fix(skill): bound catalog discovery retries

.agents/notes/implemented/feature/2026-07-27-skill-catalog-hot-refresh.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-07-27-skill-catalog-hot-refresh.md
-2026-07-27-skill-catalog-hot-refresh.md: 7c81b287cecde60c42f7e1e3ec171244ffc18aa6
-2026-07-27-skill-catalog-hot-refresh.zh.md: ca570a3c6a0402e6764824e3101e15054769b85d
+2026-07-27-skill-catalog-hot-refresh.md: 8e70fb10e7da4292325b72f3a0392bef2271738c
+2026-07-27-skill-catalog-hot-refresh.zh.md: 9a6b6d944baa4a9cc4dddb5158fdcbac2b05f5db

.agents/notes/implemented/feature/2026-07-27-skill-catalog-hot-refresh.md

@@ -12,7 +12,7 @@ Filesystem updates are also non-atomic from the observer's perspective. An edito
 
 ## Decision
 
-The skill capability separates catalog membership from instruction-body loading. `ctx.skills.snapshot()` returns summaries plus a completeness bit. `ctx.skills.registerProvider(factory)` gives the synchronous factory one registration-scoped `{ signal, invalidate }` control: `invalidate()` dirties only that exact active registration and discards completed catalog caches, while the signal aborts when registration fails or is disposed. Provider arrays are complete-discovery shorthand; an explicit incomplete observation can retain readable candidates for direct loads without becoming cacheable or authoritative for model-facing consumers. A provider or runtime generation change during discovery retries before returning. A late invalidation after disposal or replacement is a no-op because the capability has been revoked.
+The skill capability separates catalog membership from instruction-body loading. `ctx.skills.snapshot()` returns summaries plus a completeness bit. `ctx.skills.registerProvider(factory)` gives the synchronous factory one registration-scoped `{ signal, invalidate }` control: `invalidate()` dirties only that exact active registration and discards completed catalog caches, while the signal aborts when registration fails or is disposed. Provider arrays are complete-discovery shorthand; an explicit incomplete observation can retain readable candidates for direct loads without becoming cacheable or authoritative for model-facing consumers. A provider or runtime generation change during discovery retries once; if the retry is also superseded, the latest candidates return as an incomplete, uncached observation. A late invalidation after disposal or replacement is a no-op because the capability has been revoked.
 
 `@deepseek-ai/dsh-skill-local` directly depends on Chokidar and observes catalog-relevant host paths. Existing roots watch direct skill bundle directories, flat Markdown entries, and direct `SKILL.md` entry files. Additions, removals, and directory changes invalidate membership; file changes support frontmatter `name` and `description` refresh. Resource files below a bundle are ignored. Events in one microtask batch coalesce to one invalidation. Project watchers use a bounded least-recently-observed set.
 
@@ -26,7 +26,7 @@ Instruction bodies keep progressive disclosure. Every `skill(name)` call asks th
 
 ## Verification
 
-Registry tests pin registration-scoped invalidation, revocation, signal abort, contained observer failures, incomplete candidates, generation retries, and stale-name rejection. Local-provider tests cover bundle and flat-file creation, removal, rename, root creation/deletion/recreation including an unobserved root `unlinkDir`, description changes, body-only edits, first-party observation, symlinks, polling options, persistent watcher failures with loadable candidates, event coalescing, bounded projects, teardown, and transient reads. Tool tests pin full replacement messages, empty tombstones, digest stability for body-only edits, incomplete-state retention, visibility, and resume metadata. TUI tests pin last-complete retention, authoritative empty removal, latest-wins refresh, teardown, and the already-open slash-draft race; a real Loader/PTY smoke adds a local skill after startup and observes its completion without restarting. A keyless assembled agent-spine snapshot creates a project skill through model-facing filesystem tools, observes its replacement catalog on the next request, and loads its current body with the real `skill` tool.
+Registry tests pin registration-scoped invalidation, revocation, signal abort, contained observer failures, incomplete candidates, bounded generation retries, and stale-name rejection. Local-provider tests cover bundle and flat-file creation, removal, rename, root creation/deletion/recreation including an unobserved root `unlinkDir`, description changes, body-only edits, first-party observation, symlinks, polling options, persistent watcher failures with loadable candidates, event coalescing, bounded projects, teardown, and transient reads. Tool tests pin full replacement messages, empty tombstones, digest stability for body-only edits, incomplete-state retention, visibility, and resume metadata. TUI tests pin last-complete retention, authoritative empty removal, latest-wins refresh, teardown, and the already-open slash-draft race; a real Loader/PTY smoke adds a local skill after startup and observes its completion without restarting. A keyless assembled agent-spine snapshot creates a project skill through model-facing filesystem tools, observes its replacement catalog on the next request, and loads its current body with the real `skill` tool.
 
 ## Alternatives considered
 

.agents/notes/implemented/feature/2026-07-27-skill-catalog-hot-refresh.zh.md

@@ -12,7 +12,7 @@ skill（技能）摘要是模型的路由输入，但本地 skill 可在会话
 
 ## 决策
 
-skill 服务将目录成员关系与指令正文加载分离。`ctx.skills.snapshot()` 返回摘要及一个完整性位。`ctx.skills.registerProvider(factory)` 会向同步工厂提供一项注册作用域内的 `{ signal, invalidate }` 控制能力：`invalidate()` 只会将该精确活动注册标记为脏，并丢弃已完成目录缓存；注册失败或释放时，信号会中止。提供方返回的数组是完整发现的简写形式；显式的不完整观测可以保留可读候选项供直接加载，但不能缓存，也不能作为面向模型消费方的权威结果。在发现期间，如果提供方或运行时 generation 发生变化，系统会先重试再返回。资源释放或替换后的延迟失效操作不会执行任何操作，因为该能力已被撤销。
+skill 服务将目录成员关系与指令正文加载分离。`ctx.skills.snapshot()` 返回摘要及一个完整性位。`ctx.skills.registerProvider(factory)` 会向同步工厂提供一项注册作用域内的 `{ signal, invalidate }` 控制能力：`invalidate()` 只会将该精确活动注册标记为脏，并丢弃已完成目录缓存；注册失败或释放时，信号会中止。提供方返回的数组是完整发现的简写形式；显式的不完整观测可以保留可读候选项供直接加载，但不能缓存，也不能作为面向模型消费方的权威结果。在发现期间，如果提供方或运行时 generation 发生变化，系统会重试一次；如果这次重试也被后续修订取代，则最新候选项会作为不完整且不缓存的观测返回。资源释放或替换后的延迟失效操作不会执行任何操作，因为该能力已被撤销。
 
 `@deepseek-ai/dsh-skill-local` 直接依赖 Chokidar，并观察与目录相关的宿主路径。已有根目录会监视其直属 skill bundle 目录、平铺的 Markdown 条目和直属 `SKILL.md` 条目文件。新增、移除和目录变更会使成员关系失效；文件变更还支持刷新 frontmatter 中的 `name` 和 `description`。bundle 内更深层的资源文件会被忽略。同一微任务批次中的事件会合并为一次失效。项目 watcher 使用有界集合，并按最久未观察顺序淘汰。
 
@@ -26,7 +26,7 @@ TUI 将同一失效通知作为界面状态而非会话历史来消费。`skills
 
 ## 验证
 
-注册表测试固定了注册作用域内的失效、能力撤销、信号中止、监听器失败隔离、不完整候选项、generation 重试和陈旧名称拒绝。local-provider 测试覆盖 bundle 与平铺文件的创建、移除和重命名，以及根目录创建／删除／重建（包括未观测到根目录 `unlinkDir` 事件的情形）、描述变更、仅正文编辑、第一方观察、符号链接、轮询选项、候选项仍可加载的持续 watcher 失败、事件合并、项目 watcher 容量上限、资源销毁和暂时读取。工具测试固定了完整替换消息、空 tombstone、仅修改正文时 digest 稳定、不完整状态保留、可见性和恢复元数据。TUI 测试固定了上一份完整结果保留、权威空结果清除、刷新时以最新结果为准、资源销毁和已打开斜杠草稿的竞态；一项使用真实 Loader/PTY 的 smoke 测试会在启动后添加本地 skill，并观察其补全项出现，而无需重启。一个无密钥、装配完成的 agent-spine 快照测试通过面向模型的文件系统工具创建项目 skill，观察下一次请求中的替换目录，并使用真实 `skill` 工具加载当前正文。
+注册表测试固定了注册作用域内的失效、能力撤销、信号中止、监听器失败隔离、不完整候选项、有界 generation 重试和陈旧名称拒绝。local-provider 测试覆盖 bundle 与平铺文件的创建、移除和重命名，以及根目录创建／删除／重建（包括未观测到根目录 `unlinkDir` 事件的情形）、描述变更、仅正文编辑、第一方观察、符号链接、轮询选项、候选项仍可加载的持续 watcher 失败、事件合并、项目 watcher 容量上限、资源销毁和暂时读取。工具测试固定了完整替换消息、空 tombstone、仅修改正文时 digest 稳定、不完整状态保留、可见性和恢复元数据。TUI 测试固定了上一份完整结果保留、权威空结果清除、刷新时以最新结果为准、资源销毁和已打开斜杠草稿的竞态；一项使用真实 Loader/PTY 的 smoke 测试会在启动后添加本地 skill，并观察其补全项出现，而无需重启。一个无密钥、装配完成的 agent-spine 快照测试通过面向模型的文件系统工具创建项目 skill，观察下一次请求中的替换目录，并使用真实 `skill` 工具加载当前正文。
 
 ## 考虑过的替代方案
 

docs/config-catalog.md

@@ -1241,7 +1241,7 @@ export interface Config {
 }
 ```
 
-Source: [`packages/skill/skill/src/index.ts:138`](../packages/skill/skill/src/index.ts)
+Source: [`packages/skill/skill/src/index.ts:139`](../packages/skill/skill/src/index.ts)
 
 ## `@deepseek-ai/dsh-skill-local`
 

docs/cordis-catalog/events.md

@@ -659,7 +659,7 @@ A skill provider, runtime contribution, or provider-backed catalog may have chan
 'skills/change'(): void
 ```
 
-Source: [`packages/skill/skill/src/index.ts:156`](../../packages/skill/skill/src/index.ts)
+Source: [`packages/skill/skill/src/index.ts:157`](../../packages/skill/skill/src/index.ts)
 
 ## `slash/*`
 

docs/cordis-catalog/services.md

@@ -1660,11 +1660,11 @@ register(skill: SkillRegistration): () => void
 async list(options: SkillLookupOptions = {}): Promise<SkillSummary[]>
 
 /**
- * Observe the current model-invocable catalog and whether all providers completed discovery.
+ * Observe the current model-invocable catalog and whether discovery completed within a stable revision.
  * Incomplete observations are never cached, allowing consumers to retain last-good state and
  * retry on their next request boundary.
  * @param options - lookup options; `cwd` selects project roots and `signal` cancels discovery.
- * @returns sorted summaries plus provider-completeness state.
+ * @returns sorted summaries plus discovery-completeness state.
  */
 async snapshot(options: SkillLookupOptions = {}): Promise<SkillCatalogSnapshot>
 
@@ -1681,7 +1681,7 @@ async get(name: string, options: SkillLookupOptions = {}): Promise<SkillDefiniti
 
 Types: [SkillCatalogSnapshot](../core-data-structures/skills.md) · [SkillDefinition](../core-data-structures/skills.md) · [SkillLookupOptions](../core-data-structures/skills.md) · [SkillProvider](../core-data-structures/skills.md) · [SkillProviderControl](../core-data-structures/skills.md) · [SkillRegistration](../core-data-structures/skills.md) · [SkillSummary](../core-data-structures/skills.md)
 
-Source: [`packages/skill/skill/src/index.ts:177`](../../packages/skill/skill/src/index.ts)
+Source: [`packages/skill/skill/src/index.ts:178`](../../packages/skill/skill/src/index.ts)
 
 ## `ctx.spillStore` — `SpillStore` (abstract seam)
 

docs/core-data-structures/skills.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/core-data-structures/skills.md
-skills.md: 34c85ddfb4b8eec7183202fe54cd70642894b0d7
-skills.zh.md: 2fd2f73f0471bdc0c8b65b4467b5ae066943fb7c
+skills.md: a52a21d2bfe4c4e4529953fdf09998a9a281cefd
+skills.zh.md: 6ebc167f7e5dc436ac1f79b8747940e4beb30222

docs/core-data-structures/skills.md

@@ -10,7 +10,7 @@ Source: [`packages/skill/skill/src/index.ts`](../../packages/skill/skill/src/ind
 
 `ctx.skills` combines local, embedded, remote, or other providers. Registration is synchronous; remote initialization and discovery belong in awaited `list()`. Provider objects, options, and candidates are borrowed readonly, while semantic fields are validated.
 
-Duplicate names resolve by rank, provider order, then local order; summaries sort by name. A rejected `list()` is logged and omitted from an incomplete observation, while an explicit incomplete observation contributes usable candidates without making the result cacheable; malformed candidates fail fast. Each provider factory receives a registration-scoped control whose `invalidate()` clears completed catalogs only while that exact registration remains active and whose signal aborts on failed registration or disposal. An in-flight discovery retries when its provider generation changes. Provider and runtime mutations emit the unfiltered `skills/change` invalidation event; it carries no diff, so consumers refetch `snapshot()` with their own lookup options.
+Duplicate names resolve by rank, provider order, then local order; summaries sort by name. A rejected `list()` is logged and omitted from an incomplete observation, while an explicit incomplete observation contributes usable candidates without making the result cacheable; malformed candidates fail fast. Each provider factory receives a registration-scoped control whose `invalidate()` clears completed catalogs only while that exact registration remains active and whose signal aborts on failed registration or disposal. An in-flight discovery retries once when its provider generation changes; a second change returns the latest candidates incomplete and uncached. Provider and runtime mutations emit the unfiltered `skills/change` invalidation event; it carries no diff, so consumers refetch `snapshot()` with their own lookup options.
 
 An array returned by `SkillProvider.list()` is complete-discovery shorthand. `SkillProviderObservation` lets a provider expose candidates that remain directly loadable while reporting that the observation is not authoritative.
 
@@ -109,14 +109,14 @@ interface SkillSummary {
 }
 ```
 
-`SkillCatalogSnapshot` distinguishes authoritative absence from transient provider failure. `skills` contains the sorted summaries collected in that observation; `complete` is true only when every registered provider completed. Incomplete snapshots are not cached, allowing a consumer to retain its last-good model catalog and retry.
+`SkillCatalogSnapshot` distinguishes authoritative absence from transient provider failure or a catalog that kept changing during discovery. `skills` contains the sorted summaries collected in that observation; `complete` is true only when every registered provider completed without a concurrent catalog revision. Incomplete snapshots are not cached, allowing a consumer to retain its last-good model catalog and retry.
 
 ```ts type-equiv
-/** One catalog observation plus whether every registered provider completed discovery. */
+/** One catalog observation plus whether discovery completed within a stable catalog revision. */
 interface SkillCatalogSnapshot {
   /** Sorted model-invocable summaries collected in this observation. */
   readonly skills: SkillSummary[]
-  /** Whether every registered provider completed discovery for this observation. */
+  /** Whether every registered provider completed without a concurrent catalog revision. */
   readonly complete: boolean
 }
 ```

docs/core-data-structures/skills.zh.md

@@ -10,7 +10,7 @@
 
 `ctx.skills` 组合本地、内嵌、远程或其他提供方。注册是同步的；远程初始化与发现属于 `list()` 的 await 阶段。提供方对象、选项与候选项以只读方式借用，语义字段会被校验。
 
-重名按 rank、提供方顺序、本地顺序依次解决；摘要按名称排序。`list()` 拒绝时会记录日志并从不完整观测中省略；显式的不完整观测会提供可用候选项，但不会使结果变得可缓存；格式错误的候选项快速失败。每个提供方工厂都会接收一项注册作用域内的控制能力；仅当该精确注册仍处于活动状态时，其 `invalidate()` 才会清除已完成目录；注册失败或释放时，其信号会中止。若提供方代次在发现进行期间发生变化，该发现会重试。提供方和运行时变更会发出不带过滤条件的 `skills/change` 失效事件；该事件不携带 diff，因此消费方会使用自身的查找选项重新获取 `snapshot()`。
+重名按 rank、提供方顺序、本地顺序依次解决；摘要按名称排序。`list()` 拒绝时会记录日志并从不完整观测中省略；显式的不完整观测会提供可用候选项，但不会使结果变得可缓存；格式错误的候选项快速失败。每个提供方工厂都会接收一项注册作用域内的控制能力；仅当该精确注册仍处于活动状态时，其 `invalidate()` 才会清除已完成目录；注册失败或释放时，其信号会中止。若提供方代次在发现进行期间发生变化，该发现会重试一次；若再次变化，则返回最新候选项，并将结果标为不完整且不予缓存。提供方和运行时变更会发出不带过滤条件的 `skills/change` 失效事件；该事件不携带 diff，因此消费方会使用自身的查找选项重新获取 `snapshot()`。
 
 `SkillProvider.list()` 返回的数组是完整发现的简写形式。`SkillProviderObservation` 允许提供方公开仍可直接加载的候选项，同时报告该观测不具权威性。
 
@@ -109,14 +109,14 @@ interface SkillSummary {
 }
 ```
 
-`SkillCatalogSnapshot` 用于区分已确定的不存在和提供方的瞬时失败。`skills` 包含该次观测中收集并排序的摘要；只有每个已注册提供方都已完成发现，`complete` 才为 true。不完整快照不会缓存，因此消费方可以保留上一份可用模型目录并重试。
+`SkillCatalogSnapshot` 用于区分已确定的不存在与提供方的瞬时失败或发现期间持续变化的目录。`skills` 包含该次观测中收集并排序的摘要；只有每个已注册提供方都在没有并发目录修订时完成发现，`complete` 才为 true。不完整快照不会缓存，因此消费方可以保留上一份可用模型目录并重试。
 
 ```ts type-equiv
-/** One catalog observation plus whether every registered provider completed discovery. */
+/** One catalog observation plus whether discovery completed within a stable catalog revision. */
 interface SkillCatalogSnapshot {
   /** Sorted model-invocable summaries collected in this observation. */
   readonly skills: SkillSummary[]
-  /** Whether every registered provider completed discovery for this observation. */
+  /** Whether every registered provider completed without a concurrent catalog revision. */
   readonly complete: boolean
 }
 ```

docs/event-producer-consumer.md

@@ -35,7 +35,7 @@ This matrix shows which packages dispatch each harness-owned event and which pac
 | `session/disposed` | `emit` | [`packages/core/session/src/index.ts:81`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`agent-loop`](../packages/core/agent-loop), `apiproxy`, [`session-persistence`](../packages/session-persistence/session-persistence), [`session-projection-cache`](../packages/session-projection/session-projection-cache), [`session-telemetry`](../packages/telemetry/session-telemetry), [`session-title`](../packages/session-title/session-title) |
 | `session/event` | `emit` | [`packages/core/session/src/index.ts:93`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/acp/acp), `apiproxy`, [`cli-demo`](../packages/examples/cli-demo), [`compact`](../packages/compact/compact), [`compact-basic`](../packages/compact/compact-basic), [`goal`](../packages/goal/goal), [`goal-session`](../packages/goal/goal-session), [`hook-protocol`](../packages/hooks/hook-protocol), [`jsonrpc`](../packages/ui/jsonrpc), [`session`](../packages/core/session), [`session-persistence`](../packages/session-persistence/session-persistence), [`session-projection`](../packages/session-projection/session-projection), [`session-projection-cache`](../packages/session-projection/session-projection-cache), [`session-telemetry`](../packages/telemetry/session-telemetry), [`session-title`](../packages/session-title/session-title), [`token-meter`](../packages/llm/token-meter), [`tools`](../packages/core/tools), [`tui`](../packages/ui/tui), [`user-approval`](../packages/ui/user-approval), [`workspace-context`](../packages/context/workspace-context) |
 | `session/flush` | `parallel` | [`packages/core/session/src/index.ts:103`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session-persistence/session-persistence), [`session-telemetry`](../packages/telemetry/session-telemetry) |
-| `skills/change` | `emit` | [`packages/skill/skill/src/index.ts:156`](../packages/skill/skill/src/index.ts) | [`skill`](../packages/skill/skill) (`events.dispatch`) | [`tui`](../packages/ui/tui) |
+| `skills/change` | `emit` | [`packages/skill/skill/src/index.ts:157`](../packages/skill/skill/src/index.ts) | [`skill`](../packages/skill/skill) (`events.dispatch`) | [`tui`](../packages/ui/tui) |
 | `slash/input-begin-command` | `bail` | [`packages/client/ui-slash/src/types.ts:230`](../packages/client/ui-slash/src/types.ts) | - | `ui-conversation` |
 | `slash/input-consume-token` | `bail` | [`packages/client/ui-slash/src/types.ts:244`](../packages/client/ui-slash/src/types.ts) | - | `ui-conversation` |
 | `slash/input-insert-reference` | `bail` | [`packages/client/ui-slash/src/types.ts:237`](../packages/client/ui-slash/src/types.ts) | - | `ui-conversation` |

packages/cordis/tool-cordis/src/api-catalog.ts

@@ -762,7 +762,7 @@ export const SERVICE_API: readonly ServiceApiEntry[] = [
       },
       {
         signature: 'async snapshot(options: SkillLookupOptions = {}): Promise<SkillCatalogSnapshot>',
-        jsDoc: '/**\n * Observe the current model-invocable catalog and whether all providers completed discovery.\n * Incomplete observations are never cached, allowing consumers to retain last-good state and\n * retry on their next request boundary.\n * @param options - lookup options; `cwd` selects project roots and `signal` cancels discovery.\n * @returns sorted summaries plus provider-completeness state.\n */',
+        jsDoc: '/**\n * Observe the current model-invocable catalog and whether discovery completed within a stable revision.\n * Incomplete observations are never cached, allowing consumers to retain last-good state and\n * retry on their next request boundary.\n * @param options - lookup options; `cwd` selects project roots and `signal` cancels discovery.\n * @returns sorted summaries plus discovery-completeness state.\n */',
       },
       {
         signature: 'async get(name: string, options: SkillLookupOptions = {}): Promise<SkillDefinition | undefined>',

packages/skill/skill/README.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/skill/skill/README.md
-README.md: 9813338dcec82fc2db7e149be1bd80ec5239684d
-README.zh.md: 70558875a2d223367e2b35f326fba4b5d966fae3
+README.md: 65ff110999ea416648f3d676eb813dd4ceb194f8
+README.zh.md: 79e77d2a2846489125ba0ecaafc138172ba2fd86

packages/skill/skill/README.md

@@ -11,7 +11,7 @@ This package owns the `ctx.skills` interface. It does not know whether skills co
 ### Public API
 
 - `ctx.skills.registerProvider(create): () => void` Calls a synchronous provider factory with `{ signal, invalidate }`, then registers its readonly result by unique `provider.name`. Duplicate names throw, `runtime` is reserved, and failed registration aborts the signal. The exact Cordis disposer unregisters the provider, aborts the signal, and preserves ordered composite teardown.
-- `ctx.skills.snapshot({ cwd?, signal? })` Returns `{ skills, complete }`. `complete` is false when any provider rejects or explicitly reports incomplete discovery; candidates supplied with an incomplete observation remain in this result, which is never cached.
+- `ctx.skills.snapshot({ cwd?, signal? })` Returns `{ skills, complete }`. `complete` is false when any provider rejects or explicitly reports incomplete discovery, or when a second catalog revision races the bounded retry; candidates supplied by that observation remain in this result, which is never cached.
 - `ctx.skills.list({ cwd?, signal? })` Borrows the readonly lookup options, then returns model-invocable summaries for the current workspace, merged across providers and sorted by name.
 - `ctx.skills.get(name, { cwd?, signal? })` Uses the same readonly options and winning candidate for discovery and loading, rechecks cancellation after discovery or a cache hit, races provider loading against the signal, validates the loaded definition, then returns it, including disabled-for-model skills.
 - `ctx.skills.register(skill): () => void` Registers a readonly runtime embedded skill, adding `provider: "runtime"` when omitted. Same-name runtime registrations are first-wins: a duplicate logs a warning and gets a no-op disposer. Successful registrations return the exact Cordis disposer for ordered composite teardown.
@@ -32,7 +32,7 @@ A provider factory runs synchronously and receives one registration-scoped contr
 
 The registry validates candidates before caching and definitions before returning them. The winning provider receives the same candidate and opaque `locator` it returned from `list()`, allowing backend-specific file, URL, id, or version handles. Callers and providers must preserve the readonly contract.
 
-Contract violations fail fast. A rejected provider `list()` is treated as a transient source failure and omitted. An explicit incomplete observation still contributes its candidates for `list()` and `get()`, but makes the aggregate snapshot incomplete and uncacheable. A provider or runtime revision change discards an in-flight result and retries before returning. Duplicate names resolve by rank, provider registration order, then provider-local order. Summaries are sorted by skill name.
+Contract violations fail fast. A rejected provider `list()` is treated as a transient source failure and omitted. An explicit incomplete observation still contributes its candidates for `list()` and `get()`, but makes the aggregate snapshot incomplete and uncacheable. A provider or runtime revision change discards an in-flight result and retries once. If the retry is also superseded, its candidates are returned incomplete and uncached so a continuously invalidating provider cannot monopolize the caller. Duplicate names resolve by rank, provider registration order, then provider-local order. Summaries are sorted by skill name.
 
 Definitions remain progressively loaded. `get()` asks the winning provider for the body on every call rather than caching it in this registry. If the returned definition has a different name from the selected candidate, the stale selection is rejected and the registry internally invalidates that exact provider so the next snapshot rediscovers its catalog.
 

packages/skill/skill/README.zh.md

@@ -11,7 +11,7 @@
 ### 公开 API
 
 - `ctx.skills.registerProvider(create): () => void` 调用同步提供方工厂并向其传入 `{ signal, invalidate }`，随后使用唯一 `provider.name` 注册其只读结果。重复提供方名称会抛错，`runtime` 为保留名称；注册失败会中止信号。精确的 Cordis disposer 会注销提供方、中止信号，并保持有序组合拆卸。
-- `ctx.skills.snapshot({ cwd?, signal? })` 返回 `{ skills, complete }`。任一提供方调用被拒绝或显式报告发现不完整时，`complete` 为 false；不完整观测提供的候选项仍保留在该结果中，但该结果绝不缓存。
+- `ctx.skills.snapshot({ cwd?, signal? })` 返回 `{ skills, complete }`。任一提供方调用被拒绝或显式报告发现不完整，或有界重试期间又发生目录修订时，`complete` 为 false；该次观测提供的候选项仍保留在此结果中，但该结果绝不缓存。
 - `ctx.skills.list({ cwd?, signal? })` 借用只读查找选项，然后返回当前工作区中模型可调用的摘要；这些摘要跨提供方合并，并按名称排序。
 - `ctx.skills.get(name, { cwd?, signal? })` 在发现和加载中使用同一组只读选项和胜出候选项；在发现或缓存命中后重新检查取消，让提供方加载与信号竞速，验证已加载定义，然后将其返回，包括已对模型禁用的 skill。
 - `ctx.skills.register(skill): () => void` 注册只读运行时嵌入式 skill，省略时添加 `provider: "runtime"`。同名运行时注册使用先到先得：重复项会记录警告，并获得无操作 disposer。成功注册会返回精确的 Cordis disposer，以供有序组合拆卸。
@@ -32,7 +32,7 @@
 
 注册表在缓存前验证候选项，在返回前验证定义。胜出提供方会收到同一候选项和不透明 `locator`，两者都是它从 `list()` 返回的内容，从而支持后端专用文件、URL、id 或版本句柄。调用方和提供方必须保持只读契约。
 
-违反契约时会快速失败。`list()` 返回的 Promise 被拒绝会被视为瞬时来源失败，并省略其结果。显式的不完整观测仍会为 `list()` 和 `get()` 提供其候选项，但会使聚合快照不完整且不可缓存。提供方或运行时修订发生变化时，会丢弃正在进行的结果并在返回前重试。重复名称依次按 rank、提供方注册顺序和提供方本地顺序解决冲突。摘要按 skill 名称排序。
+违反契约时会快速失败。`list()` 返回的 Promise 被拒绝会被视为瞬时来源失败，并省略其结果。显式的不完整观测仍会为 `list()` 和 `get()` 提供其候选项，但会使聚合快照不完整且不可缓存。提供方或运行时修订发生变化时，会丢弃正在进行的结果并重试一次。如果这次重试也被后续修订取代，则返回其候选项，并将结果标为不完整且不予缓存，以免持续触发失效的提供方一直占用调用方。重复名称依次按 rank、提供方注册顺序和提供方本地顺序解决冲突。摘要按 skill 名称排序。
 
 定义仍采用渐进式加载。`get()` 每次调用都会向胜出提供方请求正文，而不是在此注册表中缓存正文。若返回定义的名称不同于所选候选项，系统会拒绝该陈旧选择，并由注册表在内部使该精确提供方失效，以便下一次快照重新发现其目录。
 

packages/skill/skill/src/index.ts

@@ -15,6 +15,7 @@ import type Schema from 'schemastery'
 
 const SKILL_NAME = /^[a-z0-9]+(?:-[a-z0-9]+)*$/
 const DEFAULT_COLLECT_CACHE_ENTRIES = 128
+const MAX_COLLECT_ATTEMPTS = 2
 const RUNTIME_PROVIDER = 'runtime'
 const RUNTIME_RANK = 250
 
@@ -87,11 +88,11 @@ export interface SkillLookupOptions {
   readonly signal?: AbortSignal | undefined
 }
 
-/** One catalog observation plus whether every registered provider completed discovery. */
+/** One catalog observation plus whether discovery completed within a stable catalog revision. */
 export interface SkillCatalogSnapshot {
   /** Sorted model-invocable summaries collected in this observation. */
   readonly skills: SkillSummary[]
-  /** Whether every registered provider completed discovery for this observation. */
+  /** Whether every registered provider completed without a concurrent catalog revision. */
   readonly complete: boolean
 }
 
@@ -286,11 +287,11 @@ export class SkillService extends Service {
   }
 
   /**
-   * Observe the current model-invocable catalog and whether all providers completed discovery.
+   * Observe the current model-invocable catalog and whether discovery completed within a stable revision.
    * Incomplete observations are never cached, allowing consumers to retain last-good state and
    * retry on their next request boundary.
    * @param options - lookup options; `cwd` selects project roots and `signal` cancels discovery.
-   * @returns sorted summaries plus provider-completeness state.
+   * @returns sorted summaries plus discovery-completeness state.
    */
   async snapshot(options: SkillLookupOptions = {}): Promise<SkillCatalogSnapshot> {
     const collected = await this.collect(options)
@@ -333,6 +334,7 @@ export class SkillService extends Service {
 
   private async collect(options: SkillLookupOptions): Promise<CollectResult> {
     throwIfAborted(options.signal)
+    let attempt = 1
     while (true) {
       const providerRevision = this.providerRevision
       const runtimeRevision = this.runtimeRevision
@@ -342,7 +344,13 @@ export class SkillService extends Service {
 
       const result = await this.collectFresh(options)
       throwIfAborted(options.signal)
-      if (providerRevision !== this.providerRevision || runtimeRevision !== this.runtimeRevision) continue
+      if (providerRevision !== this.providerRevision || runtimeRevision !== this.runtimeRevision) {
+        if (attempt < MAX_COLLECT_ATTEMPTS) {
+          attempt += 1
+          continue
+        }
+        return { entries: result.entries, cacheable: false }
+      }
       if (result.cacheable) {
         this.collectCache.set(key, result.entries)
         if (this.collectCache.size > this.collectCacheMaxEntries) {

packages/skill/skill/tests/skill.spec.ts

@@ -748,6 +748,40 @@ describe('SkillService registry', () => {
     expect(provider.listCalls).toBe(2)
   })
 
+  it('bounds repeated in-flight invalidation and leaves the result uncached', async () => {
+    const ctx = new Context()
+    await ctx.plugin(SkillService)
+    let listCalls = 0
+    ctx.skills.registerProvider(control => ({
+      name: 'self-invalidating',
+      async list() {
+        listCalls += 1
+        control.invalidate()
+        return [{
+          ...memorySkill('bounded-skill', `Attempt ${listCalls}`, 10),
+          provider: 'self-invalidating',
+        }]
+      },
+      async get() {
+        return undefined
+      },
+    }))
+
+    expect(await ctx.skills.snapshot()).toEqual({
+      skills: [{
+        name: 'bounded-skill',
+        description: 'Attempt 2',
+        provider: 'self-invalidating',
+        source: 'memory',
+      }],
+      complete: false,
+    })
+    expect(listCalls).toBe(2)
+
+    expect((await ctx.skills.snapshot()).skills[0]?.description).toBe('Attempt 4')
+    expect(listCalls).toBe(4)
+  })
+
   it('invalidates a provider whose loaded definition changed identity', async () => {
     const ctx = new Context()
     await ctx.plugin(SkillService)