh` 控制台命令暴露相同 profile 语法与单独打包的 `web` 应用。
Python 运行时 wheel 通过私有 `dsh-python-runtime-closure` 部署 manifest,打包来自 `node_modules/@deepseek-ai/dsh/lib/bin.js` 的普通 `@deepseek-ai/dsh` CLI。Python 客户端默认选择 `dsh --profile sdk`、有序 patch 文件与显式 Harness home;`python/sdk/examples` 下的可运行示例选择 `sdk-minimal`。安装的 `dsh` 控制台命令暴露相同 profile 语法与单独打包的 `web` 应用。
可执行文件族是 `deepseek-harness-sdk-runtime-<platform>-<arch>`。SDK 协议格式、wheel 与 import 分发名称、伴随文件名称,以及协议 identity `deepseek-harness-sdk-runtime` 保持稳定。SDK 包族是 `@deepseek-ai/dsh-sdk-client`、`@deepseek-ai/dsh-sdk-protocol` 与 `@deepseek-ai/dsh-sdk-jsonrpc-server`;`@deepseek-ai/dsh-acp` 继续作为 ACP 协议插件。仓库不保留 Python 专用 Node 应用、检入的完整配置、兼容包、转发可执行文件、后备解析器或 SDK/ACP 启动别名。[Python profile 运行时决策](2026-08-23-python-sdk-dsh-profile-runtime.zh.md)负责该启动方式,[Windows x64 运行时决策](2026-08-23-python-sdk-windows-x64-runtime.zh.md)负责第四个载体。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-23-python-sdk-dsh-profile-runtime.md
2026-08-23-python-sdk-dsh-profile-runtime.md: dcb4f77048d516f0e187b611dc09c224fba47b58
2026-08-23-python-sdk-dsh-profile-runtime.zh.md: 505b49506ae4c4be4809d588448b2c593d3ebaf5
2026-08-23-python-sdk-dsh-profile-runtime.md: 4af7812db6818b65c754a43ec1a7f973d1cbcbf9
2026-08-23-python-sdk-dsh-profile-runtime.zh.md: 155e7d2ae0b0ba3e4163dd85a31de90bef9d588a

View file

@ -24,7 +24,7 @@ Every Python launch requires either explicit `dsh_home` or a non-empty `DSH_HOME
Persistent SDK customization uses the same profile interfaces as direct CLI use. `dsh plugin --profile <name> ...` manages external dependencies and bundle order, `$DSH_HOME/profiles/<name>/cordis.patch.yml` owns persistent row changes, the home patch applies machine-local changes across profiles, and Python `patches` supplies invocation-specific overlays. A selected profile is valid only when it retains an SDK server row. Missing profiles, bundles, server rows, and invalid patches fail without a complete-config fallback; a profile that remains alive without serving JSON-RPC fails the independently bounded initialization handshake with a diagnostic naming that profile.
The [standalone sdk-minimal profile](2026-08-24-standalone-sdk-minimal-profile.md) lists one repository-owned bundle that inserts its complete explicit tree without `dsh-base`. Its persistent Bash and string-replace editor are present by composition rather than a server filter; dynamic runtime context, workspace instructions, settings, managed credentials, telemetry, compaction, and every other base row are absent. The same runtime still packages the full `sdk` and `web` profiles as separate choices.
The [standalone sdk-minimal profile](2026-08-24-standalone-sdk-minimal-profile.md) lists one repository-owned bundle that inserts its complete explicit tree without `dsh-base`. Its persistent Bash and string-replace editor are present by composition; the shared JSON-RPC server exposes no root-agent tool filter. Dynamic runtime context, workspace instructions, settings, managed credentials, telemetry, compaction, and every other base row are absent. The same runtime still packages the full `sdk` and `web` profiles as separate choices.
The runtime wheel installs a `dsh` console command. Ordinary profile and SDK execution remains Node-free; external package management requires a caller-installed `pnpm`.

View file

@ -24,7 +24,7 @@ Python SDK 分发一个私有 Node 应用,直接启动完整外部 `cordis.yml
持久 SDK 自定义使用与直接 CLI 相同的 profile 接口。`dsh plugin --profile <name> ...` 管理外部依赖与 bundle 顺序,`$DSH_HOME/profiles/<name>/cordis.patch.yml` 负责持久配置项变更,home patch 对所有 profile 应用机器本地变更,Python `patches` 则提供单次启动 overlay。所选 profile 只有保留 SDK server 配置项时才有效。缺失 profile、bundle、server 配置项或非法 patch 都会直接失败,不存在完整配置回退;保持运行却不提供 JSON-RPC 服务的 profile 会在独立有界的初始化握手中失败,诊断会指明该 profile。
[独立 sdk-minimal profile](2026-08-24-standalone-sdk-minimal-profile.zh.md)只列出一个仓库自有组合包,该组合包会插入不含 `dsh-base` 的完整显式配置树。持久 Bash 与字符串替换 editor 通过组合存在,而不是通过 server 筛选;动态运行时上下文、workspace 指令、settings、托管凭据、遥测、compaction 与其他所有 base 配置项均不存在。同一运行时仍会把完整 `sdk` 与 `web` profile 作为独立选择打包。
[独立 sdk-minimal profile](2026-08-24-standalone-sdk-minimal-profile.zh.md)只列出一个仓库自有组合包,该组合包会插入不含 `dsh-base` 的完整显式配置树。持久 Bash 与字符串替换 editor 通过组合存在;共享 JSON-RPC server 不暴露根 agent 工具筛选器。动态运行时上下文、workspace 指令、settings、托管凭据、遥测、compaction 与其他所有 base 配置项均不存在。同一运行时仍会把完整 `sdk` 与 `web` profile 作为独立选择打包。
运行时 wheel 安装 `dsh` 控制台命令。普通 profile 与 SDK 运行仍不需要 Node;外部包管理要求调用方自行安装 `pnpm`。

View file

@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-24-browser-token-authentication.md
2026-08-24-browser-token-authentication.md: c75f561d9529296ff668791c29453e522f309cc3
2026-08-24-browser-token-authentication.zh.md: d4a5059619fefda9d9060e9879d10c0a2197f8f3

View file

@ -0,0 +1,47 @@
# Agent Note: Browser launch-token authentication
Status: implemented
English | [中文](2026-08-24-browser-token-authentication.zh.md)
## Problem
The Web Host runs tool-capable Sessions with the current operating-system user's authority, but its HTTP interface identified privileged callers from request routing facts. In particular, the method-specific loopback list treated a loopback `Host` value as local authority even though an HTTP client controls that header. A caller that could reach the server could therefore name `localhost`, enter configuration methods, and use Host-side operations such as model discovery to disclose stored credentials. Binding the shipped CLI to loopback limits ordinary reachability but does not authenticate a request forwarded or otherwise delivered to that socket.
## Decision
`dsh-client-connection` authenticates the complete Host API before dispatch. Every API Proxy method, Remote unary call, generic Connection channel, and Remote WebSocket stream requires the same browser session; endpoint ownership and method names do not alter authority. The existing Host/Origin checks run first and retain their DNS-rebinding and cross-site-request role, returning 403 when they fail. A trusted Host without a valid browser session receives 401. The browser-trust rules remain owned by the [carrier-level browser trust decision](2026-07-28-api-browser-trust-boundary.md).
Each Host process generates a random launch token, retained by the root application context across Connection hot reloads. `dsh-web-app` prints and opens the normal root URL with that token in the query once per process. `frontend-static` asks Connection to authorize index responses: only `GET /?token=...` exchanges the process token for a cookie, then redirects to clean `/`; the token is not accepted on API paths or in an Authorization header. An obsolete token paired with a valid cookie redirects to clean `/`. Missing and invalid credentials receive one minimal 401 response. Static non-index assets remain public.
The cookie is a signed, authority-bound bearer. Its deterministic name and signed payload both include the normalized hostname plus port, so one Harness home can run independent Web ports without cookie collisions. The payload carries safe-integer issue and expiry times under an absolute lifetime; `cookieMaxAgeDays` defaults to 30. The cookie is host-only, `Path=/`, `HttpOnly`, and `SameSite=Strict`. It omits `Secure` because the shipped server uses loopback HTTP. There is no logout operation or reverse-proxy-specific handling.
The HMAC secret is a versioned `grant` record at `client-connection/browser-session` in `ctx.credentials`; the local provider stores it in `$DSH_HOME/.credentials.yaml`. Connection loads or creates the record during activation and retains the secret for synchronous request verification. An active Connection continues using its loaded secret if the durable record changes; the next activation loads the replacement or creates a missing record, so deleting the record and restarting the process revokes every existing cookie. Invalid owner payloads fail loud instead of being replaced. The launch token itself is never persisted and changes on every process start, while an unexpired cookie remains valid across restarts on the same authority.
The in-page Web Worker preview exposes no network socket. Its page-owned `postMessage` tunnel enters the real route first, then retries a 401 or 403 through the worker-local fetch handler. This keeps Connection interceptors while limiting the authentication bypass to the page that created the Host worker.
The shipped CLI continues to reject `--host 0.0.0.0`. Authentication does not imply supported network deployment, TLS, forwarding-header interpretation, or proxy configuration.
## Verification
Unit coverage pins process-token retention across Connection reloads, one secret load per activation, synchronous verification without credential-provider reads, cookie attributes, HMAC and payload validation, authority and lifetime checks, record deletion taking effect on the next activation, invalid durable records, and cleanup of obsolete token URLs backed by valid cookies. Host transport suites pin uniform 401/403 behavior for API Proxy, generic RPC, Typert Remote HTTP, and WebSocket upgrade paths. The frontend real-composition test boots credentials, Connection, webserver, and static serving through Loader and proves token exchange before index reads while static assets remain public. Packed-worker tests prove portable cookie encoding and worker-local retry for both authentication and trust rejection. A real-CLI test starts `dsh web` twice on one port with a temporary `DSH_HOME`, proves that forged `Host: localhost` is unauthenticated, calls `host.describe` with the exchanged cookie, observes a new process token, and reuses the old cookie after restart.
## Alternatives considered
**Determine privileged callers from the TCP peer address.** A direct peer address still identifies a local forwarding process rather than the browser user, retains a second authority model beside the API's command-execution capability, and requires proxy policy to answer who the original caller was. One application credential is the enforceable identity used for every operation.
**Keep a method-specific privileged list and restrict stored credentials to configured targets.** The list can omit new endpoints and does not constrain callers that already control a tool-capable Session. A `discoverModels` target rule would not form a security boundary because the same authenticated principal can update settings and run commands. Uniform authentication covers the operation that grants process control.
**Persist or accept the launch token as an API bearer.** A durable launch token would become a second long-lived credential, while Authorization-header support would add a non-browser client contract with no current consumer. The process token performs one browser-cookie exchange only.
**Rotate the signing secret on every restart.** This prevents an existing browser from reconnecting after an ordinary DSH restart. Persisting only the signing secret keeps that workflow while process-token rotation limits the startup URL to one process lifetime.
**Add logout, TLS-proxy, and forwarding-header configuration.** None is required by the loopback Web application or the reported authentication gap. Adding them would define deployment contracts without current consumers. Browser site-data controls revoke one browser session; deleting the credential record and restarting the process revokes all sessions.
## Consequences
Possession of the browser cookie authorizes the complete tool-capable Host API, matching the authority the Web application exposes after Session creation. `Host` does not grant a higher method tier, and a method migration between API Proxy and Typert Remote cannot change its caller set.
The persistent secret makes cookies survive restarts but gives a stolen cookie up to the configured absolute lifetime. Deleting the record and restarting the process is the global revocation mechanism; the active Connection intentionally avoids credential-provider work on each request. Omitting `Secure` preserves loopback HTTP and permits plaintext transmission if an operator makes the same cookie authority reachable over an unencrypted network. The startup URL contains a process credential and must be treated as sensitive output; runtime diagnostics do not repeat it.
The decision partially supersedes the authentication deferral and unauthenticated non-loopback consequences in the [browser trust note](2026-07-28-api-browser-trust-boundary.md). That note remains active authority for media-type, Host, Origin, Fetch-Metadata, and configured-authority validation. No active Agent Note is archived: the overlap is partial and both security rules retain future decision value.

View file

@ -0,0 +1,47 @@
# Agent Note: 浏览器启动令牌认证
Status: implemented
[English](2026-08-24-browser-token-authentication.md) | 中文
## 问题
Web Host 以当前操作系统用户的权限运行具有工具能力的 Session,但其 HTTP 接口用请求路由事实识别特权调用者。具体而言,按方法维护的 loopback 列表把 loopback `Host` 值视为本地 authority,尽管 HTTP 客户端可以控制该 header。能够到达服务器的调用者因此可以声明 `localhost`、进入配置方法,再利用模型发现等 Host 侧操作披露存储的凭据。随附 CLI 绑定 loopback 可以限制普通可达性,却不能认证被转发或以其他方式送达该 socket 的请求。
## 决策
`dsh-client-connection` 在分发前认证完整 Host API。每个 API Proxy 方法、Remote 一元调用、通用 Connection channel 和 Remote WebSocket stream 都要求同一个浏览器会话;endpoint 所有权与方法名称不改变 authority。既有 Host/Origin 校验先执行,继续负责 DNS rebinding 和跨站请求防御,失败时返回 403。Host 可信但没有有效浏览器会话时返回 401。浏览器信任规则仍由[载体级浏览器信任决策](2026-07-28-api-browser-trust-boundary.zh.md)持有。
每个 Host 进程生成随机启动令牌,并由应用根 context 跨 Connection 热重载保留。`dsh-web-app` 每个进程只打印并打开一次 query 中带该令牌的普通根 URL。`frontend-static` 请求 Connection 授权 index 响应:只有 `GET /?token=...` 会把进程令牌交换为 cookie,再重定向到干净的 `/`;API 路径和 Authorization header 都不接受该令牌。过时令牌如果同时带有有效 cookie,会重定向到干净的 `/`。缺失与无效凭据得到同一份最小 401 响应。非 index 静态资产保持公开。
cookie 是签名且绑定 authority 的 bearer。确定性名称与签名 payload 都包含规范化 hostname 和 port,因此同一 Harness home 可以在不同 Web port 运行而不发生 cookie 冲突。payload 在绝对有效期内携带安全整数形式的签发与过期时间;`cookieMaxAgeDays` 默认为 30。cookie 是 host-only、`Path=/`、`HttpOnly`、`SameSite=Strict`。随附服务器使用 loopback HTTP,因此不设置 `Secure`。这里没有 logout 操作或反向代理专用处理。
HMAC 密钥是 `ctx.credentials` 中位于 `client-connection/browser-session` 的版本化 `grant` 记录;本地提供方将其存入 `$DSH_HOME/.credentials.yaml`。Connection 在激活期间加载或创建该记录,并保留密钥以同步校验请求。持久记录发生变化后,当前 Connection 继续使用已加载的密钥;下一次激活会加载替换记录或创建缺失记录,因此删除记录并重启进程会撤销全部既有 cookie。无效 owner payload 会明确失败,而不是被覆盖。启动令牌本身绝不持久化并在每次进程启动时变化;未过期 cookie 则能在相同 authority 上跨重启继续有效。
页内 Web Worker preview 不暴露网络 socket。其由页面持有的 `postMessage` tunnel 先进入真实 route,收到 401 或 403 后再经 worker 本地 fetch handler 重试。这样既保留 Connection interceptor,又把认证绕过限制在创建 Host worker 的页面内。
随附 CLI 继续拒绝 `--host 0.0.0.0`。认证不代表支持网络部署、TLS、转发 header 解释或代理配置。
## 验证
单元覆盖 Connection 重载时保留进程令牌、每次激活只加载一次密钥、无需读取凭据提供方的同步校验、cookie 属性、HMAC 与 payload 校验、authority 与有效期校验、记录删除在下一次激活时生效、无效持久记录,以及用有效 cookie 清理过时令牌 URL。Host 传输套件固定 API Proxy、通用 RPC、Typert Remote HTTP 和 WebSocket upgrade 路径上一致的 401/403 行为。frontend 真实组合测试经 Loader 启动 credentials、Connection、webserver 与静态服务,证明读取 index 前完成令牌交换,同时静态资产仍公开。打包 worker 测试证明 cookie 编码可移植,并覆盖认证与信任拒绝后的 worker 本地重试。真实 CLI 测试在临时 `DSH_HOME` 上用同一端口两次启动 `dsh web`,证明伪造 `Host: localhost` 仍未认证,以交换所得 cookie 调用 `host.describe`,观测新的进程令牌,并在重启后复用旧 cookie。
## 曾考虑的替代方案
**从 TCP 对端地址判定特权调用者。** 直接对端地址仍可能只识别本地转发进程,而非浏览器用户;它会在 API 的命令执行能力旁保留第二套 authority 模型,并要求代理策略回答原始调用者是谁。一个应用凭据才是每项操作都能执行的身份。
**保留按方法的特权列表,并把存储凭据限制在已配置目标。** 列表可能漏掉新 endpoint,也不能约束已经控制工具型 Session 的调用者。`discoverModels` 目标规则不构成安全边界,因为同一已认证主体可以更新 settings 并运行命令。统一认证覆盖授予进程控制权的操作。
**持久化启动令牌或把它作为 API bearer 接受。** 持久启动令牌会成为第二份长期凭据;Authorization header 支持则会增加没有当前 consumer 的非浏览器客户端约定。进程令牌只完成一次浏览器 cookie 交换。
**每次重启都轮换签名密钥。** 这会阻止既有浏览器在普通 DSH 重启后重连。只持久化签名密钥既保留该工作流,又由进程令牌轮换把启动 URL 限定在一个进程生命周期。
**增加 logout、TLS 代理和转发 header 配置。** loopback Web 应用与已报告认证缺口都不需要这些能力;加入它们会在没有当前 consumer 时定义部署约定。浏览器站点数据控制会撤销单个浏览器会话;删除凭据记录并重启进程会撤销全部会话。
## 后果
持有浏览器 cookie 就能调用完整的工具型 Host API,这与 Web 应用在创建 Session 后暴露的 authority 一致。`Host` 不授予更高的方法层级,方法在 API Proxy 与 Typert Remote 之间迁移也不会改变调用者集合。
持久密钥使 cookie 跨重启生效,也让被盗 cookie 最多保有配置的绝对有效期。删除记录并重启进程是全局撤销机制;当前 Connection 刻意避免在每个请求上访问凭据提供方。不设置 `Secure` 保留 loopback HTTP,但如果操作者让同一 cookie authority 经未加密网络可达,cookie 会以明文传输。启动 URL 含进程凭据,必须视为敏感输出;运行时诊断不会重复它。
本决策部分取代[浏览器信任说明](2026-07-28-api-browser-trust-boundary.zh.md)中的认证延期与未认证非 loopback 后果。该说明仍是媒体类型、Host、Origin、Fetch-Metadata 和配置 authority 校验的有效权威。没有 active Agent Note 被归档:重叠只发生在局部,两条安全规则都保有未来决策价值。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-24-standalone-sdk-minimal-profile.md
2026-08-24-standalone-sdk-minimal-profile.md: 692bf9763f4ad710ff5cc819a7480b2f3e8d9b5f
2026-08-24-standalone-sdk-minimal-profile.zh.md: f39baad1b4429b73f13d601716dadd8376c71bfb
2026-08-24-standalone-sdk-minimal-profile.md: 9c8afbaaf1a8e522af119c1d42ca5ad714eaa879
2026-08-24-standalone-sdk-minimal-profile.zh.md: b1cd2a348c4b77f197e30a658c13691b2f35d26e

View file

@ -40,11 +40,11 @@ It also supersedes the minimal-overlay realization in [Python SDK runtime throug
## Verification
The bundle test pins the exact row and dependency roster. Profile-template and config-dump tests pin the one-bundle manifest, startup-only lifecycle, absence of `dsh-base`, and absence of module HMR. The keyless Python example test boots the real `dsh --profile sdk-minimal` process and asserts the generated manifest, complete system prompt, and two advertised tools. The installed-wheel minimal scenario exercises persistent shell state, editor effects, JSONL persistence, and the committed model-visible snapshot through the packaged executable.
The bundle test pins the exact row and dependency roster. Profile-template and config-dump tests pin the one-bundle manifest, startup-only lifecycle, absence of `dsh-base`, and absence of module HMR. The keyless source test boots the real `dsh --profile sdk-minimal` process, completes a turn, and asserts the generated manifest. The installed-wheel minimal scenario owns the complete system prompt and two advertised tools in its committed model-visible snapshot while exercising persistent shell state, editor effects, and JSONL persistence through the packaged executable.
## Alternatives considered
**Keep the minimal mode as an overlay on `sdk`.** Rejected because filtering model-visible tools does not remove base services, prompt contributors, persistence choices, or later runtime behavior. It also makes the minimal application depend on controls in shared SDK server and system-prompt interfaces.
**Keep the minimal mode as an overlay on `sdk`.** Rejected because filtering model-visible tools does not remove base services, prompt contributors, persistence choices, or later runtime behavior. It also required root-tool filtering in the shared SDK server and a complete-persona shortcut in the system-prompt config; neither shared interface carries those composition controls.
**Restore a Python `cordis` argument or environment-selected complete config.** Rejected because it recreates a Python-owned application composition and bypasses profile plugin management and launcher lifecycle.

View file

@ -40,11 +40,11 @@ Python 运行时继续打包 `dsh-web-app` 与前端产物。`dsh web` 会从已
## 验证
组合包测试固定确切配置项与依赖清单。Profile 模板与配置 dump 测试固定单组合包 manifest、仅启动时生命周期、`dsh-base` 缺席与模块 HMR 缺席。Keyless Python 示例测试启动真实 `dsh --profile sdk-minimal` 进程,并断言生成的 manifest、完整系统提示词与两个对外公布的工具。Installed-wheel 极简场景通过打包可执行程序验证持久 shell 状态、editor 文件效果、JSONL 持久化与已提交的模型可见快照。
组合包测试固定确切配置项与依赖清单。Profile 模板与配置 dump 测试固定单组合包 manifest、仅启动时生命周期、`dsh-base` 缺席与模块 HMR 缺席。Keyless 源码测试启动真实 `dsh --profile sdk-minimal` 进程、完成一个回合并断言生成的 manifest。Installed-wheel 极简场景通过已提交的模型可见快照固定完整系统提示词和两个对外公布的工具,同时经由打包可执行程序验证持久 shell 状态、editor 文件效果与 JSONL 持久化。
## 考虑过的替代方案
**继续把极简模式作为 `sdk` 上的 overlay。** 否决:筛选面向模型的工具不会移除 base 服务、提示词贡献方、持久化选择或后续运行时行为,还会让极简应用依赖共享 SDK server 与系统提示词接口中的控制项。
**继续把极简模式作为 `sdk` 上的 overlay。** 否决:筛选面向模型的工具不会移除 base 服务、提示词贡献方、持久化选择或后续运行时行为。该方案还要求共享 SDK server 提供根工具筛选,并要求 system-prompt 配置提供 complete-persona 快捷项;这两个共享接口均不再携带这些组合控制项。
**恢复 Python `cordis` 参数或由环境选择的完整配置。** 否决:这会重新创建 Python 自有应用组合,并绕过 profile 插件管理与 launcher 生命周期。

View file

@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-25-session-observations-and-projection-owned-client-state.md
2026-08-25-session-observations-and-projection-owned-client-state.md: e47f2fc75ecbca51d01af077f6c6ab98f4e275f9
2026-08-25-session-observations-and-projection-owned-client-state.zh.md: 527a4eb6b6765cba95d6067f2be60bff8f31a559

View file

@ -0,0 +1,201 @@
# Agent Note: Session observations and projection-owned client state
Status: implemented
English | [中文](2026-08-25-session-observations-and-projection-owned-client-state.zh.md)
## Problem
Session-facing consumers needed the same logical data but resolved it independently. List, follow, page, attachment and fork reads, and subagent inspection each chose between an attached Session, persisted metadata, a prepared Session, and projection cache entries. One page visit could therefore materialize the same cold log more than once, and independently assembled header, event, cursor, and projection values could describe different cuts.
Client features also kept Session-derived facts in several forms. Title had dedicated list and update handling; model selection mixed a Session-specific catalog request with local state; agent preset display could infer a global default before the current Session arrived; and subagent listing scanned or reconstructed identity separately. These mirrors introduced intermediate states in which the UI showed a guessed default, a raw id, or an unavailable state even though the durable Session already determined the answer.
Unifying only the persistence read would leave those Client mirrors as competing authorities. Unifying only the Client fields would leave each Host endpoint free to obtain a different source cut. The read unit and the derived-state unit therefore need one coordinated ownership rule.
## Decision
Exact Session reads use a retained `SessionObservation`, and replayable Session-derived values exposed to the Client use registered projections. Observation owns source selection and one immutable read cut; projection owns derivation from that cut. API layers select what to publish, while Client code consumes finished values and does not reconstruct Session facts from events or duplicate them in domain-specific mirrors.
### Data flow
The two ownership rules meet at the observation's projection snapshot. Lightweight listing may stop at cached hints; every exact opening reaches the same observation path and gives the Client a complete replacement baseline.
```mermaid
flowchart LR
List["list / search"] --> Corpus["SessionQuery corpus"]
Follow["follow"] --> Observe["observeSession"]
Page["page / attachment / fork"] --> Observe
Subagent["subagent list / continuation"] --> Corpus
Subagent --> Observe
Corpus --> Cache["projection cache hints"]
Cache --> ClientList["Client Session list"]
Cache -->|"small miss"| Observe
Observe --> Source{"live or cold"}
Source --> Live["attached Session cut"]
Source --> Borrow["borrowSession"]
Borrow --> Prepared["SessionPreparations.borrow"]
Live --> Mode{"all or none"}
Prepared --> Mode
Mode --> Snapshot["SessionObservation"]
Snapshot --> Opening["follow opening snapshot"]
Snapshot --> Read["page / inspection"]
Opening --> Store["Client projection store"]
Store --> Domain["title / model / preset / subagent"]
```
### Observation is the point-read unit
`SessionQueryEngine.observeSession(sessionId, options)` returns a disposable `SessionObservation` containing one source kind, header, contiguous event prefix, cursor, optional projection snapshot, and the durable revision for a prepared source. An attached Session wins. Otherwise `SessionPersistence.borrowSession()` and `SessionPreparations.borrow()` share and pin one prepared Session, including an in-flight cold load.
Every owner disposes its observation. `retain()` creates another lease over the same cut, which lets `session.follow` publish a snapshot and then transfer that exact prepared source to background Agent promotion without rereading the log. A live Session that appears during cold resolution wins before publication; a disappeared live source is retried as cold.
### Source resolution and lifetime
An observation binds all returned fields to one lifecycle witness. Callers do not combine a header from corpus listing, events from persistence, and projections from a later live Session. The selected header and event prefix produce the cursor and projection snapshot together.
Live preference is checked both before and after a cold borrow. The second check closes the race in which an Agent attaches while persistence is loading. If persistence itself reports that a live source won but that source has already detached by the time SessionQuery examines it, resolution restarts instead of publishing an unowned reference.
Persistence absence maps to Session-not-found only after no attached Session exists. Durable corruption, source-identity conflict, cancellation, and operational persistence failure remain distinct `SessionQueryError` outcomes so API owners can preserve their own public error vocabulary without duplicating source detection.
The observation owns no mutation authority. Its event array is an immutable prefix, and its prepared Session remains unpublished. Promotion is an explicit ownership transfer performed by the Session Controller after it has emitted the opening snapshot; other readers cannot turn an observation into a live Agent.
Projection work is deliberately `all | none`. `all` computes every registered projection at the observation's event cursor; `none` leaves projection state untouched. There is no per-key preparation state, `projectionKeys` mode, or cached `viewedState`/`viewedValue` layer. A publisher may filter the completed values for an audience, but the underlying observation is never partly projected.
### Projection execution boundary
For a live source, `all` reads one synchronous registry snapshot. For a prepared source, the projection cache may seed valid state rows, after which every registered unit advances over the exact remaining event prefix. The resulting client values share one `asOfSeq`.
Filtering belongs after computation because it changes disclosure, not state. A page authorization check may consume only `subagent`, and a list row may publish only list-relevant values, while both still rely on a complete projected cut when they request projection work.
The registry owns fold state; each domain owns its `init`, `apply`, `view`, schemas, and `stateVersion`. SessionQuery knows only whether projection work is required. It does not know title, model, preset, subagent, token, image, plan, todo, or goal semantics.
`view` remains an uncached synchronous conversion over folded state. Its cost is bounded by the registered projection units and is paid at snapshot publication; introducing a second cache would add invalidation states without reducing event replay.
Corpus listing remains a separate lightweight operation. `listSessions()` returns live-preferred headers without materializing every log. Session list and subagent list first use live projection state or durable projection-cache rows. Session list may take one complete observation for an individually stored artifact within its configured small-log limit when cached metadata cannot establish whether it is blank; a large or unreadable cache miss remains visible with unknown hints.
`session.follow` publishes a required opening snapshot containing header, cursor, the initial event window, and a complete projection baseline. Reconnect replaces the previous generation from another complete snapshot. `session.page` is reserved for older-history reads and gap repair. Observation-only reads never activate an Agent; only an ordinary follow may retain its prepared observation and request promotion after the opening snapshot has been delivered.
### Read audiences
Each public operation chooses one query and projection policy. The choice is part of that operation's behavior rather than a heuristic inside persistence or transport.
| Operation | Read path | Projection policy | Agent activation |
|---|---|---|---|
| `session.list` | Corpus headers, live state, and cached rows; bounded small-log fallback | Partial hints, or one full small-log observation | Never |
| `session.search` | Corpus authorization plus the configured search provider | None for result listing | Never |
| `session.follow` | One exact observation | All, carried in the opening snapshot | Ordinary cold Session only, after snapshot delivery |
| `session.page` | One exact observation | None, except projection-backed subagent authorization | Never |
| Attachment and fork source | One exact observation | None unless authorization requires it | Never for the source |
| Subagent list and continuation | Corpus plus live/cache/observation resolution | All on a cold fallback; audience consumes identity or inherited values | Never for listing; continuation follows its explicit command semantics |
### Replayable Client facts are projection-owned
A Client-visible fact belongs to `SessionProjectionMap` when its value is determined by the Session header or event log and must survive reload, cold access, or reconnect. The rule covers title, list metadata, model selection, agent preset selection, subagent identity, and subagent timing. Their domain packages own pure projection definitions; the Session transport and Client value store remain domain-neutral.
The three projection delivery states have different meanings:
- A Session-list hint is optional, partial, and possibly stale. A missing key means unknown, so a list consumer must not invent an empty value or deployment default.
- A follow opening baseline is the complete set of client-visible projection capabilities registered at its cursor. A missing key there means the capability is absent for that Host composition.
- An explicit `null` is a domain-computed no-value result. It is distinct from a missing list hint and survives JSON transport.
These distinctions prevent one overloaded `undefined` from representing cache miss, unloaded plugin, and a real domain answer. API types name list data as hints and opening data as a baseline so a consumer cannot assume equivalent completeness merely because both carry projection values.
### Client merge rules
| Input | Completeness | Freshness | Meaning of missing key |
|---|---|---|---|
| Session list hints | Partial | Last durable checkpoint or bounded fallback cut | Unknown |
| Follow opening baseline | Complete for the Host composition | Exact opening cursor | Capability absent |
| Projection frame | One whole key | Event sequence carried by the frame | Not applicable |
The Client stores one row per key with its sequence number. A newer hint, baseline, or frame replaces a row; an equal or older input is ignored. Reconnect can therefore replace the event window without rolling back a projection frame that was already accepted at a later sequence.
The list view reads the same per-Session store as the opened Session. Hints can populate title, preset, and other list presentation before follow completes; the opening baseline then converges that state without creating a second summary-only authority.
The per-Session Client projection store accepts list hints, the follow baseline, and later whole-value frames under one higher-sequence-wins rule. It never folds Session events. A baseline or frame may advance a hinted value, while an older cut cannot overwrite a newer row.
Data that is not derived from one Session remains outside projections. `llm.models` owns the Host-generation model catalog, and `agentPreset.list` owns the configurable preset roster. A selector combines the relevant catalog with the Session's `modelSelection` or `agentPreset` projection only when both inputs are ready. During refresh it may retain the last complete catalog; before the first complete pair it reports loading instead of rendering a guessed name or availability verdict.
Client-local interaction state also remains local: loading and error status, an open menu, an in-flight selection, and a staged choice for a not-yet-created Session are not replayable Session facts. Once a choice applies to a Session, its durable event and projection become authoritative.
### Domain applications
- **Title and list metadata.** Cached projection hints may render an existing title and determine blankness or recency. Missing hints leave those facts unknown; only the bounded small-log policy may resolve them during listing.
- **Model selection.** `model/selection` records a complete provider, model, and optional reasoning effort. `modelSelection` distinguishes the last request's route from a later selection pending consumption by a request header.
- **Agent preset.** The projection initializes from immutable Session metadata and advances on preset-selection events. A missing or `null` value is not replaced with the deployment default for an existing Session.
- **Subagent identity.** The `subagent` unit remains the sole descriptor interpreter. Listing obtains candidates from the shared corpus and resolves values through live state, projection cache, or an observation rather than scanning events itself.
- **Subagent presentation.** Opening projection values establish timing and identity before the Client declares the child interactive or offline, so transport loading does not masquerade as a durable state.
These migrations remove special-case Client state without making projection own provider catalogs or interaction mechanics. A domain still owns mutations and commands; projection owns only their replayable Session result.
### Failure and readiness boundaries
- A list cache miss is not an error and does not hide the row. Unknown hints remain absent until a bounded fallback or exact opening supplies them.
- A projection failure during an exact cold observation makes that observation fail as corrupt Session data; callers do not publish a mixture of successful keys and failed keys.
- A subagent candidate's failed cold observation is isolated to that candidate's diagnostic row; sibling candidates remain usable.
- A catalog load failure is Client-visible catalog state. It does not erase a previously complete catalog during refresh and does not synthesize a Session selection.
- A follow carrier generation is not accepted until its opening snapshot is validated and applied. The previous generation remains visible during reconnect.
Cancellation stops queued or in-flight cold resolution at documented checkpoints and releases every acquired lease. Cancellation does not convert into not-found, nor may it leave a prepared entry pinned.
### Ownership matrix
| Concern | Owner | Non-owner |
|---|---|---|
| Cold materialization and revision checks | Session persistence | API Controller and Client |
| Exact live-preferred read cut | SessionQuery observation | Individual endpoint helpers |
| Fold state and client-value computation | Projection registry and domain unit | SessionQuery and Client |
| Partial list acceleration | Projection cache and list policy | Follow protocol |
| Opening and reconnect replacement | Session follow and journal stream | Session page |
| Per-key value ordering | Client projection store | Domain UI components |
| Provider or preset catalog lifecycle | Its catalog directory | Session projection |
| Rendering and transient interaction state | Domain UI package | Host projection units |
### Extension rules
1. Determine whether a new value is a replayable fact of one Session. If it is, define or reuse its durable header/event input before adding a Client field.
2. Register one pure projection unit in the owning domain. Keep fold state and Client view types distinct when their representations differ.
3. Let exact readers request `projectionMode: 'all'`; filter only when constructing an audience-specific response.
4. Let list consumers accept an optional hint. Do not force full corpus hydration merely to avoid an explicit unknown state.
5. Feed the generic Client projection store. Do not add a dedicated reconnect fetch, event reducer, or Session summary mirror for the same fact.
6. Keep non-Session catalogs and ephemeral UI state in their own owners, and define readiness before combining them with a projection value.
These rules apply to new Session-derived Client state even when a direct event scan appears cheap. Complexity is measured across cold reads, reconnect, multiple tabs, plugin lifetime, and future consumers rather than at the first call site.
### Relationship to existing decisions
- [Reusable Session preparation](2026-08-05-session-preparation.md) owns cold materialization, repair, reservation, and publication. Observation adds a shared read lease over that prepared object; it does not move preparation into SessionQuery.
- [Session history and Remote event transport](2026-08-18-session-history-and-event-transport.md) owns stream generations and replacement semantics. This decision supplies the exact snapshot that opens each journal generation.
- [Projection state and Client views](2026-08-19-session-projection-state-and-client-views.md) owns the distinction between Host fold state and Client values. This decision governs where those values are consumed and how partial list hints differ from a complete baseline.
- [Subagent identity projection](2026-08-06-subagent-list-identity-projection.md) continues to own descriptor folding, the serializable `null` sentinel, and the own-suffix sequence check. This decision supersedes only its independent corpus merge and direct cold-inspection path: listing now uses SessionQuery's corpus and observation.
- The broader [session projection and command-log proposal](../../proposed/architecture/2026-07-27-session-projection-and-command-log.md) remains proposed for the portions not represented by shipped code. This decision records the shipped observation and Client-ownership subset.
## Verification
Persistence and SessionQuery tests pin shared cold loading, cancellation, live-source races, retained observations, disposal, and all-or-none projection calculation. Session Controller and Gateway tests pin snapshot-first opening, replacement reconnect, older-page reads, gap repair, list-cache hints, bounded small-log fallback, and promotion after snapshot delivery.
Client tests pin higher-sequence-wins projection storage, title updates, model catalog and selection readiness, preset roster refresh and Session-specific selection, and subagent loading without transient offline presentation. Subagent tests pin corpus enumeration, cache and observation fallback, lifecycle witnesses, bounded cold reads, and no Agent activation during listing.
## Alternatives considered
**Keep source resolution in each consumer.** Rejected because every caller would continue to implement its own live race, persistence error mapping, preparation lifetime, cancellation, and projection cut, allowing both duplicate work and inconsistent results.
**Activate an Agent for every exact read.** Rejected because list, history, attachment, search, and subagent inspection are read operations. Activation loads plugins and changes process state, and it has no natural retirement point for pagination or catalog reads.
**Prepare only requested projection keys.** Rejected because a partially projected Session creates another lifecycle state that every cache, restore, plugin-registration, and caller path must track. Projection units are pure and few; computing all registered units for an exact observation is simpler than maintaining `O(E*k)` partial state instead of `O(E*P)` complete state.
**Cache each projection's viewed value separately.** Rejected because `view` is a pure synchronous conversion over already folded state. A second `viewReady`/`viewedState`/`viewedValue` cache adds invalidation and plugin-lifetime states without avoiding event folding.
**Keep dedicated summary fields, RPCs, or Client reducers.** Rejected because each creates a second authority beside the event log and projection registry. It also requires separate baseline, reconnect, and race handling for every domain value.
**Require complete projections on every list row.** Rejected because listing a large cold corpus would force full-log work before navigation can render. Partial cache hints preserve a cheap list path; consumers already have an explicit unknown state until opening supplies the exact baseline.
**Render guessed defaults while catalog or projection input is missing.** Rejected because the guess can visibly disagree with the Session and then change after loading. Initial uncertainty renders as loading; refresh retains the last complete value until the replacement is ready.
## Consequences
Session consumers share one live-preferred read model and one prepared cold object. Header, events, cursor, and projections belong to the same observation, and ordinary page opening can reuse that object for later promotion. New point-read consumers use SessionQuery instead of composing persistence and registry calls themselves.
Session-derived Client state has one extension path: record or identify the durable input, register a pure projection unit, and consume its finished value through the generic store. Domain-specific catalogs may remain separate when they are not Session-derived, but they cannot substitute a default for an unknown Session projection.
The simpler state model accepts bounded extra computation. An exact projected cold observation evaluates every registered unit, and a small cache-missing list artifact may be read in full. Large list rows can remain partially described until opened, so every list consumer must preserve the distinction among unknown, absent capability, and explicit no value. Observation leases also make disposal part of the caller contract; retaining a prepared source without releasing it prevents normal cache retirement.

View file

@ -0,0 +1,201 @@
# Agent Note: Session observation 与 projection 所有的客户端状态
Status: implemented
[English](2026-08-25-session-observations-and-projection-owned-client-state.md) | 中文
## 问题
面向 Session 的多个消费方需要相同的逻辑数据,却各自完成解析。list、follow、page、附件/fork 读取和 subagent 检查分别在已挂载 Session、持久化元数据、prepared Session 与 projection cache 之间作选择。因此一次页面访问可能多次物化同一份冷日志,各自拼装的 header、事件、cursor 和 projection 值也可能来自不同的读取切面。
客户端功能还以多种形式保存 Session 派生事实。title 有专门的 list 和更新逻辑;模型选择把 Session 专用 catalog 请求与本地状态混在一起;agent preset 展示可能在当前 Session 到达前猜测全局默认值;subagent list 则单独扫描或重建 identity。这些镜像产生了中间状态:即使持久化 Session 已经决定答案,UI 仍会短暂显示猜测出的默认值、原始 id 或不可用状态。
仅统一持久化读取,Client 镜像仍会成为互相竞争的真源。仅统一 Client 字段,各 Host 入口仍可能获得不同的数据切面。因此读取单元与派生状态单元需要一条配套的 ownership 规则。
## 决策
Session 精确读取使用可保留的 `SessionObservation`,向 Client 暴露的可回放 Session 派生值使用已注册 projection。Observation 负责选择数据源和提供一份不可变读取切面;projection 负责从该切面派生状态。API 层只选择要发布的内容,Client 只消费成品值,不从事件重建 Session 事实,也不在各领域镜像中重复保存这些事实。
### 数据动线
两条 ownership 规则在 observation 的 projection snapshot 处汇合。轻量 list 可以止于 cache hints;每次精确 opening 都进入同一 observation 路径,并向 Client 提供完整 replacement baseline。
```mermaid
flowchart LR
List["list / search"] --> Corpus["SessionQuery corpus"]
Follow["follow"] --> Observe["observeSession"]
Page["page / attachment / fork"] --> Observe
Subagent["subagent list / continuation"] --> Corpus
Subagent --> Observe
Corpus --> Cache["projection cache hints"]
Cache --> ClientList["Client Session list"]
Cache -->|"small miss"| Observe
Observe --> Source{"live or cold"}
Source --> Live["attached Session cut"]
Source --> Borrow["borrowSession"]
Borrow --> Prepared["SessionPreparations.borrow"]
Live --> Mode{"all or none"}
Prepared --> Mode
Mode --> Snapshot["SessionObservation"]
Snapshot --> Opening["follow opening snapshot"]
Snapshot --> Read["page / inspection"]
Opening --> Store["Client projection store"]
Store --> Domain["title / model / preset / subagent"]
```
### Observation 是 point read 单元
`SessionQueryEngine.observeSession(sessionId, options)` 返回可 dispose(资源释放)的 `SessionObservation`,其中包含同一份 source kind、header、连续事件前缀、cursor、可选 projection snapshot,以及 prepared source 的持久化 revision。已挂载 Session 优先;否则 `SessionPersistence.borrowSession()` 与 `SessionPreparations.borrow()` 共享并固定一份 prepared Session,包括尚未完成的冷加载。
每个 owner 都会 dispose 自己的 observation。`retain()` 为同一切面创建另一份 lease,使 `session.follow` 能够先发布 snapshot,再把完全相同的 prepared source 转交给后台 Agent promotion,而无需重读日志。冷解析期间出现的 live Session 会在发布前胜出;已经消失的 live source 会按 cold source 重试。
### 数据源解析与生命周期
一份 observation 把所有返回字段绑定到同一 lifecycle witness。调用方不会把 corpus list 的 header、persistence 的 events 和稍后 live Session 的 projections 拼在一起。选中的 header 与事件前缀共同产生 cursor 和 projection snapshot。
系统在 cold borrow 前后都检查 live 优先级。第二次检查封住 persistence 加载期间 Agent 完成 attach 的竞态。如果 persistence 报告由 live source 胜出,但 SessionQuery 检查时该 source 已经 detach,解析会重新开始,而不是发布一份无人持有的引用。
只有在不存在已挂载 Session 后,persistence absence 才映射为 Session-not-found。持久数据损坏、source identity 冲突、取消和 persistence 操作失败分别保留不同的 `SessionQueryError`,API owner 因而可以维持自身公开错误词汇,而不用重复数据源判定。
Observation 不拥有任何 mutation 权限。其事件数组是不可变前缀,prepared Session 保持未发布。Promotion 是 Session Controller 在 opening snapshot 发出后执行的显式 ownership transfer;其他读方不能把 observation 变成 live Agent。
Projection 工作明确只有 `all | none` 两种模式。`all` 在 observation 的事件 cursor 上计算所有已注册 projection;`none` 完全不触碰 projection 状态。系统不存在按 key preparation 的状态、`projectionKeys` 模式或额外的 `viewedState`/`viewedValue` cache。发布方可以按 audience 筛选已完成的值,但底层 observation 不会处于只算完部分 projection 的状态。
### Projection 执行边界
对于 live source,`all` 读取一份同步 registry snapshot。对于 prepared source,projection cache 可以播种有效 state row,随后每个已注册 unit 在精确的剩余事件前缀上推进。得到的 Client value 共用一个 `asOfSeq`。
筛选发生在计算完成之后,因为它改变的是披露内容,而不是状态。Page 鉴权可以只消费 `subagent`,list row 可以只发布 list 相关值;只要它们请求 projection 工作,仍然依赖一份完整 projected cut。
Registry 拥有 fold state;各领域拥有自己的 `init`、`apply`、`view`、schema 和 `stateVersion`。SessionQuery 只知道是否需要 projection 工作,不理解 title、model、preset、subagent、token、image、plan、todo 或 goal 的语义。
`view` 保持为 folded state 上无 cache 的同步转换。其成本由已注册 projection unit 数量界定,并在 snapshot 发布时支付;引入第二层 cache 只会增加 invalidation 状态,无法减少 event replay。
语料库 list 仍是独立的轻量操作。`listSessions()` 返回 live-preferred header,而不物化每份日志。Session list 与 subagent list 先读取 live projection 状态或持久 projection-cache row。当 cache 无法判断 Session 是否为空,且该 Session 拥有的独立产物未超过配置的小日志限制时,Session list 可以执行一次完整 observation;大型或不可读的 cache miss 仍以 hints 未知但 row 可见的方式返回。
`session.follow` 发布必需的 opening snapshot,其中包含 header、cursor、首个事件窗口和完整 projection baseline。重连使用另一份完整 snapshot 替换上一 generation。`session.page` 仅用于旧历史读取与 gap repair。只读 observation 不激活 Agent;只有普通 follow 可以保留 prepared observation,并在 opening snapshot 已交付后请求 promotion。
### 读取 audience
每个公开操作选择一组查询与 projection 策略。该选择属于操作本身的行为,而不是 persistence 或 transport 内部的启发式判断。
| 操作 | 读取路径 | Projection 策略 | Agent 激活 |
|---|---|---|---|
| `session.list` | Corpus header、live state 和 cached row;有界小日志 fallback | 部分 hints,或一次完整小日志 observation | 从不 |
| `session.search` | Corpus 鉴权加已配置 search provider | 结果列表不计算 | 从不 |
| `session.follow` | 一份精确 observation | 全算,并由 opening snapshot 携带 | 仅普通 cold Session,且在 snapshot 交付后 |
| `session.page` | 一份精确 observation | 不计算,但 projection-backed subagent 鉴权除外 | 从不 |
| Attachment 与 fork source | 一份精确 observation | 鉴权不要求时不计算 | source 从不激活 |
| Subagent list 与 continuation | Corpus 加 live/cache/observation 解析 | cold fallback 全算;audience 只消费 identity 或继承值 | Listing 从不;continuation 遵循显式命令语义 |
### 可回放的 Client 事实归 projection 所有
当一个 Client 可见值由 Session header 或事件日志决定,并且必须在刷新、冷访问或重连后恢复时,它属于 `SessionProjectionMap`。这条规则覆盖 title、list metadata、model selection、agent preset selection、subagent identity 和 subagent timing。各领域包拥有纯 projection definition;Session transport 与 Client value store 不理解具体领域。
Projection 的三种交付状态含义不同:
- Session-list hint 是可选、部分且可能陈旧的数据。key 缺失表示未知,因此 list 消费方不得自行补成空值或部署默认值。
- Follow opening baseline 是其 cursor 上所有已注册 Client 可见 projection capability 的完整集合。此处缺少 key 表示当前 Host composition 不具备该 capability。
- 显式 `null` 是领域计算出的无值结果。它不同于 list hint 缺失,并且能够完整通过 JSON transport。
这些区别避免由一个重载的 `undefined` 同时表示 cache miss、plugin 未加载和真实领域答案。API 类型把 list 数据命名为 hints,把 opening 数据命名为 baseline,因此消费方不能只因两者都携带 projection value 就假定其完整性相同。
### Client 合并规则
| 输入 | 完整性 | 新鲜度 | key 缺失的含义 |
|---|---|---|---|
| Session list hints | 部分 | 上次持久 checkpoint 或有界 fallback cut | 未知 |
| Follow opening baseline | 对当前 Host composition 完整 | 精确 opening cursor | Capability 不存在 |
| Projection frame | 单个完整 key | Frame 携带的 event sequence | 不适用 |
Client 为每个 key 保存带 sequence number 的一行。更新的 hint、baseline 或 frame 会替换 row;相同或更旧的输入被忽略。因此 reconnect 可以替换 event window,而不会回退已经在更晚 sequence 接受的 projection frame。
List view 与已打开 Session 读取同一个 per-Session store。Hints 可以在 follow 完成前填充 title、preset 和其他 list presentation;opening baseline 随后收敛这份状态,而不会建立第二套 summary-only authority。
每个 Session 的 Client projection store 按一条 higher-sequence-wins 规则接收 list hints、follow baseline 和后续 whole-value frame。它从不折叠 Session event。Baseline 或 frame 可以推进 hinted value,较旧切面不能覆盖较新的 row。
不由单个 Session 派生的数据不进入 projection。`llm.models` 拥有当前 Host generation 的 model catalog,`agentPreset.list` 拥有可配置 preset roster。Selector 只在相应 catalog 与 Session 的 `modelSelection` 或 `agentPreset` projection 均就绪后组合两者。刷新时可以保留上一份完整 catalog;第一次获得完整输入前显示 loading,而不是展示猜测的名称或可用性结论。
Client 本地交互状态也继续留在本地:loading 和 error 状态、打开的菜单、进行中的选择,以及为尚未创建 Session 暂存的选择都不是可回放 Session 事实。选择一旦应用到 Session,其持久事件与 projection 就成为权威。
### 领域应用
- **Title 与 list metadata。** Cached projection hints 可以渲染已有 title,并判断 blankness 或 recency。Hints 缺失时这些事实保持未知;listing 期间只有有界小日志策略可以解析它们。
- **Model selection。** `model/selection` 记录完整 provider、model 和可选 reasoning effort。`modelSelection` 区分上一请求使用的 route,以及等待 request header 消费的较晚 selection。
- **Agent preset。** Projection 从不可变 Session metadata 初始化,并随 preset-selection event 推进。对于现有 Session,缺失或 `null` 值不会替换成部署默认值。
- **Subagent identity。** `subagent` unit 仍是唯一 descriptor interpreter。Listing 从共享 corpus 获得 candidate,并通过 live state、projection cache 或 observation 解析值,不自行扫描 event。
- **Subagent presentation。** Opening projection value 在 Client 宣布 child 可交互或离线前建立 timing 与 identity,因此 transport loading 不会伪装成 durable state。
这些迁移删除特殊 Client state,但不会让 projection 接管 provider catalog 或交互机制。领域仍拥有 mutation 和 command;projection 只拥有其可回放 Session 结果。
### 失败与 readiness 边界
- List cache miss 不是错误,也不会隐藏 row。未知 hints 保持缺失,直到有界 fallback 或精确 opening 提供值。
- 精确 cold observation 中的 projection failure 使整份 observation 按损坏的 Session data 失败;调用方不会发布成功 key 与失败 key 的混合结果。
- 一个 subagent candidate 的 cold observation 失败只影响该 candidate 的 diagnostic row;sibling candidate 继续可用。
- Catalog load failure 是 Client 可见的 catalog state。它不会在 refresh 期间清除上一份完整 catalog,也不会合成 Session selection。
- Follow carrier generation 只有在 opening snapshot 完成校验与应用后才被接受。Reconnect 期间继续显示上一 generation。
取消会在文档规定的检查点终止排队中或进行中的 cold resolution,并释放每一份已获得 lease。取消不会变成 not-found,也不能让 prepared entry 保持 pinned。
### Ownership 矩阵
| 事项 | Owner | 非 owner |
|---|---|---|
| Cold materialization 与 revision 检查 | Session persistence | API Controller 与 Client |
| 精确 live-preferred read cut | SessionQuery observation | 各 endpoint helper |
| Fold state 与 Client-value 计算 | Projection registry 与 domain unit | SessionQuery 与 Client |
| 部分 list acceleration | Projection cache 与 list policy | Follow protocol |
| Opening 与 reconnect replacement | Session follow 与 journal stream | Session page |
| Per-key value ordering | Client projection store | Domain UI component |
| Provider 或 preset catalog lifecycle | 对应 catalog directory | Session projection |
| Rendering 与瞬时 interaction state | Domain UI package | Host projection unit |
### 扩展规则
1. 判断新值是否属于单个 Session 的可回放事实;如果属于,先定义或复用其持久 header/event 输入,再添加 Client 字段。
2. 在 owning domain 注册一个 pure projection unit。Fold state 与 Client view 表示不同时,分别定义其类型。
3. 让精确读方请求 `projectionMode: 'all'`;仅在构造 audience-specific response 时筛选。
4. 让 list 消费方接受 optional hint。不能只为消除显式 unknown state 而强制 hydrate 整个 corpus。
5. 把值送入通用 Client projection store。不能为同一事实再增加 dedicated reconnect fetch、event reducer 或 Session summary mirror。
6. 非 Session catalog 与 ephemeral UI state 保留各自 owner,并在和 projection value 组合前定义 readiness。
这些规则适用于新的 Session-derived Client state,即使在第一个调用点直接扫描 event 看似廉价。复杂度需要覆盖 cold read、reconnect、多 tab、plugin lifetime 和未来消费方,而不是只看首次实现。
### 与既有决策的关系
- [可复用 Session preparation](2026-08-05-session-preparation.zh.md)拥有冷物化、修复、reservation 和发布。Observation 在该 prepared object 之上增加共享读取 lease,并未把 preparation 移入 SessionQuery。
- [Session 历史与 Remote event transport](2026-08-18-session-history-and-event-transport.zh.md)拥有 stream generation 与 replacement 语义。本决策提供每个日志 generation 的精确 opening snapshot。
- [Projection state 与 Client view](2026-08-19-session-projection-state-and-client-views.zh.md)拥有 Host fold state 和 Client value 的区分。本决策规定这些值在哪里消费,以及部分 list hints 与完整 baseline 的差别。
- [Subagent identity projection](2026-08-06-subagent-list-identity-projection.zh.md)继续拥有 descriptor folding、可序列化 `null` sentinel 和 own-suffix sequence 检查。本决策只取代其中独立 corpus merge 和直接 cold inspection 路径:listing 改为使用 SessionQuery corpus 和 observation。
- 更广泛的 [session projection 与 command-log 提案](../../proposed/architecture/2026-07-27-session-projection-and-command-log.zh.md)仍为 proposed,其中尚未由已交付代码体现的部分不受影响。本决策记录已经交付的 observation 与 Client ownership 子集。
## 验证
Persistence 与 SessionQuery 测试固定共享冷加载、取消、live-source race、retained observation、dispose 和 all-or-none projection 计算。Session Controller 与 Gateway 测试固定 snapshot-first opening、replacement reconnect、旧分页读取、gap repair、list-cache hints、小日志有界 fallback,以及 snapshot 交付后的 promotion。
Client 测试固定 higher-sequence-wins projection store、title 更新、model catalog 与 selection readiness、preset roster refresh 与 Session 专属选择,以及不会短暂展示离线状态的 subagent loading。Subagent 测试固定 corpus 枚举、cache 与 observation fallback、lifecycle witness、有界冷读,以及 listing 期间不激活 Agent。
## 考虑过的替代方案
**由各消费方继续解析数据源。** 否决,因为每个调用方都需要重复实现 live race、persistence error mapping、preparation lifetime、cancellation 和 projection cut,既会重复工作,也会产生不一致结果。
**每次精确读取都激活 Agent。** 否决,因为 list、history、attachment、search 与 subagent inspection 都是读取操作。Activation 会加载插件并改变进程状态,也没有适合分页或 catalog 读取的自然退出点。
**只 prepare 被请求的 projection key。** 否决,因为只完成部分 projection 的 Session 会增加一种生命周期状态,所有 cache、restore、plugin registration 和调用路径都必须追踪它。Projection unit 数量少且为纯函数;精确 observation 计算全部已注册 unit,比为了 `O(E*k)` 而维护部分状态、取代 `O(E*P)` 完整状态更简单。
**单独缓存每个 projection 的 viewed value。** 否决,因为 `view` 只是已折叠 state 上的纯同步转换。第二层 `viewReady`/`viewedState`/`viewedValue` cache 会增加 invalidation 和 plugin lifetime 状态,却不能减少 event folding。
**保留专用 summary 字段、RPC 或 Client reducer。** 否决,因为每一项都会在 event log 与 projection registry 之外建立第二个真源,还要求每个领域分别实现 baseline、reconnect 和 race handling。
**要求每个 list row 都携带完整 projection。** 否决,因为列出大型 cold corpus 时必须先读取完整日志,navigation 才能渲染。部分 cache hints 保留了轻量 list 路径;在 opening 给出精确 baseline 前,消费方已经拥有明确的 unknown 状态。
**在 catalog 或 projection 输入缺失时渲染猜测默认值。** 否决,因为猜测可能明显违背 Session,并在加载后发生跳变。初次不确定时显示 loading;刷新时保留上一份完整值,直到替代值就绪。
## 后果
Session 消费方共享一份 live-preferred read model 和一个 prepared cold object。Header、events、cursor 与 projections 属于同一 observation,普通页面打开还可以为后续 promotion 复用该对象。新的 point-read 消费方使用 SessionQuery,而不再自行拼接 persistence 与 registry 调用。
Session 派生 Client 状态只有一条扩展路径:记录或识别持久输入、注册纯 projection unit,再通过通用 store 消费其成品值。不由 Session 派生的领域 catalog 可以独立存在,但不能用默认值替代未知的 Session projection。
更简单的状态模型接受有界的额外计算。精确的 projected cold observation 会计算所有已注册 unit,小型且 cache miss 的 list artifact 可能完整读取。大型 list row 在打开前可以保持部分描述,因此每个 list 消费方必须保留 unknown、capability absent 和 explicit no value 的区别。Observation lease 还使 dispose 成为调用方约定的一部分;保留 prepared source 而不释放会阻止正常 cache retirement。

View file

@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-25-sparse-first-party-prompt-section-orders.md
2026-08-25-sparse-first-party-prompt-section-orders.md: 2bf2e7441b449a6cfbd5b845f1d7e97b3fab09ae
2026-08-25-sparse-first-party-prompt-section-orders.zh.md: 624ed51f4d40848c72091097900ef05ec35fdd2e

View file

@ -0,0 +1,57 @@
# Agent Note: Centralize sparse first-party prompt-section orders
Status: implemented
English | [中文](2026-08-25-sparse-first-party-prompt-section-orders.zh.md)
## Problem
Repository-owned system-prompt sections declared unrelated numeric literals across more than twenty packages. The main tool sequence occupied consecutive values from 100 through 117 and then used half-step values for insertions. A later change could therefore collide with an existing section without seeing the complete allocation.
Equal orders used stable JavaScript sort behavior, which made plugin activation order the effective tie-breaker. The [Cordis/workflow prompt-order fix](../../archived/bug-fix/2026-08-24-system-prompt-section-order-ties.md) showed that clean compositions can activate the same plugins in different orders and produce different request headers and snapshot results. Fixing one collision locally did not prevent another package from reusing that value.
The shell guidance also followed filesystem guidance even though shell commands have the broadest execution and failure semantics. A model should read the shell result obligation before the narrower instructions that route file work to dedicated tools.
## Decision
`@deepseek-ai/dsh-system-prompt` exports `FIRST_PARTY_SECTION_ORDER` as the single allocation for repository-owned sections. Every first-party contributor imports its named placement instead of declaring a numeric literal. Values are unique integers, and adjacent allocated values differ by at least ten.
The allocation preserves the established first-party sequence except for two deliberate changes: Bash, or PowerShell in the Windows composition, leads per-tool guidance; and sections that shared an order receive an explicit sequence. The groups are:
| Group | Entries |
|---|---|
| Product opening | `harness:identity` −1000, `harness:source` −900, `app:web-surface` −800, `deployment:persona` 0 |
| Work modes | `plan:policy` 500, `team:policy` 600 |
| Invocation prelude | `tools:code-only` 800, `context:file-reference` 900 |
| Local tools | `tool:bash` 1000, `tool:pwsh` 1010, `tool:read` 1100, `tool:write` 1200, `tool:edit` 1300, `tool:glob` 1400, `tool:grep` 1500, `tool:jobs` 1600, `tool:pty` 1700 |
| Higher-level tools | `tool:web_search` 2000, `tool:web_fetch` 2100, `tool:lsp` 2200, `tool:session-query` 2300, `tool:goal` 2400, `tool:cordis` 2500, `tool:workflow` 2600, `tool:ralph` 2700, continuable-subagent guidance 2800, `tool:report` 2900 |
| Generated protocol | `tools:sdk` 5000 |
| Final-output obligations | deliverable file references 9000, `tool:structured_output` 9900 |
`SystemPrompt.assemble()` sorts equal-order sections by code-unit section name after comparing `order`. This makes third-party collisions deterministic without locale-sensitive comparison. First-party contributors still receive distinct ranks so their intended sequence remains explicit rather than depending on the fallback.
Dynamic `PromptContext` order and tool-schema `toolOrder` are separate sequences and remain unchanged. A scoped `deployment:persona` continues to shadow the global section by name before section sorting, so it shares `PERSONA_ORDER` rather than consuming another placement.
## Verification
The system-prompt unit suite verifies that every exported first-party value is an integer, every value is unique, adjacent values differ by at least ten, and opposite registration permutations produce the same code-unit name order for a tie. Real-composition snapshots pin the model-visible ordering change, including Bash before filesystem guidance and the explicit Cordis, workflow, Ralph, subagent, and report sequence.
## Alternatives considered
**Keep package-local numeric literals and review collisions manually.** Rejected because a contributor cannot see the complete allocation locally, and the collision that motivated the earlier fix recurred after that fix merged.
**Continue inserting fractional values.** Rejected because fractions provide no durable spacing rule, obscure the semantic groups, and still permit unrelated packages to choose the same value.
**Normalize only snapshot comparisons.** Rejected because the runtime request header and model prompt would remain activation-order dependent while the test hid the difference.
**Preserve activation order for equal ranks.** Rejected because activation order is not a prompt-order decision and varies across valid compositions. Name order is deterministic for external collisions; explicit named placements carry first-party intent.
**Renumber dynamic contexts and tool schemas in the same allocation.** Rejected because they are independently assembled sequences. Combining them would imply cross-sequence ordering that the runtime does not perform.
## Consequences
Numeric ranks are not rendered, so the renumbering alone does not change model text. Bash or PowerShell moves before other per-tool guidance, and previously tied sections acquire deterministic order; those model-visible changes update request-header snapshots and may invalidate provider prefix reuse from the first moved paragraph.
An external plugin that chose a raw number specifically to sit between old first-party values may move relative to repository sections. This repository is pre-release and provides no compatibility shim for the old allocation; extensions can select positions from the exported current allocation. Equal external ranks remain supported and deterministic by name.
The system-prompt package now knows the names and relative placement of repository features. That centralized coupling is deliberate: the registry already owns the ordering semantics, while distributed numeric literals made the same relationship implicit and uncheckable.

View file

@ -0,0 +1,57 @@
# Agent Note: 集中管理稀疏的 first-party 提示词段顺序
Status: implemented
[English](2026-08-25-sparse-first-party-prompt-section-orders.md) | 中文
## 问题
仓库自带的系统提示词段分散在二十多个包中,各自声明互不关联的数字字面量。主要工具序列连续占用 100 到 117,后续插入还使用半步数值。因此,后续更改可能在无法看到完整分配表的情况下与已有段发生冲突。
相同 order 依赖 JavaScript 稳定排序,使插件激活顺序成为实际的平局规则。[Cordis/workflow 提示词顺序修复](../../archived/bug-fix/2026-08-24-system-prompt-section-order-ties.md)表明,完整且有效的组合可能按不同顺序激活同一组插件,进而产生不同的请求 header 和快照结果。局部修复一次冲突,无法阻止另一个包再次使用同一数值。
此外,shell 指导位于文件系统指导之后,但 shell 命令具有最广泛的执行和失败语义。模型应先读到 shell 结果义务,再阅读将文件操作分流到专用工具的更窄指令。
## 决策
`@deepseek-ai/dsh-system-prompt` 导出 `FIRST_PARTY_SECTION_ORDER`,作为仓库自带提示词段的唯一分配表。每个 first-party 贡献方都导入具名位置,不再声明数字字面量。所有值都是互不相同的整数,相邻已分配值之差至少为十。
除两项有意调整外,该分配保留既有 first-party 顺序:Bash,或 Windows 组合中的 PowerShell,位于逐工具指导的首位;原先共享 order 的段获得明确顺序。分组如下:
| 分组 | 条目 |
|---|---|
| 产品开场 | `harness:identity` −1000、`harness:source` −900、`app:web-surface` −800、`deployment:persona` 0 |
| 工作模式 | `plan:policy` 500、`team:policy` 600 |
| 调用前置说明 | `tools:code-only` 800、`context:file-reference` 900 |
| 本地工具 | `tool:bash` 1000、`tool:pwsh` 1010、`tool:read` 1100、`tool:write` 1200、`tool:edit` 1300、`tool:glob` 1400、`tool:grep` 1500、`tool:jobs` 1600、`tool:pty` 1700 |
| 高层工具 | `tool:web_search` 2000、`tool:web_fetch` 2100、`tool:lsp` 2200、`tool:session-query` 2300、`tool:goal` 2400、`tool:cordis` 2500、`tool:workflow` 2600、`tool:ralph` 2700、可继续运行的 subagent 指导 2800、`tool:report` 2900 |
| 生成协议 | `tools:sdk` 5000 |
| 最终输出义务 | 可交付文件引用 9000、`tool:structured_output` 9900 |
`SystemPrompt.assemble()` 比较 `order` 后,按提示词段名称的代码单元顺序排列同号项。这样无需使用受区域设置影响的比较,也能让第三方冲突产生确定结果。first-party 贡献方仍使用不同 rank,其预期顺序由分配表明确表达,而不依赖兜底规则。
动态 `PromptContext` 顺序和工具 schema 的 `toolOrder` 是独立序列,保持不变。带作用域的 `deployment:persona` 仍会在段排序之前按名称遮蔽全局段,因此共享 `PERSONA_ORDER`,而不占用另一个位置。
## 验证
系统提示词单元测试验证:导出的每个 first-party 值都是整数、所有值互不重复、相邻值之差至少为十,并且顺序相反的两种注册排列会对同号项产生相同的代码单元名称顺序。真实组合快照固定面向模型的顺序变化,包括 Bash 位于文件系统指导之前,以及 Cordis、workflow、Ralph、subagent 和 report 的明确序列。
## 考虑过的替代方案
**保留包内数字字面量并通过评审人工检查冲突。**未采用,因为贡献方无法在局部看到完整分配表,而且早期修复合入后,触发该修复的同类冲突再次出现。
**继续插入小数值。**未采用,因为小数没有持久的间距规则,难以表达语义分组,也无法阻止无关包选择同一数值。
**只规范化快照比较。**未采用,因为运行时请求 header 和模型提示词仍依赖激活顺序,测试只会隐藏差异。
**同 rank 时保留激活顺序。**未采用,因为激活顺序不是提示词顺序决策,并且会在有效组合之间变化。名称顺序为外部冲突提供确定结果;具名位置负责表达 first-party 意图。
**在同一分配表中重新编号动态上下文和工具 schema。**未采用,因为运行时独立组装这些序列。合并分配会暗示运行时并不执行的跨序列顺序。
## 后果
数字 rank 不会被渲染,因此单纯重新编号不会改变模型文本。Bash 或 PowerShell 会移到其他逐工具指导之前,原先同号的段会获得确定顺序;这些面向模型的变化会更新请求 header 快照,并可能从第一个移动的段落起使提供方前缀复用失效。
如果外部插件专门选择一个原始数字以插入旧 first-party 数值之间,它相对仓库段的位置可能改变。本仓库处于预发布阶段,不为旧分配提供兼容层;扩展可以根据当前导出的分配表选择位置。外部段仍可使用相同 rank,并会按名称获得确定顺序。
系统提示词包现在了解仓库功能的名称和相对位置。这种集中耦合是有意的:注册表本就拥有排序语义,而分散的数字字面量只是让同一关系变得隐式且无法检查。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-07-24-empty-model-response-is-retryable.md
2026-07-24-empty-model-response-is-retryable.md: 021c368a25ce8dd277ba431fa905c0621b752fd0
2026-07-24-empty-model-response-is-retryable.zh.md: 67809931f959086a4525cec4c879af190e653d4a
2026-07-24-empty-model-response-is-retryable.md: 62e45385fbcf1cdebd1b2c19abcbfe6888abdfe7
2026-07-24-empty-model-response-is-retryable.zh.md: 459a5def0bbd5a82c83af67ea1c035259ec4f0a5

View file

@ -33,4 +33,4 @@ The classification uses the existing loop machinery — `finishError` → `agent
- A transiently misbehaving provider consumes a bounded retry instead of a turn with no output; a persistently empty model produces an actionable `EMPTY_RESPONSE` turn failure.
- A model that genuinely intends to say nothing (rare, but possible after a tool result) is retried and, if consistently empty, fails the turn. This trade was accepted deliberately: an empty assistant message is indistinguishable from the provider defect and has no value to the user.
- The `empty-response-retry` ACP snapshot (an authored keyless scenario with a deterministic 1 ms zero-jitter retry overlay, `examples/acp-agent/retry.cordis.yml`) pins the product-visible behavior: a durable `llm/retry` event, no ACP output for the discarded attempt, the recovered reply, and a clean completed turn.
- The [`empty-response-retry` headless snapshot](../../../../snapshots/session/empty-response-retry/) is an authored keyless scenario with a deterministic 1 ms zero-jitter retry overlay. It pins a durable `llm/retry` event, no output for the discarded attempt, the recovered reply, and a clean completed turn.

View file

@ -33,4 +33,4 @@ Status: implemented
- 一个偶发异常的提供方会消耗一次有界重试,而不是一个没有输出的轮次;一个持续返回空内容的模型则会产生用户可据以行动的 `EMPTY_RESPONSE` 轮次失败。
- 一个确实打算什么都不说的模型(罕见,但在一次工具结果之后有可能出现)会被重试,若始终为空,则该轮次失败。这个取舍是经过审慎权衡后接受的:一条空的 assistant 消息与提供方缺陷无法区分,且对用户毫无价值。
- `empty-response-retry` ACP(Agent Client Protocol)快照(一个人工编写的无密钥场景,配有确定性的 1 ms 零抖动重试 overlay,`examples/acp-agent/retry.cordis.yml`)钉住了产品可见的行为:持久的 `llm/retry` 事件、被丢弃的尝试不产生任何 ACP 输出、恢复后的回复,以及一次正常完成的轮次。
- [`empty-response-retry` headless 快照](../../../../snapshots/session/empty-response-retry/)是配有确定性 1 ms 零抖动重试 overlay 的人工无密钥场景。它钉住持久的 `llm/retry` 事件、被丢弃的尝试不产生输出、恢复后的回复,以及一次正常完成的轮次。

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-07-30-source-checkout-workdir-distinction.md
2026-07-30-source-checkout-workdir-distinction.md: ba6d9dd12b55a54d4ae8d2e91ad83ac3c1dc47fd
2026-07-30-source-checkout-workdir-distinction.zh.md: 06b09f82f294a70795090a9aec1b2220bd6e7512
2026-07-30-source-checkout-workdir-distinction.md: 407d11bd2d0bb27bf1953f1a3a84848480e03328
2026-07-30-source-checkout-workdir-distinction.zh.md: 91251aa040226d72cf431101aa0d064ba107823e

View file

@ -14,7 +14,7 @@ A blanket statement that the checkout is not the working directory would also be
The section identifies the path as the “DeepSeek Harness implementation checkout.” It says that the checkout location and current working directory are separate values that may differ, forbids inferring the working directory from the checkout path, directs the model to use `pwd`, and limits the checkout's purpose to inspecting or extending DSH itself.
The path derivation, global `harness:source` ownership, and `-99` ordering remain unchanged. Describing the values as conceptually separate rather than always unequal keeps the instruction accurate in both ordinary project sessions and `dsh meta`.
The path derivation and global `harness:source` ownership remain unchanged. The section uses first-party order −900, immediately after `harness:identity`. Describing the values as conceptually separate rather than always unequal keeps the instruction accurate in both ordinary project sessions and `dsh meta`.
## Verification

View file

@ -14,7 +14,7 @@ Status: implemented
该提示词段将路径标识为「DeepSeek Harness implementation checkout」。它说明 checkout 位置与当前工作目录是两个可能不同的值,禁止从 checkout 路径推断工作目录,指示模型使用 `pwd`,并限定该 checkout 只用于检查或扩展 DSH 自身。
路径推导方式、全局 `harness:source` 所有权和 `-99` 顺序均保持不变。将两者描述为概念上独立、而不是始终不相等,使这条指令在普通项目会话和 `dsh meta` 中都准确。
路径推导方式与全局 `harness:source` 所有权保持不变。该段使用 first-party 顺序 −900,紧随 `harness:identity`。将两者描述为概念上独立、而不是始终不相等,使这条指令在普通项目会话和 `dsh meta` 中都准确。
## 验证

View file

@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-08-06-api-key-format-validation.md
2026-08-06-api-key-format-validation.md: e2dcb91cba703cf95994426c67c4203effad290a
2026-08-06-api-key-format-validation.zh.md: e74cb58ee0e0be9ec9ec30123e7eeffa3f3db657
2026-08-06-api-key-format-validation.md: 853ae9a4e2f21a6b2cfa92b5059181b2c2869623
2026-08-06-api-key-format-validation.zh.md: 923798e5264a29fe95e6ee7616fdefeabea766b4

View file

@ -98,4 +98,4 @@ The costliest way to get this wrong would have been to treat absence as invalidi
`packages/client/ui-settings-models/tests/` pins `apiKeyFailure` over the same table plus the paste-shape cases, and drives both cards: a blank field submits without writing a credential, a whitespace-only field fails on the field, an illegal or wrapped key blocks submit and the interrogation alike, a padded key is trimmed before `credentials.set` and before an interrogation, and a hand-declared route can be created with no key at all.
The user-visible terminal state is pinned where it is actually assembled: `examples/headless-agent/tests/headless.snapshot.ts` runs the one-shot app against a stored key no header can carry, over the same keyless composition its missing-credential sibling uses, and records that the turn ends on `INVALID_CREDENTIAL` with an actionable message carrying neither the key nor the word `ByteString`. A package test could not have shown that, and the web e2e covers only the browser half.
The user-visible terminal state is pinned where it is actually assembled: `apps/cli/tests/profiles/headless/tests/headless.expected.e2e.ts` runs the one-shot app against a stored key no header can carry, over the same keyless composition its missing-credential sibling uses, and records that the turn ends on `INVALID_CREDENTIAL` with an actionable message carrying neither the key nor the word `ByteString`. A package test could not have shown that, and the web e2e covers only the browser half.

View file

@ -98,4 +98,4 @@ Status: implemented
`packages/client/ui-settings-models/tests/` 以同一张表加上形状用例钉住 `apiKeyFailure`,并驱动两张卡片:留空的输入框可提交且不写入凭据、只含空白的输入框在字段上失败、非法或被包裹的 Key 同时拦截提交与探测、带首尾空白的 Key 在 `credentials.set` 与探测之前被 trim,以及手工声明的路由可以完全不带 Key 创建。
用户可见的终态则钉在它真正被组装的位置:`examples/headless-agent/tests/headless.snapshot.ts` 让 one-shot 应用在一个 HTTP 标头无法承载的已存密钥下运行,复用其 missing-credential 兄弟场景的同一套无密钥 composition,并记录该轮次以 `INVALID_CREDENTIAL` 结束、消息可操作且既不含密钥也不含 `ByteString` 字样。包级测试无法证明这一点,而 web e2e 只覆盖了浏览器那一半。
用户可见的终态则钉在它真正被组装的位置:`apps/cli/tests/profiles/headless/tests/headless.expected.e2e.ts` 让 one-shot 应用在一个 HTTP 标头无法承载的已存密钥下运行,复用其 missing-credential 兄弟场景的同一套无密钥 composition,并记录该轮次以 `INVALID_CREDENTIAL` 结束、消息可操作且既不含密钥也不含 `ByteString` 字样。包级测试无法证明这一点,而 web e2e 只覆盖了浏览器那一半。

Some files were not shown because too many files have changed in this diff Show more