菜单
Use when editing docs/plans (PLANS_TODO, PLANS_HUB, PLAN_*), docs/MAP*, docs/README hub rows, docs/adr index, or AGENTS ADR pointer—keep hubs aligned with repo truth and ADR policy.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/DataBoar/data-boar --skill doc-hubs-plans-sync命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Guides Python3 programs that access databases and filesystems to detect possible personal or sensitive data using regex and ML, aligned with LGPD, GDPR and similar norms. Use when implementing or reviewing PII/sensitive data detection, compliance scanning, crawlers, or pattern recognition for data protection laws.
Use when deciding how to run lint, tests, commit/PR, or Docker homelab steps. Prefer repo scripts (check-all, lint-only, quick-test, commit-or-pr, docker-lab-build, docker-hub-pull, docker-prune-local) over ad-hoc commands to save tokens and keep behaviour consistent.
Checklist for enforcing NASA-style active voice, LCM Level 1/2 verbs, and canonical Data Boar lexicon stability in code, docs, commit messages, and PR reviews. Use when writing or reviewing docstrings, module comments, PowerShell CBH, Python function names, technical docs, or any prose produced for this repo.
基于 SOC 职业分类
正在显示 SKILL.md
| name | doc-hubs-plans-sync |
| description | Use when editing docs/plans (PLANS_TODO, PLANS_HUB, PLAN_*), docs/MAP*, docs/README hub rows, docs/adr index, or AGENTS ADR pointer—keep hubs aligned with repo truth and ADR policy. |
| Gate | What it guards | Command |
|---|---|---|
| plans-hub | PLANS_HUB.md table matches every PLAN_*.md (open + completed/) | python scripts/plans_hub_sync.py --check / --write |
| plans-stats | Status dashboard block in PLANS_TODO.md matches tables | python scripts/plans-stats.py --check / --write |
| Pre-commit / CI | Same hooks as .\scripts\check-all.ps1 Lint job | uv run pre-commit run --all-files |
| Tests | tests/test_plans_hub_sync.py, tests/test_plans_stats.py, tests/test_docs_external_no_plan_links.py, markdown lint | Part of check-all |
These do not auto-edit MAP.md, README pitch tables, or AGENTS.md ADR guidance prose—that is human/agent ritual below. AGENTS.md does not carry a hardcoded “current last ADR” number (it goes stale); agents discover the latest file on disk or run .\scripts\new-adr.ps1 (auto-numbering).
PLANS_TODO.md: update rows / dependency narrative (English-only history per policy).python scripts/plans-stats.py --write if any status table row changed.git mv archived PLAN_*.md: python scripts/plans_hub_sync.py --write; fix broken links per .cursor/rules/docs-plans.mdc..\scripts\check-all.ps1 (or at least uv run pre-commit run --all-files) before commit.docs/adr/README.md and docs/adr/README.pt_BR.md: index row + consequences if needed..\scripts\new-adr.ps1 for the next id or list docs/adr/ADR-*.md sorted to see the latest.MAP.pt_BR.md / docs/README.pt_BR.md in the same PR as EN when the change is not EN-only typo.(...) target under docs/ should resolve; run uv run pytest tests/test_markdown_lint.py if you touched many links.docs/plans/; MAP already points to docs/README for plans entry.PLANS_TODO.md + PLANS_HUB.md after the scripts above—not social posts or chat summaries.cd <repo-root>
python scripts/plans_hub_sync.py --write
python scripts/plans-stats.py --write
.\scripts\check-all.ps1
When only docs/markdown changed: .\scripts\lint-only.ps1 may be enough if you did not touch Python—still run plans scripts if plans files changed.