{"skill":{"slug":"knowledge-health-checker","displayName":"knowledge-health-checker","summary":"Audit and improve Markdown knowledge-base health across Obsidian, Logseq, Notion exports, docs folders, and wiki repositories. Detect empty placeholder notes...","tags":{"latest":"1.0.1"},"stats":{"comments":0,"downloads":49,"installsAllTime":0,"installsCurrent":0,"stars":1,"versions":1},"createdAt":1777095896700,"updatedAt":1777096006777},"latestVersion":{"version":"1.0.1","createdAt":1777095896700,"changelog":"# Changelog\n\nAll notable changes to `knowledge-health-checker` will be documented in this file.\n\n## [1.1.0] - 2026-04-25\n\n### Added\n\n- Added ClawHub-ready metadata and public-facing skill positioning.\n- Added a complete Markdown knowledge-base health audit workflow:\n  - confirm scan scope and safety\n  - build file and heading index\n  - detect hollow or low-value notes\n  - detect broken wiki and markdown links\n  - analyze content density and structure\n  - analyze knowledge graph connectivity\n  - calculate a health score\n  - generate a report and safe fix plan\n- Added support guidance for multiple knowledge-base formats:\n  - Obsidian vaults\n  - Logseq exports\n  - Notion Markdown exports\n  - documentation repositories\n  - plain Markdown wiki folders\n- Added a 0-100 knowledge health scoring model.\n- Added health labels:\n  - Excellent\n  - Healthy\n  - Needs maintenance\n  - Fragile\n  - Critical\n- Added a safe fix policy with three risk levels:\n  - Low-risk reporting and suggestions\n  - Medium-risk generated fix plans\n  - High-risk destructive or global changes requiring explicit confirmation\n- Added a standard output format for:\n  - health summary\n  - findings\n  - highest-impact fixes\n  - safe fix plan\n  - generated artifacts\n- Added Darwin-style regression prompts in `test-prompts.json`.\n\n### Improved\n\n- Reframed the skill as a general-purpose Markdown knowledge-base health auditor instead of an internal-only checker.\n- Improved safety guidance so the default behavior is now:\n\n  ```text\n  scan only → report only → generate fix plan → user reviews → user applies\nImproved audit clarity by separating knowledge-base health into four dimensions:\nhollow note rate\nbroken link rate\ncontent density\nnetwork connectivity\nImproved scalability guidance for large knowledge bases.\nImproved compatibility guidance for non-English filenames and mixed Markdown conventions.\nImproved output usefulness by requiring prioritized P0 / P1 / P2 repair suggestions.\nChanged\n\nReorganized the original long workflow into a cleaner ClawHub-friendly structure.\nChanged “automatic repair” guidance into a safer reviewed fix-plan workflow.\nMoved the skill from a local archived copy into the active skills directory:\nCopy\n/Users/mac/.openclaw/skills/knowledge-health-checker\nVerified\n\nRan a smoke test using a temporary Markdown knowledge-base sample.\nVerified that the bundled scanner can process Markdown files and generate health output.\nCommand used:\nCopy\npython3 scripts/health_check.py <tmp-sample>\nResult:\nCopy\n扫描完成：发现 3 个Markdown文件\n检查完成！健康分：27.0\nFiles\n\nUpdated SKILL.md\nAdded CHANGELOG.md\nAdded test-prompts.json\nVerified scripts/health_check.py","license":"MIT-0"},"metadata":null,"owner":{"handle":"xb19960921","userId":"s172f3jjxf40vh438bf24d3cj983g86d","displayName":"xb19960921","image":"https://avatars.githubusercontent.com/u/83778953?v=4"},"moderation":null}