@colin4k1024/tsp 2.4.1 → 2.4.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (216) hide show
  1. package/README.md +12 -6
  2. package/docs/.vitepress/config.mts +199 -0
  3. package/docs/adr/ADR-001-doc-architecture-integration.md +33 -0
  4. package/docs/guides/README.md +5 -0
  5. package/docs/guides/installation.md +33 -0
  6. package/docs/guides/user-guide.md +36 -0
  7. package/docs/index.md +65 -0
  8. package/docs/memory/backlog.md +10 -0
  9. package/docs/memory/decisions.md +43 -0
  10. package/docs/memory/lessons-learned.md +87 -0
  11. package/docs/plans/2026-04-03-python-remnants-audit.md +265 -0
  12. package/docs/plans/2026-04-03-scripts-python-to-js-migration.md +372 -0
  13. package/docs/plans/2026-04-03-solo-delivery-execution-checklist.md +413 -0
  14. package/docs/plans/2026-04-03-solo-delivery-gap-plan.md +377 -0
  15. package/docs/plans/2026-04-03-team-skills-workflow-gates.md +548 -0
  16. package/docs/plans/2026-04-21-open-source-readiness-gap-plan.md +217 -0
  17. package/docs/plans/llm-surface-reduction-audit.md +147 -0
  18. package/docs/plans/llm-surface-reduction-execution-checklist.md +217 -0
  19. package/docs/plans/llm-surface-reduction-execution-history.md +124 -0
  20. package/docs/plans/team-skills-platform-migration.md +54 -0
  21. package/docs/presentation/README.md +42 -0
  22. package/docs/presentation/audience-presentation-route-map.md +84 -0
  23. package/docs/presentation/executive-briefing-talk-track.md +50 -0
  24. package/docs/presentation/generate_capability_matrix.py +396 -0
  25. package/docs/presentation/generate_ppt.py +354 -0
  26. package/docs/presentation/implementation-onboarding-brief.md +38 -0
  27. package/docs/presentation/presentation-talk-track.md +97 -0
  28. package/docs/presentation/vertical-scenario-route-map.md +99 -0
  29. package/docs/presentation/workshop-facilitator-guide.md +47 -0
  30. package/docs/runbooks/actionlint-workflow-gates.md +80 -0
  31. package/docs/runbooks/agent-governance.md +131 -0
  32. package/docs/runbooks/ai-eval-platform-demo-execution-log.md +147 -0
  33. package/docs/runbooks/ai-eval-platform-demo-script.md +136 -0
  34. package/docs/runbooks/ai-eval-platform-walkthrough.md +113 -0
  35. package/docs/runbooks/ai-pr-review-automation.md +56 -0
  36. package/docs/runbooks/api-breaking-change-gates.md +58 -0
  37. package/docs/runbooks/api-design-evolution-walkthrough.md +42 -0
  38. package/docs/runbooks/api-lint-gates.md +57 -0
  39. package/docs/runbooks/api-mocking-strategy-and-lifecycle-guide.md +47 -0
  40. package/docs/runbooks/architect-daily-operations.md +63 -0
  41. package/docs/runbooks/architect-design-conversation-example.md +83 -0
  42. package/docs/runbooks/artifact-attestation-gates.md +75 -0
  43. package/docs/runbooks/artifact-persistence.md +257 -0
  44. package/docs/runbooks/backend-engineer-daily-operations.md +63 -0
  45. package/docs/runbooks/batch-optimization-completion-checklist.md +104 -0
  46. package/docs/runbooks/biz-service-designer-end-to-end-conversation-example.md +5 -0
  47. package/docs/runbooks/biz-service-designer-toolkit.md +5 -0
  48. package/docs/runbooks/bug-fix-complete-walkthrough.md +60 -0
  49. package/docs/runbooks/build-failure-recovery-walkthrough.md +40 -0
  50. package/docs/runbooks/canary-decision-matrix.md +41 -0
  51. package/docs/runbooks/canary-staging-release-walkthrough.md +46 -0
  52. package/docs/runbooks/checkov-iac-gates.md +104 -0
  53. package/docs/runbooks/claude-code-review-workflow.md +72 -0
  54. package/docs/runbooks/claude-conversation-prompt-recipes.md +132 -0
  55. package/docs/runbooks/claude-end-to-end-conversation-example.md +198 -0
  56. package/docs/runbooks/claude-feature-development-guide.md +112 -0
  57. package/docs/runbooks/claude-quick-start.md +227 -0
  58. package/docs/runbooks/claude-usage-scenarios.md +176 -0
  59. package/docs/runbooks/code-review-collaboration-walkthrough.md +65 -0
  60. package/docs/runbooks/codeql-pr-security-gates.md +64 -0
  61. package/docs/runbooks/codex-end-to-end-conversation-example.md +166 -0
  62. package/docs/runbooks/codex-multi-agent-orchestration.md +65 -0
  63. package/docs/runbooks/codex-parallel-prompt-recipes.md +131 -0
  64. package/docs/runbooks/codex-quick-start.md +223 -0
  65. package/docs/runbooks/codex-usage-scenarios.md +168 -0
  66. package/docs/runbooks/codex-workflow-essentials.md +88 -0
  67. package/docs/runbooks/command-and-capability-matrix.md +162 -0
  68. package/docs/runbooks/conftest-policy-gates.md +84 -0
  69. package/docs/runbooks/consumer-driven-contract-testing-with-mock-alignment.md +45 -0
  70. package/docs/runbooks/contract-testing-playbook.md +78 -0
  71. package/docs/runbooks/cosign-signing-gates.md +71 -0
  72. package/docs/runbooks/cross-role-issue-triage-walkthrough.md +47 -0
  73. package/docs/runbooks/cursor-quick-start.md +123 -0
  74. package/docs/runbooks/custom-overlay.md +115 -0
  75. package/docs/runbooks/data-ml-pipeline-demo-execution-log.md +141 -0
  76. package/docs/runbooks/data-ml-pipeline-demo-script.md +102 -0
  77. package/docs/runbooks/data-ml-pipeline-walkthrough.md +119 -0
  78. package/docs/runbooks/data-observability-quality-demo-execution-log.md +36 -0
  79. package/docs/runbooks/data-observability-quality-demo-script.md +42 -0
  80. package/docs/runbooks/data-observability-quality-walkthrough.md +86 -0
  81. package/docs/runbooks/demo-deliverables-overview.md +278 -0
  82. package/docs/runbooks/demo-execution-log.md +530 -0
  83. package/docs/runbooks/demo-scenario.md +129 -0
  84. package/docs/runbooks/dependency-review-gates.md +63 -0
  85. package/docs/runbooks/dependency-update-automation.md +83 -0
  86. package/docs/runbooks/design-md-workflow.md +185 -0
  87. package/docs/runbooks/devops-engineer-daily-operations.md +60 -0
  88. package/docs/runbooks/devops-release-conversation-example.md +88 -0
  89. package/docs/runbooks/doc-architecture-integration.md +59 -0
  90. package/docs/runbooks/doc-architecture-quick-start.md +122 -0
  91. package/docs/runbooks/document-execution-audit.md +32 -0
  92. package/docs/runbooks/documentation-update-walkthrough.md +37 -0
  93. package/docs/runbooks/ecc-harness-usage.md +93 -0
  94. package/docs/runbooks/error-experience-usage.md +116 -0
  95. package/docs/runbooks/evolution-usage.md +162 -0
  96. package/docs/runbooks/executive-value-one-page.md +55 -0
  97. package/docs/runbooks/external-capability-approval-and-enablement-workflow.md +39 -0
  98. package/docs/runbooks/external-capability-intake.md +160 -0
  99. package/docs/runbooks/first-team-command-60-seconds.md +96 -0
  100. package/docs/runbooks/first-team-workflow-walkthrough.md +245 -0
  101. package/docs/runbooks/frontend-backend-integration-acceptance-checklist.md +46 -0
  102. package/docs/runbooks/frontend-backend-parallel-integration-walkthrough.md +48 -0
  103. package/docs/runbooks/frontend-bugfix-one-page.md +82 -0
  104. package/docs/runbooks/frontend-engineer-daily-operations.md +60 -0
  105. package/docs/runbooks/frontend-enterprise-style-profile.md +5 -0
  106. package/docs/runbooks/frontend-governance.md +47 -0
  107. package/docs/runbooks/frontend-refactor-walkthrough.md +42 -0
  108. package/docs/runbooks/git-pr-workflow.md +63 -0
  109. package/docs/runbooks/github-actions-supply-chain-demo-execution-log.md +158 -0
  110. package/docs/runbooks/github-actions-supply-chain-demo-script.md +150 -0
  111. package/docs/runbooks/github-actions-supply-chain-walkthrough.md +117 -0
  112. package/docs/runbooks/github-token-permissions-baseline.md +92 -0
  113. package/docs/runbooks/gitlab-manual-pipeline-release.md +5 -0
  114. package/docs/runbooks/gitlab-release-integration-playbook.md +5 -0
  115. package/docs/runbooks/gitnexus-code-intelligence-usage.md +133 -0
  116. package/docs/runbooks/graphify-knowledge-graph-usage.md +88 -0
  117. package/docs/runbooks/handoff-filling-guide-with-examples.md +70 -0
  118. package/docs/runbooks/handoff-governance.md +250 -0
  119. package/docs/runbooks/helm-unittest-playbook.md +101 -0
  120. package/docs/runbooks/hotfix-emergency-release-walkthrough.md +60 -0
  121. package/docs/runbooks/iac-kubernetes-platform-demo-execution-log.md +144 -0
  122. package/docs/runbooks/iac-kubernetes-platform-demo-script.md +130 -0
  123. package/docs/runbooks/iac-kubernetes-platform-walkthrough.md +120 -0
  124. package/docs/runbooks/implementation-onboarding-reading-path.md +67 -0
  125. package/docs/runbooks/in-toto-attestation-framework.md +94 -0
  126. package/docs/runbooks/incident-severity-triage-tree.md +43 -0
  127. package/docs/runbooks/incident-triage-one-page.md +65 -0
  128. package/docs/runbooks/internal-developer-platform-demo-execution-log.md +36 -0
  129. package/docs/runbooks/internal-developer-platform-demo-script.md +42 -0
  130. package/docs/runbooks/internal-developer-platform-walkthrough.md +91 -0
  131. package/docs/runbooks/karpathy-guidelines-usage.md +27 -0
  132. package/docs/runbooks/kubeconform-schema-gates.md +100 -0
  133. package/docs/runbooks/kubectl-server-dry-run-gates.md +103 -0
  134. package/docs/runbooks/kyverno-policy-gates.md +90 -0
  135. package/docs/runbooks/langfuse-and-observability-integration-guide.md +43 -0
  136. package/docs/runbooks/langfuse-coding-trace.md +44 -0
  137. package/docs/runbooks/mobile-miniapp-delivery-walkthrough.md +112 -0
  138. package/docs/runbooks/mobile-miniapp-demo-execution-log.md +139 -0
  139. package/docs/runbooks/mobile-miniapp-demo-script.md +129 -0
  140. package/docs/runbooks/multi-service-backend-integration-walkthrough.md +61 -0
  141. package/docs/runbooks/open-design-integration.md +163 -0
  142. package/docs/runbooks/open-source-release-checklist.md +90 -0
  143. package/docs/runbooks/opencode-quick-start.md +128 -0
  144. package/docs/runbooks/parallel-development-coordination-walkthrough.md +47 -0
  145. package/docs/runbooks/parallel-execution-usage.md +179 -0
  146. package/docs/runbooks/platform-capability-demo-execution-log.md +184 -0
  147. package/docs/runbooks/platform-capability-demo-script.md +192 -0
  148. package/docs/runbooks/plugin-extension-platform-demo-execution-log.md +136 -0
  149. package/docs/runbooks/plugin-extension-platform-demo-script.md +102 -0
  150. package/docs/runbooks/plugin-extension-platform-walkthrough.md +111 -0
  151. package/docs/runbooks/policy-controller-gates.md +75 -0
  152. package/docs/runbooks/post-rollback-verification-checklist.md +37 -0
  153. package/docs/runbooks/pre-release-checklist.md +50 -0
  154. package/docs/runbooks/product-manager-clarification-conversation-example.md +90 -0
  155. package/docs/runbooks/product-manager-daily-operations.md +60 -0
  156. package/docs/runbooks/production-incident-response-walkthrough.md +50 -0
  157. package/docs/runbooks/project-claude-design-rationale.md +188 -0
  158. package/docs/runbooks/project-manager-daily-operations.md +61 -0
  159. package/docs/runbooks/project-manager-planning-conversation-example.md +82 -0
  160. package/docs/runbooks/project-onboarding.md +452 -0
  161. package/docs/runbooks/qa-engineer-daily-operations.md +63 -0
  162. package/docs/runbooks/qa-review-conversation-example.md +87 -0
  163. package/docs/runbooks/release-closure-one-page.md +65 -0
  164. package/docs/runbooks/release-governance-reading-path.md +56 -0
  165. package/docs/runbooks/release-notes-automation.md +48 -0
  166. package/docs/runbooks/release-rollback-recovery-walkthrough.md +47 -0
  167. package/docs/runbooks/requirement-clarity-and-scope-walkthrough.md +46 -0
  168. package/docs/runbooks/reviewdog-pr-gates.md +49 -0
  169. package/docs/runbooks/role-prompt-recipes.md +130 -0
  170. package/docs/runbooks/rtk-integration-intake.md +45 -0
  171. package/docs/runbooks/rtk-token-optimization-usage.md +107 -0
  172. package/docs/runbooks/runner-egress-hardening.md +81 -0
  173. package/docs/runbooks/runtime-capabilities-overview.md +113 -0
  174. package/docs/runbooks/sbom-generation-gates.md +71 -0
  175. package/docs/runbooks/scorecard-supply-chain-gates.md +82 -0
  176. package/docs/runbooks/secret-scanning-gates.md +85 -0
  177. package/docs/runbooks/security-compliance-platform-demo-execution-log.md +36 -0
  178. package/docs/runbooks/security-compliance-platform-demo-script.md +49 -0
  179. package/docs/runbooks/security-compliance-platform-walkthrough.md +98 -0
  180. package/docs/runbooks/slsa-generator-patterns.md +73 -0
  181. package/docs/runbooks/slsa-verification-gates.md +75 -0
  182. package/docs/runbooks/solo-delivery-mode.md +142 -0
  183. package/docs/runbooks/solo-delivery-one-page.md +111 -0
  184. package/docs/runbooks/specialist-commands-playbook.md +85 -0
  185. package/docs/runbooks/sub-agent-invocation-map.md +144 -0
  186. package/docs/runbooks/system-architecture-design-walkthrough.md +49 -0
  187. package/docs/runbooks/team-closeout-example.md +73 -0
  188. package/docs/runbooks/team-command-output-contracts.md +358 -0
  189. package/docs/runbooks/team-commands-quick-prompts.md +125 -0
  190. package/docs/runbooks/team-execute-example.md +63 -0
  191. package/docs/runbooks/team-handoff-example.md +49 -0
  192. package/docs/runbooks/team-intake-example.md +70 -0
  193. package/docs/runbooks/team-plan-example.md +62 -0
  194. package/docs/runbooks/team-release-example.md +63 -0
  195. package/docs/runbooks/team-review-example.md +61 -0
  196. package/docs/runbooks/team-skills-test-run.md +184 -0
  197. package/docs/runbooks/team-skills-usage.md +336 -0
  198. package/docs/runbooks/team-training-reading-path.md +64 -0
  199. package/docs/runbooks/tech-lead-closure-conversation-example.md +78 -0
  200. package/docs/runbooks/tech-lead-daily-operations.md +67 -0
  201. package/docs/runbooks/trivy-security-gates.md +79 -0
  202. package/docs/runbooks/troubleshooting.md +234 -0
  203. package/docs/runbooks/vertical-scenario-capability-matrix.md +107 -0
  204. package/docs/runbooks/witness-policy-gates.md +78 -0
  205. package/docs/runbooks/zizmor-workflow-audits.md +81 -0
  206. package/manifests/install-components.json +8 -0
  207. package/manifests/install-modules.json +34 -0
  208. package/manifests/install-profiles.json +2 -0
  209. package/package.json +2 -1
  210. package/scripts/install-apply.js +9 -0
  211. package/scripts/install-open-design.js +206 -0
  212. package/scripts/install-plan.js +17 -0
  213. package/scripts/lib/install/apply.js +31 -0
  214. package/scripts/lib/install-executor.js +56 -0
  215. package/skills/open-design/SKILL.md +87 -0
  216. package/skills/open-design/agents/openai.yaml +4 -0
package/README.md CHANGED
@@ -66,6 +66,7 @@ TSP 整合了多个社区开源框架的精华能力,而非从零构建:
66
66
  | **BMAD** | 方法来源(已吸收) | 单入口主链(`/team-help`)、Requirement Challenge、Design Review、Implementation Readiness、Story Slice、`artifact:persist` 落盘、Release→Closeout 收口 |
67
67
  | **Graphify** | 社区(`safishamsi/graphify`) | 可选知识图谱能力(brownfield 结构扫描、依赖路径分析、架构问答证据),以 runbook + 本地 skill 接入,不替换 workflow-engine |
68
68
  | **GitNexus** | 社区(`abhigyanpatwari/GitNexus`) | 受控可选代码智能能力(MCP 查询、impact、detect_changes、多仓图谱证据),以 runbook + thin skill 接入,不内置依赖 |
69
+ | **Open Design** | 社区(`nexu-io/open-design`) | 受控可选设计工作台能力(本地优先原型、deck、dashboard、mobile flow、`DESIGN.md`、导出 artifact),以 runbook + thin skill 接入,不内置 daemon |
69
70
  | **GSD** (Getting Shit Done) | 社区 | Quality Gates 分类法(4 类 gate)、wave-execution、session-continuity、discuss-phase、quick-execution、context-engineering、workflow-forensics |
70
71
  | **gstack** | 社区 | brainstorming 苏格拉底式创意探索、cross-model-review 跨模型交叉审查、multi-perspective-review 多视角评审(CEO/Design/Eng/DevEx) |
71
72
  | **Superpowers** | 社区 | session-continuity 会话暂停/恢复、subagent-driven-development 子 agent 驱动开发、git-worktree-isolation 任务隔离 |
@@ -73,15 +74,19 @@ TSP 整合了多个社区开源框架的精华能力,而非从零构建:
73
74
  | **rtk** (Rust Token Killer) | 社区 | CLI 代理透明命令重写,60-90% token 节约,100+ 命令支持(git/gh/cargo/npm/docker/kubectl/aws) |
74
75
  | 社区贡献 | 多方 | UI/UX Pro Max 设计系统、vertical workflow examples、Santa Method 对抗验证、Ralph RFC Pipeline |
75
76
 
76
- ### 外部设计 skill 接入
77
+ ### 外部设计能力接入
77
78
 
78
- TSP 当前支持把第三方设计 skill 作为外部扩展能力接入使用,其中 [huashu-design](https://github.com/alchaincyf/huashu-design) 是与本平台协同性较高的一项:它偏向高保真设计产出,覆盖 HTML 原生交互原型、浏览器演讲幻灯片、时间轴动画、信息图与 5 维度设计评审;TSP 则负责团队协作、角色分工、handoff、quality gate 与发布收口。
79
+ TSP 当前支持把第三方设计能力作为外部扩展接入使用。设计工具负责高保真 artifact,TSP 负责团队协作、角色分工、handoff、quality gate 与发布收口。
79
80
 
80
- 推荐的组合方式是:让 TSP 管任务链路,让 huashu-design 管最终视觉和动效交付。典型场景包括:
81
+ [Open Design](https://github.com/nexu-io/open-design) 是当前已纳入安装面导航的受控可选设计工作台:它提供本地优先 web/daemon、coding-agent CLI 调度、31 个设计 skills、设计系统库、sandbox preview 与 HTML/PDF/PPTX/ZIP/MP4 等导出链路。TSP 侧落点是 [skills/open-design/SKILL.md](skills/open-design/SKILL.md) 与 [docs/runbooks/open-design-integration.md](docs/runbooks/open-design-integration.md),并通过 `design-prototyping` module 进入 `team` / `full` profile。其中 `full` profile 会自动执行 [scripts/install-open-design.js](scripts/install-open-design.js),把 Open Design clone/update 到 `~/.tsp/open-design`,并在 `corepack` / `pnpm` 可用时安装依赖。TSP 不 vendoring Open Design 源码、daemon、skills、design-systems 或 SQLite 数据。
81
82
 
82
- - `frontend-engineer` 先在 TSP 流程内收敛页面目标、状态边界和验收标准,再调用 huashu-design 做高保真原型或动画样机。
83
- - `tech-lead` / `qa-engineer` 在评审阶段把 huashu-design 产出的 HTML demo、deck 或动画作为 UI 证据补充到 review / release 资料里。
84
- - `product-manager` / `architect` 在方案讨论阶段借助 huashu-design 的设计方向顾问、设计评审与 deck 产出,加速多方案对比。
83
+ [huashu-design](https://github.com/alchaincyf/huashu-design) 仍作为文档级协同能力保留:它偏向高保真设计产出,覆盖 HTML 原生交互原型、浏览器演讲幻灯片、时间轴动画、信息图与 5 维度设计评审。
84
+
85
+ 推荐的组合方式是:让 TSP 管任务链路,让外部设计工具管最终视觉和动效交付。典型场景包括:
86
+
87
+ - `frontend-engineer` 先在 TSP 流程内收敛页面目标、状态边界和验收标准,再调用 Open Design 或 huashu-design 做高保真原型或动画样机。
88
+ - `tech-lead` / `qa-engineer` 在评审阶段把 HTML demo、deck、dashboard、mobile flow 或动画作为 UI 证据补充到 review / release 资料里。
89
+ - `product-manager` / `architect` 在方案讨论阶段借助设计方向顾问、设计评审与 deck 产出,加速多方案对比。
85
90
 
86
91
  上游安装方式:
87
92
 
@@ -204,6 +209,7 @@ TSP 的公开能力是在多个社区项目、技能仓库和工程方法论的
204
209
  | [affaan-m/everything-claude-code](https://github.com/affaan-m/everything-claude-code) | 上游能力来源 | ECC harness layer、specialist agents、skills、runtime hooks 与安装工具链的重要参考来源 |
205
210
  | [safishamsi/graphify](https://github.com/safishamsi/graphify) | 已吸收并本地化 | 为 brownfield 结构扫描、依赖路径分析与架构问答补充知识图谱能力 |
206
211
  | [abhigyanpatwari/GitNexus](https://github.com/abhigyanpatwari/GitNexus) | 受控可选接入 | 为 brownfield MCP 查询、impact、detect_changes 和多仓代码图谱证据提供外部能力;因许可证与 Node 20 要求,不内置依赖 |
212
+ | [nexu-io/open-design](https://github.com/nexu-io/open-design) | 受控可选接入 | 为本地优先原型、deck、dashboard、mobile flow、`DESIGN.md` 与可导出视觉 artifact 提供外部设计工作台;不内置 daemon 或上游源码 |
207
213
  | [rtk-ai/rtk](https://github.com/rtk-ai/rtk) | 已集成运行时能力 | CLI 透明命令重写与 token 优化能力来源 |
208
214
  | [VoltAgent/awesome-design-md](https://github.com/VoltAgent/awesome-design-md) | 设计资产来源 | 为 `DESIGN.md` 品牌风格扩展提供参考品牌库 |
209
215
  | [alchaincyf/huashu-design](https://github.com/alchaincyf/huashu-design) | 外部协同能力 | 提供高保真原型、HTML slides、动画与设计评审能力;当前仅做文档级接入说明,不内置分发 |
@@ -0,0 +1,199 @@
1
+ import { defineConfig } from 'vitepress'
2
+
3
+ export default defineConfig({
4
+ base: '/harness-engineering/',
5
+ title: 'Team Skills Platform',
6
+ description: 'Company-grade role-based Team Skills Platform for AI-assisted development',
7
+ lang: 'zh-CN',
8
+ ignoreDeadLinks: true,
9
+
10
+ head: [
11
+ ['link', { rel: 'icon', type: 'image/svg+xml', href: '/logo.svg' }],
12
+ ],
13
+
14
+ themeConfig: {
15
+ logo: '/logo.svg',
16
+ siteTitle: 'TSP',
17
+
18
+ nav: [
19
+ { text: '首页', link: '/' },
20
+ { text: '快速开始', link: '/runbooks/claude-quick-start' },
21
+ { text: '用户指南', link: '/runbooks/team-skills-usage' },
22
+ {
23
+ text: '手册',
24
+ items: [
25
+ { text: 'Claude 快速开始', link: '/runbooks/claude-quick-start' },
26
+ { text: 'Codex 快速开始', link: '/runbooks/codex-quick-start' },
27
+ { text: 'Cursor 快速开始', link: '/runbooks/cursor-quick-start' },
28
+ { text: 'OpenCode 快速开始', link: '/runbooks/opencode-quick-start' },
29
+ ],
30
+ },
31
+ ],
32
+
33
+ sidebar: {
34
+ '/guides/': [
35
+ {
36
+ text: '指南',
37
+ items: [
38
+ { text: '安装(入口页)', link: '/guides/installation' },
39
+ { text: '用户指南(入口页)', link: '/guides/user-guide' },
40
+ { text: '权威安装文档', link: '/runbooks/claude-quick-start' },
41
+ { text: '权威使用手册', link: '/runbooks/team-skills-usage' },
42
+ ],
43
+ },
44
+ ],
45
+ '/runbooks/': [
46
+ {
47
+ text: '快速开始',
48
+ collapsed: false,
49
+ items: [
50
+ { text: 'Claude 快速开始', link: '/runbooks/claude-quick-start' },
51
+ { text: 'Codex 快速开始', link: '/runbooks/codex-quick-start' },
52
+ { text: 'Cursor 快速开始', link: '/runbooks/cursor-quick-start' },
53
+ { text: 'OpenCode 快速开始', link: '/runbooks/opencode-quick-start' },
54
+ { text: '首次主链演练', link: '/runbooks/first-team-workflow-walkthrough' },
55
+ { text: '项目接入', link: '/runbooks/project-onboarding' },
56
+ { text: '故障排查', link: '/runbooks/troubleshooting' },
57
+ ],
58
+ },
59
+ {
60
+ text: '使用场景',
61
+ collapsed: true,
62
+ items: [
63
+ { text: 'Claude 使用场景', link: '/runbooks/claude-usage-scenarios' },
64
+ { text: 'Codex 使用场景', link: '/runbooks/codex-usage-scenarios' },
65
+ { text: 'Solo 交付模式', link: '/runbooks/solo-delivery-mode' },
66
+ { text: 'Solo 单页速查', link: '/runbooks/solo-delivery-one-page' },
67
+ ],
68
+ },
69
+ {
70
+ text: '提示与对话',
71
+ collapsed: true,
72
+ items: [
73
+ { text: 'Claude 提示模板', link: '/runbooks/claude-conversation-prompt-recipes' },
74
+ { text: 'Codex 并行模板', link: '/runbooks/codex-parallel-prompt-recipes' },
75
+ { text: '角色提示模板', link: '/runbooks/role-prompt-recipes' },
76
+ { text: 'Claude 完整对话示例', link: '/runbooks/claude-end-to-end-conversation-example' },
77
+ { text: 'Codex 完整对话示例', link: '/runbooks/codex-end-to-end-conversation-example' },
78
+ ],
79
+ },
80
+ {
81
+ text: '角色操作手册',
82
+ collapsed: true,
83
+ items: [
84
+ { text: '架构师日常', link: '/runbooks/architect-daily-operations' },
85
+ { text: '后端工程师日常', link: '/runbooks/backend-engineer-daily-operations' },
86
+ { text: '产品经理对话示例', link: '/runbooks/product-manager-clarification-conversation-example' },
87
+ { text: '项目经理规划示例', link: '/runbooks/project-manager-planning-conversation-example' },
88
+ { text: '架构师设计对话', link: '/runbooks/architect-design-conversation-example' },
89
+ { text: 'QA 评审对话', link: '/runbooks/qa-review-conversation-example' },
90
+ { text: 'DevOps 发布对话', link: '/runbooks/devops-release-conversation-example' },
91
+ { text: 'Tech Lead 收口对话', link: '/runbooks/tech-lead-closure-conversation-example' },
92
+ ],
93
+ },
94
+ {
95
+ text: '速查手册',
96
+ collapsed: true,
97
+ items: [
98
+ { text: '前端 Bugfix 速查', link: '/runbooks/frontend-bugfix-one-page' },
99
+ { text: '后端 API 交付速查', link: '/runbooks/backend-api-delivery-one-page' },
100
+ { text: '发布收口速查', link: '/runbooks/release-closure-one-page' },
101
+ ],
102
+ },
103
+ {
104
+ text: '平台能力',
105
+ collapsed: true,
106
+ items: [
107
+ { text: '命令与能力矩阵', link: '/runbooks/command-and-capability-matrix' },
108
+ { text: 'ECC Harness 使用', link: '/runbooks/ecc-harness-usage' },
109
+ { text: '运行时能力概览', link: '/runbooks/runtime-capabilities-overview' },
110
+ { text: 'Agent 治理', link: '/runbooks/agent-governance' },
111
+ { text: 'Sub-agent 调用地图', link: '/runbooks/sub-agent-invocation-map' },
112
+ { text: 'Handoff 治理', link: '/runbooks/handoff-governance' },
113
+ { text: '错误经验库', link: '/runbooks/error-experience-usage' },
114
+ { text: '并行执行', link: '/runbooks/parallel-execution-usage' },
115
+ { text: '进化系统', link: '/runbooks/evolution-usage' },
116
+ { text: 'Graphify 知识图谱', link: '/runbooks/graphify-knowledge-graph-usage' },
117
+ ],
118
+ },
119
+ {
120
+ text: '工程实践',
121
+ collapsed: true,
122
+ items: [
123
+ { text: 'Git PR 工作流', link: '/runbooks/git-pr-workflow' },
124
+ { text: 'AI PR Review 自动化', link: '/runbooks/ai-pr-review-automation' },
125
+ { text: '契约测试', link: '/runbooks/contract-testing-playbook' },
126
+ { text: 'Helm 单元测试', link: '/runbooks/helm-unittest-playbook' },
127
+ { text: '依赖更新自动化', link: '/runbooks/dependency-update-automation' },
128
+ { text: 'Release Notes 自动化', link: '/runbooks/release-notes-automation' },
129
+ ],
130
+ },
131
+ {
132
+ text: 'CI/CD 安全门禁',
133
+ collapsed: true,
134
+ items: [
135
+ { text: 'CodeQL 安全门禁', link: '/runbooks/codeql-pr-security-gates' },
136
+ { text: 'Secret Scanning', link: '/runbooks/secret-scanning-gates' },
137
+ { text: 'Trivy 安全扫描', link: '/runbooks/trivy-security-gates' },
138
+ { text: 'Checkov IaC 门禁', link: '/runbooks/checkov-iac-gates' },
139
+ { text: 'API Lint 门禁', link: '/runbooks/api-lint-gates' },
140
+ { text: 'API Breaking Change', link: '/runbooks/api-breaking-change-gates' },
141
+ { text: 'Reviewdog PR 门禁', link: '/runbooks/reviewdog-pr-gates' },
142
+ { text: 'Scorecard 供应链', link: '/runbooks/scorecard-supply-chain-gates' },
143
+ { text: 'SBOM 生成', link: '/runbooks/sbom-generation-gates' },
144
+ { text: 'Cosign 签名', link: '/runbooks/cosign-signing-gates' },
145
+ { text: 'SLSA 验证', link: '/runbooks/slsa-verification-gates' },
146
+ ],
147
+ },
148
+ {
149
+ text: '企业扩展',
150
+ collapsed: true,
151
+ items: [
152
+ { text: '企业扩展快速开始', link: '/runbooks/enterprise-extension-quick-start' },
153
+ { text: '公司技能集成', link: '/runbooks/company-skills-integration' },
154
+ { text: '公司技能决策框架', link: '/runbooks/company-skills-decision-framework' },
155
+ { text: '外部能力引入', link: '/runbooks/external-capability-intake' },
156
+ { text: '企业标准映射', link: '/runbooks/enterprise-standards-mapping' },
157
+ ],
158
+ },
159
+ {
160
+ text: '历史档案',
161
+ collapsed: true,
162
+ items: [
163
+ { text: '版本收口(2026-03-29)', link: '/runbooks/version-closure-2026-03-29' },
164
+ { text: '平台能力演示剧本(历史)', link: '/runbooks/platform-capability-demo-script' },
165
+ { text: '平台能力演示执行记录(历史)', link: '/runbooks/platform-capability-demo-execution-log' },
166
+ { text: '平台迁移计划(历史)', link: '/plans/team-skills-platform-migration' },
167
+ ],
168
+ },
169
+ ],
170
+ },
171
+
172
+ search: {
173
+ provider: 'local',
174
+ },
175
+
176
+ socialLinks: [
177
+ { icon: 'github', link: 'https://github.com/Colin4k1024/harness-engineering' },
178
+ ],
179
+
180
+ footer: {
181
+ message: 'Released under the MIT License.',
182
+ copyright: '© 2026 Team Skills Platform',
183
+ },
184
+
185
+ outline: {
186
+ level: [2, 3],
187
+ label: '页面导航',
188
+ },
189
+
190
+ lastUpdated: {
191
+ text: '最后更新',
192
+ },
193
+
194
+ docFooter: {
195
+ prev: '上一页',
196
+ next: '下一页',
197
+ },
198
+ },
199
+ })
@@ -0,0 +1,33 @@
1
+ ---
2
+ artifact: adr
3
+ task: doc-architecture-demo
4
+ date: 2026-04-01
5
+ role: architect
6
+ status: draft
7
+ ---
8
+
9
+ # ADR-001: Doc Architecture Integration
10
+
11
+ ## 决策信息
12
+ - 编号:ADR-001
13
+ - 标题:将文档架构能力集成到 team 主链
14
+ - 状态:Accepted
15
+ - 决策日期:2026-04-01
16
+ - Owner:architect
17
+
18
+ ## 背景与约束
19
+ - 背景:原能力存在独立文档目录习惯,当前平台要求统一 artifact 持久化。
20
+ - 约束:不得新增并行主命令,不得破坏角色边界。
21
+
22
+ ## 备选方案
23
+ - 方案 A:新增 team-doc-architect 命令。
24
+ - 方案 B:角色内嵌 shared skill + 现有命令契约增强。
25
+ - 结论:采用方案 B。
26
+
27
+ ## 决策结果
28
+ - 采用 `doc-architecture` shared skill。
29
+ - 通过 artifact-standards 与输出契约扩展保证落盘一致性。
30
+ - 回退思路:若复杂度过高,先仅在 tech-lead/architect 启用。
31
+
32
+ ## 后续动作
33
+ - 在后续任务中持续验证主链闭环。
@@ -0,0 +1,5 @@
1
+ # Guides Index
2
+
3
+ | 文档 | 说明 |
4
+ |------|------|
5
+ | [user-guide.md](user-guide.md) | 完整用户指南,包含项目概述、快速开始、核心概念、命令详解等 |
@@ -0,0 +1,33 @@
1
+ ---
2
+ version: "2.3.0"
3
+ status: active
4
+ created: 2026-04-17
5
+ updated: 2026-04-18
6
+ owner: 工程团队
7
+ doc_tier: redirect
8
+ last_verified: 2026-04-18
9
+ source_of_truth:
10
+ - ../runbooks/claude-quick-start.md
11
+ - ../runbooks/codex-quick-start.md
12
+ - ../runbooks/project-onboarding.md
13
+ - ../runbooks/troubleshooting.md
14
+ ---
15
+
16
+ # 安装指南(入口页)
17
+
18
+ 本页作为**薄入口**保留,不再承载安装细节。安装与排障的权威内容统一维护在 runbooks:
19
+
20
+ - Claude:[/runbooks/claude-quick-start](../runbooks/claude-quick-start.md)
21
+ - Codex:[/runbooks/codex-quick-start](../runbooks/codex-quick-start.md)
22
+ - Cursor:[/runbooks/cursor-quick-start](../runbooks/cursor-quick-start.md)
23
+ - OpenCode:[/runbooks/opencode-quick-start](../runbooks/opencode-quick-start.md)
24
+ - 项目接入:[/runbooks/project-onboarding](../runbooks/project-onboarding.md)
25
+ - 故障排查:[/runbooks/troubleshooting](../runbooks/troubleshooting.md)
26
+
27
+ ## 最短路径
28
+
29
+ 1. 选择目标平台 quick start。
30
+ 2. 完成安装后先执行 `/team-help`。
31
+ 3. 正式任务产出按 `artifact:persist` 写回项目仓库。
32
+
33
+ > 说明:为避免入口层双写漂移,本页不再维护版本 tarball、脚本细节与示例命令。
@@ -0,0 +1,36 @@
1
+ ---
2
+ version: "2.3.0"
3
+ status: active
4
+ created: 2026-04-17
5
+ updated: 2026-04-18
6
+ owner: 工程团队
7
+ doc_tier: redirect
8
+ last_verified: 2026-04-18
9
+ source_of_truth:
10
+ - ../runbooks/team-skills-usage.md
11
+ - ../runbooks/command-and-capability-matrix.md
12
+ - ../runbooks/project-onboarding.md
13
+ - ../runbooks/troubleshooting.md
14
+ ---
15
+
16
+ # 用户指南(入口页)
17
+
18
+ 本页作为**薄入口**保留,不再维护完整使用手册正文。权威内容统一在 runbooks:
19
+
20
+ - 主手册:[/runbooks/team-skills-usage](../runbooks/team-skills-usage.md)
21
+ - 命令矩阵:[/runbooks/command-and-capability-matrix](../runbooks/command-and-capability-matrix.md)
22
+ - 项目接入:[/runbooks/project-onboarding](../runbooks/project-onboarding.md)
23
+ - 故障排查:[/runbooks/troubleshooting](../runbooks/troubleshooting.md)
24
+
25
+ ## 推荐阅读顺序
26
+
27
+ 1. `/runbooks/command-and-capability-matrix`
28
+ 2. `/runbooks/team-skills-usage`
29
+ 3. `/runbooks/project-onboarding`
30
+ 4. `/runbooks/troubleshooting`
31
+
32
+ ## 主链入口提醒
33
+
34
+ 无论从哪份手册进入,正式任务都建议先执行 `/team-help` 判断主链入口,再进入 `/team-intake` 或 `/quick`。
35
+
36
+ > 说明:本页不再维护重复命令清单与安装示例,避免与 runbooks 并行漂移。
package/docs/index.md ADDED
@@ -0,0 +1,65 @@
1
+ ---
2
+ layout: home
3
+ status: active
4
+ updated: 2026-04-17
5
+ owner: 工程团队
6
+ doc_tier: entry
7
+ last_verified: 2026-04-17
8
+ source_of_truth:
9
+ - ../README.md
10
+ - ../AGENTS.md
11
+ - ./runbooks/command-and-capability-matrix.md
12
+
13
+ hero:
14
+ name: Team Skills Platform
15
+ text: 角色化 AI 研发团队
16
+ tagline: 200+ Skills · 8 Roles · 多平台支持 — 内置 /team-help、brownfield 补齐与 story-sized execution
17
+ actions:
18
+ - theme: brand
19
+ text: 快速开始
20
+ link: /runbooks/claude-quick-start
21
+ - theme: alt
22
+ text: 用户指南
23
+ link: /runbooks/team-skills-usage
24
+ - theme: alt
25
+ text: GitHub
26
+ link: https://github.com/Colin4k1024/tsp
27
+
28
+ features:
29
+ - icon: 🧭
30
+ title: 单入口主链
31
+ details: 通过 /team-help 统一判断当前应该进入 intake、plan 还是先补 brownfield 文档、project context 与 readiness 证据,降低首次接入成本。
32
+ - icon: 👥
33
+ title: 角色化协作
34
+ details: Tech Lead 编排 + 产品、架构、前端、后端、测试、运维 8 个专业角色协作,每个角色有明确的职责边界和交接规则。
35
+ - icon: ⚡
36
+ title: 200+ AI Skills
37
+ details: 覆盖调试验证、编排效率、学习记忆、进化复用等多个维度,包含前端工程、后端模式、安全审查、TDD 等专项能力。
38
+ - icon: 🔧
39
+ title: 多平台支持
40
+ details: 一套定义,同时支持 Claude Code、Codex、Cursor、OpenCode 四个平台,通过 npx 一键安装。
41
+ - icon: 🏗️
42
+ title: ECC Harness Layer
43
+ details: Specialist agents、快捷命令、分层规则、运行时 hooks、上下文管理、成本追踪与持续学习等工程增强能力。
44
+ - icon: 🧩
45
+ title: BMAD 经验吸收
46
+ details: 已吸收 implementation-readiness、brownfield/document-project、Story Slice Plan 与阶段化 readiness gate,不额外引入第二套平行框架。
47
+ - icon: 🕸️
48
+ title: Graphify 可选能力
49
+ details: 提供 brownfield 结构扫描、依赖路径分析与架构问答证据,作为 `/team-*` 主链的可选补充而非替代引擎。
50
+ - icon: 🎨
51
+ title: 前端治理
52
+ details: React/Next 优先的工程规范 + UI/UX 设计知识库 + 强制质量门禁,从 intake 到 release 全链路覆盖。
53
+ - icon: 🏢
54
+ title: 企业级扩展
55
+ details: 支持公司领域 Skill 扩展、企业架构治理、组件基线、数据库/缓存规范,可对齐集团标准。
56
+ ---
57
+
58
+ ## 文档生命周期字段
59
+
60
+ | 字段 | 值 | 说明 |
61
+ |------|----|------|
62
+ | `status` | `active` | 首页属于公开入口,作为文档导航首屏 |
63
+ | `doc_tier` | `entry` | 入口层文档;不承载与 runbooks 重复的细节说明 |
64
+ | `last_verified` | `2026-04-17` | 最近一次入口链路回归核对日期 |
65
+ | `source_of_truth` | `README.md`、`AGENTS.md`、`runbooks/command-and-capability-matrix.md` | 命令面与入口说明的权威来源 |
@@ -0,0 +1,10 @@
1
+ # Backlog
2
+
3
+ ## 2026-04-21 - Community adoption follow-up
4
+
5
+ The release-blocking open-source readiness gaps from this round are closed: public installer surface is aligned, support depth is now documented, CI enforces the declared doc/reference gates, and prebuilt binaries are treated as release/package artifacts instead of git-tracked assets.
6
+
7
+ Next follow-ups are now narrowly adoption-oriented:
8
+
9
+ - Add minimal quick-start stubs or examples for `antigravity`, `codebuddy`, `gemini`, `copilot`, `windsurf`, and `augment` if community usage warrants it.
10
+ - Keep the new `npm run release:health` summary and `open-source-release-checklist.md` aligned with future support-surface changes.
@@ -0,0 +1,43 @@
1
+ # Decisions Log
2
+
3
+ > 追加策略:每次重大技术决策由主责角色追加,不覆盖历史行。格式:`## YYYY-MM-DD · {标题}`
4
+
5
+ ---
6
+
7
+ ## 2026-03-31 · Memory Persistence Hook 从被动等待改为主动采集
8
+
9
+ **背景**:原 `session_end.py` 设计依赖 Claude 发送结构化 JSON 结束信号,实际上 Claude 从不发送该信号,导致 hook 永远处于等待状态,经验数据完全不落盘。
10
+
11
+ **决策**:彻底重写 `session_end.py`,改为主动通过 `git log -15` + `docs/memory/` 文件读取的方式采集经验;`session_start.py` 改为双源加载(repo docs/memory/ + ~/.claude/memory/ 经验文件)。
12
+
13
+ **影响**:hook 现在不依赖 Claude 输出行为,在任何 session end 场景下都能落盘。Smoke test 验证:15 commits、61 experience capsules。待确认:session_start payload 中 repo_memory_files_found 是否非空(依赖 docs/memory/ 存在)。
14
+
15
+ ---
16
+
17
+ ## 2026-03-31 · Observability Trace Skill 归入公共增强层
18
+
19
+ **背景**:通用编码链路追踪能力不绑定特定业务系统,适合作为编码类 skill 的公共增强。
20
+
21
+ **决策**:将这类能力保留在公开增强层,配置给 backend-engineer 和 qa-engineer,而不是作为私有企业扩展的默认依赖。
22
+
23
+ **影响**:公共仓可以保留通用可观测性说明,而不会默认暴露私有企业发布或权限集成。
24
+
25
+ ---
26
+
27
+ ## 2026-04-01 · 文档架构能力并入 team 主链
28
+
29
+ **背景**:需要将 eags 风格 discovery/modeling/audit 能力与当前 team-skills 体系整合,并避免并行文档目录。
30
+
31
+ **决策**:新增 `skills/doc-architecture`,并将其装配到 tech-lead、architect、backend-engineer、frontend-engineer、qa-engineer、devops-engineer。同步增强 artifact 标准、输出契约与模板字段。
32
+
33
+ **影响**:主链可直接产出并追溯 PRD、Plan、Arch、API、Execute、Test、Release、ADR、Session Summary,不需要新增 team 主命令。
34
+
35
+ ---
36
+
37
+ ## 2026-04-02 · Internal Workflow and Permission Case Studies Moved Out of Public Memory
38
+
39
+ **背景**:部分记忆条目直接绑定私有流程系统、权限中心和组织专属案例,不适合作为公开仓的默认示例。
40
+
41
+ **决策**:将组织专属案例从公开记忆层移出,只保留公共运行时、文档架构和平台演进相关的决策记录;私有案例转入 `enterprise` overlay 或私有仓。
42
+
43
+ **影响**:公开仓的 `docs/memory/` 更适合外部用户阅读,也不会把私有领域案例误当成公共默认路径。
@@ -0,0 +1,87 @@
1
+ # Lessons Learned
2
+
3
+ > 追加策略:由 qa-engineer 或 devops-engineer 在遇到可复用经验时追加。格式:`## YYYY-MM-DD · {标题}`
4
+
5
+ ---
6
+
7
+ ## 2026-03-31 · session_end hook 不能依赖 LLM 主动发送结构化数据
8
+
9
+ **场景**:设计 memory persistence hook 时,原方案希望 Claude 在会话结束时输出结构化 JSON 摘要,由 hook 捕获存盘。
10
+
11
+ **问题**:Claude 不会主动向 hook stdin 发送任何结束信号。`while True: line = sys.stdin.readline()` 会永久阻塞,数据永远不落盘。
12
+
13
+ **建议**:session_end hook 应主动采集,而非被动等待。可读 git log、docs/memory/、changelog 等文件作为替代数据源;不依赖 LLM 在特定格式下产出数据。
14
+
15
+ ---
16
+
17
+ ## 2026-03-31 · _experience.json 文件格式与 SessionSummary dataclass 不兼容
18
+
19
+ **场景**:session_end.py 在同一目录写入 `{id}.json`(SessionSummary)和 `{id}_experience.json`(experience capsules)两类文件。
20
+
21
+ **问题**:`memory_store.py` 的 `load_latest_session_summary` 按日期排序读取所有 `.json` 文件,`_experience.json` 字段不同会导致 dataclass 反序列化失败。
22
+
23
+ **建议**:在 `load_latest_session_summary` 中添加 `if "_experience" in session_file.name: continue` 跳过逻辑;或将两类文件放入不同子目录隔离。
24
+
25
+ ---
26
+
27
+ ## 2026-03-31 · docs/memory/ 目录必须提前初始化
28
+
29
+ **场景**:session_start.py 重写后从 `docs/memory/project-context.md` 等文件读取上下文,但该目录在项目初始未被创建。
30
+
31
+ **问题**:`repo_memory_files_found: []` — session_start 运行无报错但上下文为空,导致新会话缺少项目感知。
32
+
33
+ **建议**:每个接入项目在首次配置 hooks 后,应立即创建 `docs/memory/project-context.md`、`decisions.md`、`lessons-learned.md` 三个基础文件,哪怕内容只是占位符。可在 `docs/runbooks/project-onboarding.md` 中补充这一步骤。
34
+
35
+ ---
36
+
37
+ ## 2026-03-31 · validate_library.py(当前等价 `node scripts/validate-library.js`)的 REQUIRED_ECC_SKILLS 必须与目录同步更新
38
+
39
+ **场景**:新增 ecc skill(langfuse-coding-trace、harness-audit)后,历史脚本 `validate_library.py`(当前等价 `node scripts/validate-library.js`)仍只检查旧的 9 个 required skills。
40
+
41
+ **问题**:新 skill 存在但不被验证,脚本 PASS 不代表实际完整性正确。
42
+
43
+ **建议**:每次在 `skills/` 新增 skill 时,同步更新 `scripts/validate-library.js` 中的 `REQUIRED_ECC_SKILLS` 列表;可考虑改为动态读取 `skills/` 自动发现。
44
+
45
+ ---
46
+
47
+ ## 2026-04-01 · 文档能力整合优先走 artifacts 映射而非新目录
48
+
49
+ **场景**:将外部文档能力并入 team-skills 主链。
50
+
51
+ **问题**:若沿用外部目录习惯(architecture/domains/specs/plans/runbooks),会与现有 artifact 持久化规则冲突,导致事实源分裂。
52
+
53
+ **建议**:统一把 discovery/modeling/audit 内容映射到现有 artifact 文件;新增能力优先通过 shared skill + 输出契约字段扩展落地,不新增并行主命令或并行主目录。
54
+
55
+ ---
56
+
57
+ ## 2026-04-09 · skills 子目录嵌套会阻止 Claude 自动发现
58
+
59
+ **场景**:skills 按 shared/ecc/company 三层子目录组织(`skills/shared/api-contract`、`skills/ecc/eval-harness`),认为逻辑清晰。
60
+
61
+ **问题**:Claude Code 只自动扫描 `skills/` 根目录下的 SKILL.md,嵌套一层子目录后 skill 不会出现在可用列表中,用户无法直接使用。
62
+
63
+ **建议**:所有 skill 统一平铺到 `skills/` 根目录。分类信息通过 manifest module 和 role.yaml 的 recommended_*_skills 字段表达,不依赖目录层级。
64
+
65
+ ---
66
+
67
+ ## 2026-04-09 · JSON 文件中的 Unicode 弯引号导致解析歧义
68
+
69
+ **场景**:`roles/architect/role.yaml` 中 outputs 字段包含中文弯引号 `"前端页面"`,Node.js `JSON.parse` 报错 position 422。
70
+
71
+ **问题**:中文弯引号 U+201C/U+201D 在 JSON 字符串内部时,解析器会在遇到 ASCII `"` 时错误地判断字符串终止位置。
72
+
73
+ **建议**:JSON 文件内的中文引号统一使用 `「」` 或反斜杠转义。CI 可加一条 pre-commit 检查,检测 .yaml/.json 中的 U+201C/U+201D。
74
+
75
+ ---
76
+
77
+ ## 2026-04-09 · 批量路径替换后验证脚本需单独修复
78
+
79
+ **场景**:sed 全局替换了 `skills/shared/` → `skills/`,文档和配置文件正确,但验证脚本内部 `path.join(root, "skills", "shared", skill)` 硬编码路径未被替换。
80
+
81
+ **问题**:验证脚本通过 string concat 拼路径,sed 替换 markdown/JSON 中的字符串引用时不会命中 JS/Python 代码中的路径拼接。
82
+
83
+ **建议**:路径替换后,必须单独 grep 验证脚本中的硬编码子目录名(如 `"shared"`、`"ecc"`、`"company"`),并运行验证脚本确认。
84
+
85
+ ## 2026-04-17 - BMAD Source Adoption v2.2 closeout
86
+
87
+ Closeout readiness for /team-closeout depends on complete artifacts (prd, delivery-plan, handoff, execute/review/release docs). When continuing from mid-stream work, run a gate completeness check early to avoid end-stage backfill.