@colin4k1024/tsp 2.4.4 → 2.4.6
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.
- package/README.md +16 -20
- package/bin/lib/install-surface.js +3 -3
- package/bin/lib/source-installer.js +2 -2
- package/commands/team-help.md +2 -2
- package/commands/team-plan.md +1 -1
- package/commands/update-codemaps.md +3 -3
- package/manifests/install-components.json +1 -1
- package/manifests/install-modules.json +17 -3
- package/manifests/install-profiles.json +2 -0
- package/package.json +6 -3
- package/schemas/ecc-install-config.schema.json +6 -1
- package/schemas/install-modules.schema.json +4 -1
- package/scripts/codegraph-preflight.js +179 -0
- package/scripts/gitnexus-preflight.js +8 -0
- package/scripts/install-apply.js +10 -8
- package/scripts/install-codegraph.js +158 -0
- package/scripts/install-plan.js +28 -11
- package/scripts/lib/install/apply.js +256 -5
- package/scripts/lib/install/request.js +3 -2
- package/scripts/lib/install-audit-manifest.js +3 -0
- package/scripts/lib/install-executor.js +14 -5
- package/scripts/lib/install-lifecycle.js +2 -2
- package/scripts/lib/install-manifests.js +23 -4
- package/scripts/lib/install-targets/codex-home.js +187 -1
- package/scripts/lib/install-targets/opencode-home.js +135 -2
- package/scripts/lib/install-targets/registry.js +23 -1
- package/scripts/lib/release-health.js +19 -4
- package/scripts/lib/team-skills-data.json +6 -6
- package/scripts/release-health-summary.js +1 -1
- package/scripts/workflow-help.js +3 -3
- package/skills/codegraph/SKILL.md +57 -0
- package/skills/codegraph/agents/openai.yaml +4 -0
- package/docs/.vitepress/config.mts +0 -199
- package/docs/adr/ADR-001-doc-architecture-integration.md +0 -33
- package/docs/guides/README.md +0 -5
- package/docs/guides/installation.md +0 -33
- package/docs/guides/user-guide.md +0 -36
- package/docs/index.md +0 -65
- package/docs/memory/backlog.md +0 -10
- package/docs/memory/decisions.md +0 -43
- package/docs/memory/lessons-learned.md +0 -87
- package/docs/plans/2026-04-03-python-remnants-audit.md +0 -265
- package/docs/plans/2026-04-03-scripts-python-to-js-migration.md +0 -372
- package/docs/plans/2026-04-03-solo-delivery-execution-checklist.md +0 -413
- package/docs/plans/2026-04-03-solo-delivery-gap-plan.md +0 -377
- package/docs/plans/2026-04-03-team-skills-workflow-gates.md +0 -548
- package/docs/plans/2026-04-21-open-source-readiness-gap-plan.md +0 -217
- package/docs/plans/llm-surface-reduction-audit.md +0 -147
- package/docs/plans/llm-surface-reduction-execution-checklist.md +0 -217
- package/docs/plans/llm-surface-reduction-execution-history.md +0 -124
- package/docs/plans/team-skills-platform-migration.md +0 -54
- package/docs/presentation/README.md +0 -42
- package/docs/presentation/audience-presentation-route-map.md +0 -84
- package/docs/presentation/executive-briefing-talk-track.md +0 -50
- package/docs/presentation/generate_capability_matrix.py +0 -396
- package/docs/presentation/generate_ppt.py +0 -354
- package/docs/presentation/implementation-onboarding-brief.md +0 -38
- package/docs/presentation/presentation-talk-track.md +0 -97
- package/docs/presentation/vertical-scenario-route-map.md +0 -99
- package/docs/presentation/workshop-facilitator-guide.md +0 -47
- package/docs/runbooks/actionlint-workflow-gates.md +0 -80
- package/docs/runbooks/agent-governance.md +0 -131
- package/docs/runbooks/ai-eval-platform-demo-execution-log.md +0 -147
- package/docs/runbooks/ai-eval-platform-demo-script.md +0 -136
- package/docs/runbooks/ai-eval-platform-walkthrough.md +0 -113
- package/docs/runbooks/ai-pr-review-automation.md +0 -56
- package/docs/runbooks/api-breaking-change-gates.md +0 -58
- package/docs/runbooks/api-design-evolution-walkthrough.md +0 -42
- package/docs/runbooks/api-lint-gates.md +0 -57
- package/docs/runbooks/api-mocking-strategy-and-lifecycle-guide.md +0 -47
- package/docs/runbooks/architect-daily-operations.md +0 -63
- package/docs/runbooks/architect-design-conversation-example.md +0 -83
- package/docs/runbooks/artifact-attestation-gates.md +0 -75
- package/docs/runbooks/artifact-persistence.md +0 -257
- package/docs/runbooks/backend-engineer-daily-operations.md +0 -63
- package/docs/runbooks/batch-optimization-completion-checklist.md +0 -104
- package/docs/runbooks/biz-service-designer-end-to-end-conversation-example.md +0 -5
- package/docs/runbooks/biz-service-designer-toolkit.md +0 -5
- package/docs/runbooks/bug-fix-complete-walkthrough.md +0 -60
- package/docs/runbooks/build-failure-recovery-walkthrough.md +0 -40
- package/docs/runbooks/canary-decision-matrix.md +0 -41
- package/docs/runbooks/canary-staging-release-walkthrough.md +0 -46
- package/docs/runbooks/checkov-iac-gates.md +0 -104
- package/docs/runbooks/claude-code-review-workflow.md +0 -72
- package/docs/runbooks/claude-conversation-prompt-recipes.md +0 -132
- package/docs/runbooks/claude-end-to-end-conversation-example.md +0 -198
- package/docs/runbooks/claude-feature-development-guide.md +0 -112
- package/docs/runbooks/claude-quick-start.md +0 -227
- package/docs/runbooks/claude-usage-scenarios.md +0 -176
- package/docs/runbooks/code-review-collaboration-walkthrough.md +0 -65
- package/docs/runbooks/codeql-pr-security-gates.md +0 -64
- package/docs/runbooks/codex-end-to-end-conversation-example.md +0 -166
- package/docs/runbooks/codex-multi-agent-orchestration.md +0 -65
- package/docs/runbooks/codex-parallel-prompt-recipes.md +0 -131
- package/docs/runbooks/codex-quick-start.md +0 -223
- package/docs/runbooks/codex-usage-scenarios.md +0 -168
- package/docs/runbooks/codex-workflow-essentials.md +0 -88
- package/docs/runbooks/command-and-capability-matrix.md +0 -162
- package/docs/runbooks/conftest-policy-gates.md +0 -84
- package/docs/runbooks/consumer-driven-contract-testing-with-mock-alignment.md +0 -45
- package/docs/runbooks/contract-testing-playbook.md +0 -78
- package/docs/runbooks/cosign-signing-gates.md +0 -71
- package/docs/runbooks/cross-role-issue-triage-walkthrough.md +0 -47
- package/docs/runbooks/cursor-quick-start.md +0 -123
- package/docs/runbooks/custom-overlay.md +0 -115
- package/docs/runbooks/data-ml-pipeline-demo-execution-log.md +0 -141
- package/docs/runbooks/data-ml-pipeline-demo-script.md +0 -102
- package/docs/runbooks/data-ml-pipeline-walkthrough.md +0 -119
- package/docs/runbooks/data-observability-quality-demo-execution-log.md +0 -36
- package/docs/runbooks/data-observability-quality-demo-script.md +0 -42
- package/docs/runbooks/data-observability-quality-walkthrough.md +0 -86
- package/docs/runbooks/demo-deliverables-overview.md +0 -278
- package/docs/runbooks/demo-execution-log.md +0 -530
- package/docs/runbooks/demo-scenario.md +0 -129
- package/docs/runbooks/dependency-review-gates.md +0 -63
- package/docs/runbooks/dependency-update-automation.md +0 -83
- package/docs/runbooks/design-md-workflow.md +0 -185
- package/docs/runbooks/devops-engineer-daily-operations.md +0 -60
- package/docs/runbooks/devops-release-conversation-example.md +0 -88
- package/docs/runbooks/doc-architecture-integration.md +0 -59
- package/docs/runbooks/doc-architecture-quick-start.md +0 -122
- package/docs/runbooks/document-execution-audit.md +0 -32
- package/docs/runbooks/documentation-update-walkthrough.md +0 -37
- package/docs/runbooks/ecc-harness-usage.md +0 -93
- package/docs/runbooks/error-experience-usage.md +0 -116
- package/docs/runbooks/evolution-usage.md +0 -162
- package/docs/runbooks/executive-value-one-page.md +0 -55
- package/docs/runbooks/external-capability-approval-and-enablement-workflow.md +0 -39
- package/docs/runbooks/external-capability-intake.md +0 -160
- package/docs/runbooks/first-team-command-60-seconds.md +0 -96
- package/docs/runbooks/first-team-workflow-walkthrough.md +0 -245
- package/docs/runbooks/frontend-backend-integration-acceptance-checklist.md +0 -46
- package/docs/runbooks/frontend-backend-parallel-integration-walkthrough.md +0 -48
- package/docs/runbooks/frontend-bugfix-one-page.md +0 -82
- package/docs/runbooks/frontend-engineer-daily-operations.md +0 -60
- package/docs/runbooks/frontend-enterprise-style-profile.md +0 -5
- package/docs/runbooks/frontend-governance.md +0 -47
- package/docs/runbooks/frontend-refactor-walkthrough.md +0 -42
- package/docs/runbooks/git-pr-workflow.md +0 -63
- package/docs/runbooks/github-actions-supply-chain-demo-execution-log.md +0 -158
- package/docs/runbooks/github-actions-supply-chain-demo-script.md +0 -150
- package/docs/runbooks/github-actions-supply-chain-walkthrough.md +0 -117
- package/docs/runbooks/github-token-permissions-baseline.md +0 -92
- package/docs/runbooks/gitlab-manual-pipeline-release.md +0 -5
- package/docs/runbooks/gitlab-release-integration-playbook.md +0 -5
- package/docs/runbooks/gitnexus-code-intelligence-usage.md +0 -133
- package/docs/runbooks/graphify-knowledge-graph-usage.md +0 -88
- package/docs/runbooks/handoff-filling-guide-with-examples.md +0 -70
- package/docs/runbooks/handoff-governance.md +0 -250
- package/docs/runbooks/helm-unittest-playbook.md +0 -101
- package/docs/runbooks/hotfix-emergency-release-walkthrough.md +0 -60
- package/docs/runbooks/iac-kubernetes-platform-demo-execution-log.md +0 -144
- package/docs/runbooks/iac-kubernetes-platform-demo-script.md +0 -130
- package/docs/runbooks/iac-kubernetes-platform-walkthrough.md +0 -120
- package/docs/runbooks/implementation-onboarding-reading-path.md +0 -67
- package/docs/runbooks/in-toto-attestation-framework.md +0 -94
- package/docs/runbooks/incident-severity-triage-tree.md +0 -43
- package/docs/runbooks/incident-triage-one-page.md +0 -65
- package/docs/runbooks/internal-developer-platform-demo-execution-log.md +0 -36
- package/docs/runbooks/internal-developer-platform-demo-script.md +0 -42
- package/docs/runbooks/internal-developer-platform-walkthrough.md +0 -91
- package/docs/runbooks/karpathy-guidelines-usage.md +0 -27
- package/docs/runbooks/kubeconform-schema-gates.md +0 -100
- package/docs/runbooks/kubectl-server-dry-run-gates.md +0 -103
- package/docs/runbooks/kyverno-policy-gates.md +0 -90
- package/docs/runbooks/langfuse-and-observability-integration-guide.md +0 -43
- package/docs/runbooks/langfuse-coding-trace.md +0 -44
- package/docs/runbooks/mobile-miniapp-delivery-walkthrough.md +0 -112
- package/docs/runbooks/mobile-miniapp-demo-execution-log.md +0 -139
- package/docs/runbooks/mobile-miniapp-demo-script.md +0 -129
- package/docs/runbooks/multi-service-backend-integration-walkthrough.md +0 -61
- package/docs/runbooks/open-design-integration.md +0 -163
- package/docs/runbooks/open-source-release-checklist.md +0 -90
- package/docs/runbooks/opencode-quick-start.md +0 -128
- package/docs/runbooks/parallel-development-coordination-walkthrough.md +0 -47
- package/docs/runbooks/parallel-execution-usage.md +0 -179
- package/docs/runbooks/platform-capability-demo-execution-log.md +0 -184
- package/docs/runbooks/platform-capability-demo-script.md +0 -192
- package/docs/runbooks/plugin-extension-platform-demo-execution-log.md +0 -136
- package/docs/runbooks/plugin-extension-platform-demo-script.md +0 -102
- package/docs/runbooks/plugin-extension-platform-walkthrough.md +0 -111
- package/docs/runbooks/policy-controller-gates.md +0 -75
- package/docs/runbooks/post-rollback-verification-checklist.md +0 -37
- package/docs/runbooks/pre-release-checklist.md +0 -50
- package/docs/runbooks/product-manager-clarification-conversation-example.md +0 -90
- package/docs/runbooks/product-manager-daily-operations.md +0 -60
- package/docs/runbooks/production-incident-response-walkthrough.md +0 -50
- package/docs/runbooks/project-claude-design-rationale.md +0 -188
- package/docs/runbooks/project-manager-daily-operations.md +0 -61
- package/docs/runbooks/project-manager-planning-conversation-example.md +0 -82
- package/docs/runbooks/project-onboarding.md +0 -452
- package/docs/runbooks/qa-engineer-daily-operations.md +0 -63
- package/docs/runbooks/qa-review-conversation-example.md +0 -87
- package/docs/runbooks/release-closure-one-page.md +0 -65
- package/docs/runbooks/release-governance-reading-path.md +0 -56
- package/docs/runbooks/release-notes-automation.md +0 -48
- package/docs/runbooks/release-rollback-recovery-walkthrough.md +0 -47
- package/docs/runbooks/requirement-clarity-and-scope-walkthrough.md +0 -46
- package/docs/runbooks/reviewdog-pr-gates.md +0 -49
- package/docs/runbooks/role-prompt-recipes.md +0 -130
- package/docs/runbooks/rtk-integration-intake.md +0 -45
- package/docs/runbooks/rtk-token-optimization-usage.md +0 -107
- package/docs/runbooks/runner-egress-hardening.md +0 -81
- package/docs/runbooks/runtime-capabilities-overview.md +0 -113
- package/docs/runbooks/sbom-generation-gates.md +0 -71
- package/docs/runbooks/scorecard-supply-chain-gates.md +0 -82
- package/docs/runbooks/secret-scanning-gates.md +0 -85
- package/docs/runbooks/security-compliance-platform-demo-execution-log.md +0 -36
- package/docs/runbooks/security-compliance-platform-demo-script.md +0 -49
- package/docs/runbooks/security-compliance-platform-walkthrough.md +0 -98
- package/docs/runbooks/slsa-generator-patterns.md +0 -73
- package/docs/runbooks/slsa-verification-gates.md +0 -75
- package/docs/runbooks/solo-delivery-mode.md +0 -142
- package/docs/runbooks/solo-delivery-one-page.md +0 -111
- package/docs/runbooks/specialist-commands-playbook.md +0 -85
- package/docs/runbooks/sub-agent-invocation-map.md +0 -144
- package/docs/runbooks/system-architecture-design-walkthrough.md +0 -49
- package/docs/runbooks/team-closeout-example.md +0 -73
- package/docs/runbooks/team-command-output-contracts.md +0 -358
- package/docs/runbooks/team-commands-quick-prompts.md +0 -125
- package/docs/runbooks/team-execute-example.md +0 -63
- package/docs/runbooks/team-handoff-example.md +0 -49
- package/docs/runbooks/team-intake-example.md +0 -70
- package/docs/runbooks/team-plan-example.md +0 -62
- package/docs/runbooks/team-release-example.md +0 -63
- package/docs/runbooks/team-review-example.md +0 -61
- package/docs/runbooks/team-skills-test-run.md +0 -184
- package/docs/runbooks/team-skills-usage.md +0 -336
- package/docs/runbooks/team-training-reading-path.md +0 -64
- package/docs/runbooks/tech-lead-closure-conversation-example.md +0 -78
- package/docs/runbooks/tech-lead-daily-operations.md +0 -67
- package/docs/runbooks/trivy-security-gates.md +0 -79
- package/docs/runbooks/troubleshooting.md +0 -234
- package/docs/runbooks/vertical-scenario-capability-matrix.md +0 -107
- package/docs/runbooks/witness-policy-gates.md +0 -78
- package/docs/runbooks/zizmor-workflow-audits.md +0 -81
|
@@ -1,217 +0,0 @@
|
|
|
1
|
-
# Open-source Release Readiness Gap Plan
|
|
2
|
-
|
|
3
|
-
本文把当前仓库作为公开开源项目时仍未收口的事项整理成可执行清单,优先回答三个问题:
|
|
4
|
-
|
|
5
|
-
1. 哪些问题会直接影响外部用户信任或安装成功率。
|
|
6
|
-
2. 哪些问题已经在文档中被承认,但还没有真正修掉。
|
|
7
|
-
3. 下一轮应该按什么顺序推进,才能最快把“可发布”变成“可稳定采用”。
|
|
8
|
-
|
|
9
|
-
## Status Update — 2026-04-21
|
|
10
|
-
|
|
11
|
-
本轮 release-blocking 缺口已完成收口:
|
|
12
|
-
|
|
13
|
-
- `commands-core` 的 Codex selective-install gap 已修复,并有 regression test 保护。
|
|
14
|
-
- `tsp-create` / wizard / README / SUPPORT 的公开 target/profile 口径已统一。
|
|
15
|
-
- CI 已接入 `validate-doc-freshness`、strict reference validation,以及轻量 target coverage。
|
|
16
|
-
- GitHub issue 入口链接已改为可点击的绝对 URL。
|
|
17
|
-
- prebuilt binaries 已明确为 release/package artifacts;`build-prebuilt.yml` 不再尝试把被 `.gitignore` 忽略的 `bin/prebuilt/` 回写 Git。
|
|
18
|
-
- `npm run release:health` 与 `open-source-release-checklist.md` 已把发布前的 docs / support / prebuilt / tarball / install-surface 校验收成统一入口。
|
|
19
|
-
|
|
20
|
-
后续剩余工作主要是 adoption-oriented 的增强项,而不是下一次公开发布的阻塞项。
|
|
21
|
-
|
|
22
|
-
## 1. 当前结论
|
|
23
|
-
|
|
24
|
-
当前仓库已经具备公开仓库的基础面:
|
|
25
|
-
|
|
26
|
-
- `README.md`、`LICENSE`、`CONTRIBUTING.md`、`CODE_OF_CONDUCT.md`、`SECURITY.md`、`SUPPORT.md` 已存在。
|
|
27
|
-
- `node scripts/validate-library.js`、`node scripts/validate-file-references.js --strict`、`node tests/run-all.js` 当前可通过。
|
|
28
|
-
- npm 发布工作流、CI、Issue / PR 模板、CODEOWNERS 已具备基础骨架。
|
|
29
|
-
|
|
30
|
-
本轮 release-blocking 缺口已经收口,剩余事项主要转为 adoption-oriented follow-up:
|
|
31
|
-
|
|
32
|
-
- 为次级 targets 补更轻量的 quick-start / example。
|
|
33
|
-
- 把 release health 汇总成更容易复用的一页 checklist 或 summary 命令。
|
|
34
|
-
- 持续保持 README / SUPPORT / install doctor / runbooks 的 support-level 口径一致。
|
|
35
|
-
|
|
36
|
-
## 2. Closed Release Blockers (Historical Record)
|
|
37
|
-
|
|
38
|
-
### 2.1 Clarify support depth after the Codex command-gap fix
|
|
39
|
-
|
|
40
|
-
**Evidence**
|
|
41
|
-
|
|
42
|
-
- `commands-core` 已补上 `codex` target,`install-plan --profile full --target codex` 不再跳过核心命令模块。
|
|
43
|
-
- 但 `copilot` 等次级 targets 仍会在 `team` profile 下跳过大量模块,说明“支持 target”与“支持深度”仍然不是一回事。
|
|
44
|
-
- `README.md` 目前已声明 10 个公开 targets,但没有把成熟度或能力边界表达清楚。
|
|
45
|
-
|
|
46
|
-
**Why this matters**
|
|
47
|
-
|
|
48
|
-
修掉核心命令缺口后,下一步风险不再是“能不能装上命令”,而是“用户是否会把 partial support 误读成 full parity”。
|
|
49
|
-
|
|
50
|
-
**Required work**
|
|
51
|
-
|
|
52
|
-
- 保留 Codex command-gap regression test,防止 `commands-core` 再次回退。
|
|
53
|
-
- 为次级 targets 标注 support level 或 capability depth,而不是只列 target 名称。
|
|
54
|
-
- 把 README / quick-start / support docs 的承诺改成与实际模块覆盖一致。
|
|
55
|
-
|
|
56
|
-
**Definition of done**
|
|
57
|
-
|
|
58
|
-
- `install-plan --profile full --target codex` 持续保持 `commands-core` 选中。
|
|
59
|
-
- 有明确测试在 CI 中覆盖该场景。
|
|
60
|
-
- README 与 support docs 对次级 targets 的能力边界表达一致。
|
|
61
|
-
|
|
62
|
-
### 2.2 Reconcile the public installer surface
|
|
63
|
-
|
|
64
|
-
**Evidence**
|
|
65
|
-
|
|
66
|
-
- `bin/tsp-create.js` 的帮助文本仍暴露 `enterprise` profile,并只列 7 个 targets。
|
|
67
|
-
- `bin/lib/wizard.js` 也只展示 7 个 targets,并把 `enterprise` 当作可选 profile。
|
|
68
|
-
- `manifests/install-profiles.json` 的公开 profile 中已经没有 `enterprise`。
|
|
69
|
-
- `scripts/lib/install-targets/registry.js` 的真实 target 适配器已经是 10 个。
|
|
70
|
-
|
|
71
|
-
**Why this matters**
|
|
72
|
-
|
|
73
|
-
外部用户最先接触的是 `npx @colin4k1024/tsp --help` 和交互向导。这里一旦和真实能力面不一致,用户会第一时间撞上“文档说可以、实际报错”的失败体验。
|
|
74
|
-
|
|
75
|
-
**Required work**
|
|
76
|
-
|
|
77
|
-
- 统一 `tsp-create` help、wizard、README、Support / quick-start 对 target 和 profile 的公开口径。
|
|
78
|
-
- 明确 public repo 是否真的对外承诺全部 10 个 targets;如果不是,就缩减声明,而不是继续半暴露。
|
|
79
|
-
- 对已移出公开仓的 `enterprise` profile 做更明确的错误提示或入口说明,避免现在这种看起来可用、实际报错的状态。
|
|
80
|
-
|
|
81
|
-
**Definition of done**
|
|
82
|
-
|
|
83
|
-
- `tsp-create --help`、交互式 wizard、README、SUPPORT 文档中的 targets / profiles 完全一致。
|
|
84
|
-
- 不存在帮助文本暴露但 runtime 拒绝的 profile。
|
|
85
|
-
- 至少为当前公开支持的每个 target 提供一句明确的安装边界说明。
|
|
86
|
-
|
|
87
|
-
### 2.3 Promote declared release gates into enforced CI
|
|
88
|
-
|
|
89
|
-
**Evidence**
|
|
90
|
-
|
|
91
|
-
- `CONTRIBUTING.md`、`package.json`、`AGENTS.md` 已要求运行 `validate-doc-freshness`、strict reference validation 等检查。
|
|
92
|
-
- `.github/workflows/ci.yml` 当前只跑 build / validate / tests / Claude install smoke,没有把这些文档门禁接入。
|
|
93
|
-
- `tests/test_platform_smoke.js` 只验证 `~/.claude` 安装面,与“10 个 targets”口径不匹配。
|
|
94
|
-
|
|
95
|
-
**Why this matters**
|
|
96
|
-
|
|
97
|
-
开源项目最怕“规则只写在文档里”。如果贡献者看见 PR 模板和贡献指南要求某些门禁,但 GitHub checks 并不执行,项目治理会快速失真。
|
|
98
|
-
|
|
99
|
-
**Required work**
|
|
100
|
-
|
|
101
|
-
- 在 CI 中加入 `node scripts/validate-doc-freshness.js`。
|
|
102
|
-
- 在 CI 中加入 `node scripts/validate-file-references.js --strict`。
|
|
103
|
-
- 评估是否把 `npm run lint` 也升级为正式门禁。
|
|
104
|
-
- 为非-Claude targets 至少补一层轻量安装验证:可接受 install-plan / doctor matrix,也可以做更窄的 smoke,但不能完全没有。
|
|
105
|
-
|
|
106
|
-
**Definition of done**
|
|
107
|
-
|
|
108
|
-
- PR required checks 与贡献文档列出的最低门禁一致。
|
|
109
|
-
- target coverage 不再只剩 Claude 一条路径。
|
|
110
|
-
- 文档、引用、安装面漂移能在 PR 阶段而不是发布后暴露。
|
|
111
|
-
|
|
112
|
-
### 2.4 Keep entry docs publish-clean
|
|
113
|
-
|
|
114
|
-
**Evidence**
|
|
115
|
-
|
|
116
|
-
- 当前 `validate-doc-freshness` 会因为 `AGENTS.md` 含 Claude memory context block 失败。
|
|
117
|
-
- 该检查说明入口文档存在会话残留污染风险。
|
|
118
|
-
|
|
119
|
-
**Why this matters**
|
|
120
|
-
|
|
121
|
-
这类内容一旦进入公开仓库,不只是噪音,还会泄露内部工作痕迹并破坏入口文档可信度。
|
|
122
|
-
|
|
123
|
-
**Required work**
|
|
124
|
-
|
|
125
|
-
- 在发布前移除 `AGENTS.md` 中的会话记忆块。
|
|
126
|
-
- 明确是否需要额外的 pre-commit / CI 保护,避免类似残留再次进入入口文档。
|
|
127
|
-
|
|
128
|
-
**Definition of done**
|
|
129
|
-
|
|
130
|
-
- `node scripts/validate-doc-freshness.js` 在干净工作区内通过。
|
|
131
|
-
- 入口文档没有 session residue、temporary notes 或会话上下文块。
|
|
132
|
-
|
|
133
|
-
## 3. Should Fix Next
|
|
134
|
-
|
|
135
|
-
### 3.1 Repair GitHub issue entry links
|
|
136
|
-
|
|
137
|
-
**Evidence**
|
|
138
|
-
|
|
139
|
-
- `.github/ISSUE_TEMPLATE/config.yml` 里的 `contact_links` 使用 `./docs/...` 相对路径。
|
|
140
|
-
|
|
141
|
-
**Why this matters**
|
|
142
|
-
|
|
143
|
-
GitHub issue chooser 需要真实可跳转的 URL。相对路径看起来像链接,但对外部贡献者通常不可点击或行为不符合预期。
|
|
144
|
-
|
|
145
|
-
**Required work**
|
|
146
|
-
|
|
147
|
-
- 改成仓库绝对 URL 或文档站 URL。
|
|
148
|
-
- 顺手确认 Discussions / Security / Support 入口与公开协作方式一致。
|
|
149
|
-
|
|
150
|
-
### 3.2 Decide the prebuilt binary source-of-truth
|
|
151
|
-
|
|
152
|
-
**Evidence**
|
|
153
|
-
|
|
154
|
-
- `build-prebuilt.yml` 试图把 `bin/prebuilt/` 回写仓库。
|
|
155
|
-
- `.gitignore` 明确忽略了 `bin/prebuilt/`。
|
|
156
|
-
- 当前 `git add --dry-run bin/prebuilt/...` 会被 ignore 拦住。
|
|
157
|
-
|
|
158
|
-
**Why this matters**
|
|
159
|
-
|
|
160
|
-
这是一个“自动化看起来完整,实际落不到 Git”的典型失效路径。要么 prebuilt 是 release-only artifact,要么它们应该真正进入版本管理,二者必须二选一。
|
|
161
|
-
|
|
162
|
-
**Required work**
|
|
163
|
-
|
|
164
|
-
- 明确 prebuilt 的权威存储位置:Git 仓库、GitHub Release artifact、还是 npm pack staging。
|
|
165
|
-
- 删除与最终策略冲突的 workflow 步骤。
|
|
166
|
-
- 把 README / release docs 中对 prebuilt 来源的描述同步到最终策略。
|
|
167
|
-
|
|
168
|
-
### 3.3 Clarify support depth for non-primary targets
|
|
169
|
-
|
|
170
|
-
**Evidence**
|
|
171
|
-
|
|
172
|
-
- README 对 10 个 targets 做了高层声明。
|
|
173
|
-
- runbooks / examples / smoke tests 仍主要围绕 Claude、Codex、Cursor、OpenCode。
|
|
174
|
-
|
|
175
|
-
**Why this matters**
|
|
176
|
-
|
|
177
|
-
“支持 target”不等于“有完整上手路径”。如果某些 target 只有 adapter,没有 docs、examples、troubleshooting,就应该明确标注成熟度。
|
|
178
|
-
|
|
179
|
-
**Required work**
|
|
180
|
-
|
|
181
|
-
- 给每个非主 target 标注 support level,例如 experimental / basic / recommended。
|
|
182
|
-
- 若暂时不打算投入文档和回归覆盖,就下调 README 的公开承诺。
|
|
183
|
-
|
|
184
|
-
## 4. Nice to Have
|
|
185
|
-
|
|
186
|
-
### 4.1 Add an open-source release checklist
|
|
187
|
-
|
|
188
|
-
- 目标:把发布前的 README / tarball / workflow / docs / support / security 校验收成一页 checklist。
|
|
189
|
-
- 价值:减少“这次是记得的,下次靠回忆”的人工波动。
|
|
190
|
-
|
|
191
|
-
### 4.2 Add community-facing adoption examples beyond Claude/Codex
|
|
192
|
-
|
|
193
|
-
- 目标:为 Cursor / Copilot / Windsurf / Augment 这类 target 增加最小示例或 quick-start stub。
|
|
194
|
-
- 价值:降低公开支持矩阵的首次理解成本。
|
|
195
|
-
|
|
196
|
-
### 4.3 Add a release-health summary command or report
|
|
197
|
-
|
|
198
|
-
- 目标:在发布前输出统一状态摘要,例如 docs gates、tarball gates、prebuilt gates、support surface、known risks。
|
|
199
|
-
- 价值:让发布决策不再依赖人工拼装多个脚本输出。
|
|
200
|
-
|
|
201
|
-
## 5. Recommended Execution Order
|
|
202
|
-
|
|
203
|
-
1. 先修 `codex` `commands-core` gap,因为这是最直接的真实能力缺口。
|
|
204
|
-
2. 再统一 `tsp-create` / wizard / README 的公开口径,避免继续误导新用户。
|
|
205
|
-
3. 把 `validate-doc-freshness`、strict reference validation 和 target coverage 接到 CI,让规则真正变成门禁。
|
|
206
|
-
4. 处理 GitHub issue 入口链接和 prebuilt workflow 失效路径,收掉“看似存在、实际不可用”的外围问题。
|
|
207
|
-
5. 最后再补 release checklist 和非主 target 文档,把项目从“能发布”推进到“更好采用”。
|
|
208
|
-
|
|
209
|
-
## 6. Exit Condition
|
|
210
|
-
|
|
211
|
-
当以下条件同时满足时,可以认为这一轮开源发布收口完成:
|
|
212
|
-
|
|
213
|
-
- 公开支持矩阵中不存在已知但未修的核心安装缺口。
|
|
214
|
-
- 对外 CLI、README、Support、Issue / PR 入口描述一致。
|
|
215
|
-
- 贡献指南里声明的最低质量门禁均已进入 GitHub checks。
|
|
216
|
-
- 发布自动化中不再保留与仓库策略冲突的失效步骤。
|
|
217
|
-
- 活跃入口文档可以在干净工作区内通过文档新鲜度校验。
|
|
@@ -1,147 +0,0 @@
|
|
|
1
|
-
# LLM Surface Reduction Audit
|
|
2
|
-
|
|
3
|
-
本文只保留减法审计结论,不再承载逐阶段执行台账。
|
|
4
|
-
|
|
5
|
-
执行清单看 [llm-surface-reduction-execution-checklist.md](llm-surface-reduction-execution-checklist.md),逐阶段历史看 [llm-surface-reduction-execution-history.md](llm-surface-reduction-execution-history.md)。
|
|
6
|
-
|
|
7
|
-
## 1. 审计结论
|
|
8
|
-
|
|
9
|
-
截至 2026-03-29,这一轮 LLM surface reduction 已达到目标:
|
|
10
|
-
|
|
11
|
-
- 删除了主要在重复现代 LLM 通用能力的 shared skills、ecc skills、公开命令和通用模板。
|
|
12
|
-
- 保住了平台主链、企业专属能力、具体工具链能力和前端治理锚点。
|
|
13
|
-
- 生成链、README、runbook、examples 与角色说明均已完成同步收敛。
|
|
14
|
-
|
|
15
|
-
当前建议从“继续删减”切换为“维持收敛状态”,后续只在新增对象明显属于通用包装层时再做新一轮审计。
|
|
16
|
-
|
|
17
|
-
## 2. 判定原则
|
|
18
|
-
|
|
19
|
-
一个对象适合做减法,通常同时满足以下三条中的两条以上:
|
|
20
|
-
|
|
21
|
-
1. 本体内容很薄,主要是在重述已有 rule、runbook 或 template。
|
|
22
|
-
2. 不承载企业专属知识、具体工具链边界或严格回写约束。
|
|
23
|
-
3. 删除后,现代 LLM 在没有专门 skill 或命令包装的情况下仍能稳定产出相近结果。
|
|
24
|
-
|
|
25
|
-
一个对象不适合删,通常是因为满足以下任意一条:
|
|
26
|
-
|
|
27
|
-
1. 承载企业专属能力,例如 私有流程、权限集成、组织内部领域扩展。
|
|
28
|
-
2. 承载具体工具链或验证方法,例如 `maven-qa`、`mysql-query`、`browser-smoke-testing`。
|
|
29
|
-
3. 是平台主链、前端治理或发布门禁中的显式锚点,被大范围引用。
|
|
30
|
-
|
|
31
|
-
## 3. 当前保留项
|
|
32
|
-
|
|
33
|
-
### 3.1 Shared skills
|
|
34
|
-
|
|
35
|
-
- `api-contract`
|
|
36
|
-
- `frontend-engineering`
|
|
37
|
-
- `frontend-ui-ux-system`
|
|
38
|
-
|
|
39
|
-
原因:这三项仍然直接承载 API 契约、前端工程和 UI/UX 治理约束,不是通用包装层。
|
|
40
|
-
|
|
41
|
-
### 3.2 ECC skills
|
|
42
|
-
|
|
43
|
-
- `browser-smoke-testing`
|
|
44
|
-
- `java-unit-test`
|
|
45
|
-
- `maven-qa`
|
|
46
|
-
- `mysql-query`
|
|
47
|
-
- `pairwise-test-design`
|
|
48
|
-
- `systematic-debugging`
|
|
49
|
-
- `testcontainers-integration-testing`
|
|
50
|
-
|
|
51
|
-
原因:都带有明确工具链、执行边界或方法论,不适合并回通用描述。
|
|
52
|
-
|
|
53
|
-
### 3.3 Company skills
|
|
54
|
-
|
|
55
|
-
- `skills/*` 全部保留
|
|
56
|
-
|
|
57
|
-
原因:这是仓库最明确的非通用知识层,正是现代通用 LLM 不具备、因此应该保留的部分。
|
|
58
|
-
|
|
59
|
-
### 3.4 公开命令
|
|
60
|
-
|
|
61
|
-
- `/team-intake`
|
|
62
|
-
- `/team-plan`
|
|
63
|
-
- `/team-execute`
|
|
64
|
-
- `/handoff`
|
|
65
|
-
- `/team-review`
|
|
66
|
-
- `/team-release`
|
|
67
|
-
- `/plan`
|
|
68
|
-
- `/code-review`
|
|
69
|
-
- `/build-fix`
|
|
70
|
-
- `/verify`
|
|
71
|
-
- `/multi-frontend`
|
|
72
|
-
- `/multi-backend`
|
|
73
|
-
|
|
74
|
-
原因:这是当前平台最稳定的主链与 specialist 入口,收缩后已具备足够识别度。
|
|
75
|
-
|
|
76
|
-
### 3.5 模板
|
|
77
|
-
|
|
78
|
-
- `templates/api-contract.md`
|
|
79
|
-
- `templates/design-system-brief.md`
|
|
80
|
-
- `templates/release-plan.md`
|
|
81
|
-
- `templates/ui-implementation-plan.md`
|
|
82
|
-
- `templates/ui-review-checklist.md`
|
|
83
|
-
|
|
84
|
-
原因:这些模板仍然带有显著的平台门禁、发布结构或 UI 质量要求。
|
|
85
|
-
|
|
86
|
-
## 4. 已完成的减法结果
|
|
87
|
-
|
|
88
|
-
### 4.1 已删除 shared skills
|
|
89
|
-
|
|
90
|
-
- `requirements-analysis`
|
|
91
|
-
- `feature-backlog`
|
|
92
|
-
- `incident-response`
|
|
93
|
-
- `release-readiness`
|
|
94
|
-
- `solution-design`
|
|
95
|
-
- `test-strategy`
|
|
96
|
-
|
|
97
|
-
### 4.2 已删除 ecc skills
|
|
98
|
-
|
|
99
|
-
- `search-first`
|
|
100
|
-
- `verification-loop`
|
|
101
|
-
- `coding-standards`
|
|
102
|
-
- `frontend-patterns`
|
|
103
|
-
- `backend-patterns`
|
|
104
|
-
- `security-review`
|
|
105
|
-
- `api-design`
|
|
106
|
-
- `deployment-patterns`
|
|
107
|
-
- `tdd-workflow`
|
|
108
|
-
|
|
109
|
-
### 4.3 已删除公开命令
|
|
110
|
-
|
|
111
|
-
- `/eval`
|
|
112
|
-
- `/sessions`
|
|
113
|
-
- `/multi-plan`
|
|
114
|
-
- `/multi-execute`
|
|
115
|
-
- `/multi-workflow`
|
|
116
|
-
- `/orchestrate`
|
|
117
|
-
- `/test-coverage`
|
|
118
|
-
- `/update-docs`
|
|
119
|
-
|
|
120
|
-
### 4.4 已删除通用模板
|
|
121
|
-
|
|
122
|
-
- `templates/prd.md`
|
|
123
|
-
- `templates/delivery-plan.md`
|
|
124
|
-
- `templates/test-plan.md`
|
|
125
|
-
- `templates/incident-brief.md`
|
|
126
|
-
- `templates/adr.md`
|
|
127
|
-
|
|
128
|
-
## 5. 当前状态
|
|
129
|
-
|
|
130
|
-
最新一轮校验后的平台状态如下:
|
|
131
|
-
|
|
132
|
-
- Roles: 8
|
|
133
|
-
- Shared skills: 3
|
|
134
|
-
- ECC skills: 9
|
|
135
|
-
- Company skills: 8
|
|
136
|
-
- Specialist agents: 27
|
|
137
|
-
- Generated artifacts: 69
|
|
138
|
-
|
|
139
|
-
这组数字说明:减法已经完成,但主链、企业扩展和工具链能力仍保持完整。
|
|
140
|
-
|
|
141
|
-
## 6. 后续策略
|
|
142
|
-
|
|
143
|
-
当前不建议继续做大规模删减。更合理的后续动作是:
|
|
144
|
-
|
|
145
|
-
1. 维持收敛结果,不再恢复已删除的薄包装对象。
|
|
146
|
-
2. 新增对象时先用本审计中的判定原则过一遍,再决定是否真的需要独立 skill、命令或模板。
|
|
147
|
-
3. 文档侧优先做入口收敛和示例补齐,而不是继续扩大命名面。
|
|
@@ -1,217 +0,0 @@
|
|
|
1
|
-
# LLM Surface Reduction Execution Checklist
|
|
2
|
-
|
|
3
|
-
本文是面向执行的减法清单。目标不是继续分析,而是给出一份可以按阶段推进、逐项勾选的行动列表。
|
|
4
|
-
> 历史命令说明:文中勾选记录里的 `python3 scripts/build_platform_artifacts.py` 与 `python3 scripts/validate_library.py` 为当时执行命名;当前等价命令分别为 `node scripts/build-platform-artifacts.js` 与 `node scripts/validate-library.js`。
|
|
5
|
-
|
|
6
|
-
审计结论看 [llm-surface-reduction-audit.md](llm-surface-reduction-audit.md),逐阶段历史看 [llm-surface-reduction-execution-history.md](llm-surface-reduction-execution-history.md)。
|
|
7
|
-
|
|
8
|
-
## 1. 执行目标
|
|
9
|
-
|
|
10
|
-
- 降低 `skills`、`commands`、`templates` 的表面积。
|
|
11
|
-
- 删除主要在重复现代 LLM 通用能力的包装层。
|
|
12
|
-
- 保留平台主链、企业专属能力、具体工具链能力和前端治理门禁。
|
|
13
|
-
|
|
14
|
-
## 2. 执行原则
|
|
15
|
-
|
|
16
|
-
1. 先删命名噪音,不碰主链。
|
|
17
|
-
2. 先删“薄包装”,后合并“弱独立价值”对象。
|
|
18
|
-
3. 每一阶段都必须先完成引用清理,再删实体目录或文件。
|
|
19
|
-
4. 每一阶段完成后都重新生成产物并跑校验。
|
|
20
|
-
|
|
21
|
-
## 3. 第一阶段:先做,风险最低
|
|
22
|
-
|
|
23
|
-
执行状态:已于 2026-03-29 完成。
|
|
24
|
-
|
|
25
|
-
### 3.1 删除或并入 `skills/search-first`
|
|
26
|
-
|
|
27
|
-
- [x] 已将“先查事实再下结论”的要求并回 specialist 生成源
|
|
28
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
29
|
-
- [x] 已删除 `skills/search-first/`
|
|
30
|
-
|
|
31
|
-
### 3.2 删除或并入 `skills/requirements-analysis`
|
|
32
|
-
|
|
33
|
-
- [x] 已将需求澄清能力收敛回角色主链与示例文档
|
|
34
|
-
- [x] 已更新 `roles/tech-lead/role.yaml`
|
|
35
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
36
|
-
- [x] 已更新相关 runbook、examples、README
|
|
37
|
-
- [x] 已删除 `skills/requirements-analysis/`
|
|
38
|
-
|
|
39
|
-
### 3.3 删除或并入 `skills/feature-backlog`
|
|
40
|
-
|
|
41
|
-
- [x] 已将 backlog / milestone 拆解收敛回角色主链与迁移文档
|
|
42
|
-
- [x] 已更新 `roles/tech-lead/role.yaml`
|
|
43
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
44
|
-
- [x] 已更新相关 runbook、examples、README
|
|
45
|
-
- [x] 已删除 `skills/feature-backlog/`
|
|
46
|
-
|
|
47
|
-
### 3.4 删除或并入 `skills/incident-response`
|
|
48
|
-
|
|
49
|
-
- [x] 已将事故协作要求收敛回 runbook 和调试约束
|
|
50
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
51
|
-
- [x] 已更新相关 runbook、README
|
|
52
|
-
- [x] 已删除 `skills/incident-response/`
|
|
53
|
-
|
|
54
|
-
### 3.5 删除 `/eval`
|
|
55
|
-
|
|
56
|
-
- [x] 已将公开评估入口收敛回 `/verify` 与现有评审链路
|
|
57
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
58
|
-
- [x] 已更新相关 README / 生成产物
|
|
59
|
-
- [x] 已删除 `commands/eval.md`
|
|
60
|
-
|
|
61
|
-
### 3.6 删除 `/sessions`
|
|
62
|
-
|
|
63
|
-
- [x] 已将 `/sessions` 降级为内部流程,不再作为公开命令
|
|
64
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
65
|
-
- [x] 已更新相关 README / 生成产物
|
|
66
|
-
- [x] 已删除 `commands/sessions.md`
|
|
67
|
-
|
|
68
|
-
### 3.7 合并 `/orchestrate`、`/multi-workflow`、`/multi-execute`
|
|
69
|
-
|
|
70
|
-
- [x] 已明确最终保留 `/plan`、`/multi-frontend`、`/multi-backend`
|
|
71
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
72
|
-
- [x] 已更新相关 README / runbook / 生成产物
|
|
73
|
-
- [x] 已删除 `commands/multi-execute.md`、`commands/multi-workflow.md`、`commands/orchestrate.md`
|
|
74
|
-
|
|
75
|
-
### 3.8 合并 `/multi-plan` 到 `/plan`
|
|
76
|
-
|
|
77
|
-
- [x] 已把“并行规划”语义吸收到 `/plan`
|
|
78
|
-
- [x] 已更新 `scripts/team_skills_platform.py`
|
|
79
|
-
- [x] 已更新相关文档与生成产物
|
|
80
|
-
- [x] 已删除 `commands/multi-plan.md`
|
|
81
|
-
|
|
82
|
-
### 3.9 第一阶段收尾
|
|
83
|
-
|
|
84
|
-
- [x] 已运行历史命令 `python3 scripts/build_platform_artifacts.py`(当前等价:`node scripts/build-platform-artifacts.js`)
|
|
85
|
-
- [x] 已运行历史命令 `python3 scripts/validate_library.py`(当前等价:`node scripts/validate-library.js`)
|
|
86
|
-
- [x] 已更新 [docs/plans/llm-surface-reduction-audit.md](llm-surface-reduction-audit.md) 中第一阶段状态
|
|
87
|
-
|
|
88
|
-
## 4. 第二阶段:合并薄 skill 和薄命令
|
|
89
|
-
|
|
90
|
-
### 4.1 shared skills
|
|
91
|
-
|
|
92
|
-
- [x] 已合并 `skills/solution-design` 到 `architect` / `tech-lead`
|
|
93
|
-
- [x] 已合并 `skills/release-readiness` 到 `devops-engineer` / 发布治理 runbook
|
|
94
|
-
|
|
95
|
-
### 4.2 ecc skills
|
|
96
|
-
|
|
97
|
-
- [x] 已合并 `skills/coding-standards`
|
|
98
|
-
- [x] 已合并 `skills/verification-loop`
|
|
99
|
-
- [x] 已合并 `skills/frontend-patterns`
|
|
100
|
-
- [x] 已合并 `skills/backend-patterns`
|
|
101
|
-
- [x] 已合并 `skills/api-design`
|
|
102
|
-
- [x] 已合并 `skills/deployment-patterns`
|
|
103
|
-
- [x] 已合并 `skills/tdd-workflow`
|
|
104
|
-
- [x] 已合并 `skills/security-review`
|
|
105
|
-
|
|
106
|
-
### 4.3 commands
|
|
107
|
-
|
|
108
|
-
- [x] 已合并 `/multi-execute` 到最终保留的协调入口
|
|
109
|
-
- [x] 已合并 `/test-coverage` 到 `/tdd`
|
|
110
|
-
- [x] 已将 `/update-docs` 降级为 specialist 内部能力
|
|
111
|
-
|
|
112
|
-
### 4.4 第二阶段收尾
|
|
113
|
-
|
|
114
|
-
- [x] 运行历史命令 `python3 scripts/build_platform_artifacts.py`(当前等价:`node scripts/build-platform-artifacts.js`)
|
|
115
|
-
- [x] 运行历史命令 `python3 scripts/validate_library.py`(当前等价:`node scripts/validate-library.js`)
|
|
116
|
-
- [x] 更新 [docs/plans/llm-surface-reduction-audit.md](llm-surface-reduction-audit.md) 中第二阶段状态
|
|
117
|
-
|
|
118
|
-
## 5. 第三阶段:模板减法
|
|
119
|
-
|
|
120
|
-
### 5.1 优先评估删除
|
|
121
|
-
|
|
122
|
-
- [x] `templates/prd.md`
|
|
123
|
-
- [x] `templates/delivery-plan.md`
|
|
124
|
-
- [x] `templates/test-plan.md`
|
|
125
|
-
- [x] `templates/incident-brief.md`
|
|
126
|
-
|
|
127
|
-
### 5.2 暂缓项
|
|
128
|
-
|
|
129
|
-
- [x] `templates/adr.md` 已确认企业内控补充区可迁移,并回写到 `rules/artifact-standards.md`
|
|
130
|
-
|
|
131
|
-
### 5.3 替代动作
|
|
132
|
-
|
|
133
|
-
- [x] 将保留的字段要求回写到 `rules/artifact-standards.md`
|
|
134
|
-
- [x] 将高频模板示例内嵌到对应 runbook 中
|
|
135
|
-
- [x] 更新 README、AGENTS、CLAUDE、examples 和所有模板链接入口
|
|
136
|
-
|
|
137
|
-
### 5.4 第三阶段收尾
|
|
138
|
-
|
|
139
|
-
- [x] 运行历史命令 `python3 scripts/build_platform_artifacts.py`(当前等价:`node scripts/build-platform-artifacts.js`)
|
|
140
|
-
- [x] 运行历史命令 `python3 scripts/validate_library.py`(当前等价:`node scripts/validate-library.js`)
|
|
141
|
-
- [x] 更新 [docs/plans/llm-surface-reduction-audit.md](llm-surface-reduction-audit.md) 中第三阶段状态
|
|
142
|
-
|
|
143
|
-
## 5.5 第四阶段:剩余边界项收敛
|
|
144
|
-
|
|
145
|
-
- [x] 已评估 `skills/test-strategy`
|
|
146
|
-
- [x] 已将测试口径收敛回 `qa-engineer`、`artifact-standards.md` 和测试相关 runbook
|
|
147
|
-
- [x] 已更新 `scripts/team_skills_platform.py`、`roles/qa-engineer/role.yaml`、相关 examples 与 runbook
|
|
148
|
-
- [x] 运行历史命令 `python3 scripts/build_platform_artifacts.py`(当前等价:`node scripts/build-platform-artifacts.js`)
|
|
149
|
-
- [x] 运行历史命令 `python3 scripts/validate_library.py`(当前等价:`node scripts/validate-library.js`)
|
|
150
|
-
- [x] 更新 [docs/plans/llm-surface-reduction-audit.md](llm-surface-reduction-audit.md) 中最终状态
|
|
151
|
-
|
|
152
|
-
## 6. 当前明确不要动的项
|
|
153
|
-
|
|
154
|
-
### 6.1 skills(company 领域)
|
|
155
|
-
|
|
156
|
-
- [ ] 保留 `skills/*` 全部
|
|
157
|
-
|
|
158
|
-
### 6.2 高价值 ecc skills
|
|
159
|
-
|
|
160
|
-
- [ ] 保留 `browser-smoke-testing`
|
|
161
|
-
- [ ] 保留 `java-unit-test`
|
|
162
|
-
- [ ] 保留 `maven-qa`
|
|
163
|
-
- [ ] 保留 `mysql-query`
|
|
164
|
-
- [ ] 保留 `pairwise-test-design`
|
|
165
|
-
- [ ] 保留 `systematic-debugging`
|
|
166
|
-
- [ ] 保留 `testcontainers-integration-testing`
|
|
167
|
-
|
|
168
|
-
### 6.3 核心命令
|
|
169
|
-
|
|
170
|
-
- [ ] 保留 `/team-intake`
|
|
171
|
-
- [ ] 保留 `/team-plan`
|
|
172
|
-
- [ ] 保留 `/team-execute`
|
|
173
|
-
- [ ] 保留 `/handoff`
|
|
174
|
-
- [ ] 保留 `/team-review`
|
|
175
|
-
- [ ] 保留 `/team-release`
|
|
176
|
-
- [ ] 保留 `/plan`
|
|
177
|
-
- [ ] 保留 `/code-review`
|
|
178
|
-
- [ ] 保留 `/build-fix`
|
|
179
|
-
- [ ] 保留 `/verify`
|
|
180
|
-
- [ ] 保留 `/multi-frontend`
|
|
181
|
-
- [ ] 保留 `/multi-backend`
|
|
182
|
-
|
|
183
|
-
### 6.4 核心模板
|
|
184
|
-
|
|
185
|
-
- [ ] 保留 `templates/api-contract.md`
|
|
186
|
-
- [ ] 保留 `templates/ui-implementation-plan.md`
|
|
187
|
-
- [ ] 保留 `templates/ui-review-checklist.md`
|
|
188
|
-
- [ ] 保留 `templates/release-plan.md`
|
|
189
|
-
- [ ] 保留 `templates/design-system-brief.md`
|
|
190
|
-
|
|
191
|
-
## 7. 每一轮实际执行都要同步修改的地方
|
|
192
|
-
|
|
193
|
-
- [ ] `roles/*/role.yaml`
|
|
194
|
-
- [ ] `scripts/team_skills_platform.py`
|
|
195
|
-
- [ ] `README.md`
|
|
196
|
-
- [ ] `AGENTS.md`
|
|
197
|
-
- [ ] `CLAUDE.md`
|
|
198
|
-
- [ ] `docs/runbooks/team-skills-usage.md`
|
|
199
|
-
- [ ] 相关 `examples/*.md`
|
|
200
|
-
- [ ] 对应 usage-scenarios / reading path / 对话样例 / quick start 文档
|
|
201
|
-
|
|
202
|
-
## 8. 推荐起手批次
|
|
203
|
-
|
|
204
|
-
已完成的起手批次:
|
|
205
|
-
|
|
206
|
-
- [x] `search-first`
|
|
207
|
-
- [x] `requirements-analysis`
|
|
208
|
-
- [x] `feature-backlog`
|
|
209
|
-
- [x] `incident-response`
|
|
210
|
-
- [x] `/eval`
|
|
211
|
-
- [x] `/sessions`
|
|
212
|
-
- [x] `/multi-plan`
|
|
213
|
-
- [x] `/multi-execute`
|
|
214
|
-
- [x] `/multi-workflow`
|
|
215
|
-
- [x] `/orchestrate`
|
|
216
|
-
|
|
217
|
-
下一批建议进入文案收敛:继续清理安全与验证类 runbook 中残余的旧术语表述。
|