[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C7GD3CXB6VMARKTERWAJSF":3,"skill-same:01M6C7GD3CXB6VMARKTERWAJSF":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":25,"linkOnly":26,"markdownContent":27,"markdownBody":28,"frontMatter":29,"refName":30,"commitSha":31,"skillDirectory":32,"forksCount":33,"repositoryHtmlUrl":34,"repositoryHomepageUrl":6,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M6C7GD3CXB6VMARKTERWAJSF","audit",null,"Comprehensive codebase audit for release readiness. Parallel exploration of docs, code, config, tests, and specs to identify cruft, then interactive triage with clear action options.","lepahc\u002Fkynetic-spec","lepahc","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F255431765?v=4","01M6C7E6Z0C8SE1X6PJ2F9Q1CX","01M6C7E744QKNYBVZ7KV7Z69GF",1,"MIT","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Faudit\u002FSKILL.md",[],[5,18,19,20,21,22,23,24],"docs","triage","config","code","release","parallel","interactive","2026-11-10 10:12:30",false,"---\nname: audit\ndescription: Comprehensive codebase audit for release readiness. Parallel\n  exploration of docs, code, config, tests, and specs to identify cruft, then\n  interactive triage with clear action options.\n---\n\u003C!-- kspec-managed -->\n\n# Codebase Audit\n\nSystematic review of the entire codebase to identify cruft, outdated content, and issues before release. Uses parallel exploration for discovery, then interactive triage for decisions.\n\n## When to Use\n\n- Pre-release cleanup\n- Periodic codebase health check\n- After major refactoring\n- When onboarding to understand technical debt\n\n## Workflow\n\n### Phase 1: Discovery (Parallel Exploration)\n\nLaunch 5 parallel Explore agents to review different areas:\n\n```\nTask tool (subagent_type: Explore) - run all 5 in parallel:\n\n1. DOCS: Review *.md files for outdated content, stale TODOs,\n   inconsistencies, wrong command syntax, aspirational features\n   documented as current\n\n2. CODE: Review src\u002F for dead code, unresolved TODOs\u002FFIXMEs,\n   duplicate functions, commented-out code, debug artifacts,\n   unused imports\u002Fexports\n\n3. CONFIG: Review package.json, tsconfig, etc. for unused deps,\n   misplaced deps (dev vs prod), broken scripts, stale configs,\n   version drift\n\n4. TESTS: Review tests\u002F for skipped tests, dead mocks, duplicate\n   test names, flaky patterns, missing coverage, outdated fixtures\n\n5. SPECS: Review .kspec\u002F for stale spec items, orphaned tasks,\n   status mismatches, empty stubs, inbox items that should be\n   promoted, accumulating session files\n```\n\nEach agent should report findings with:\n- File and location\n- What the issue is\n- Why it's cruft (outdated, redundant, broken, etc.)\n- Recommended action\n\n### Phase 2: Compilation\n\nOrganize all findings by severity:\n\n| Severity | Criteria | Examples |\n|----------|----------|----------|\n| **CRITICAL** | Blocks release, breaks functionality | Runtime deps in devDeps, security vulns, broken required scripts |\n| **HIGH** | Should fix, user-facing or misleading | Wrong docs, incorrect command syntax, misleading task statuses |\n| **MEDIUM** | Recommended, technical debt | Unresolved TODOs, duplicate code, test improvements |\n| **LOW** | Nice to have, minor cleanup | Doc cross-refs, test name clarity, helper consolidation |\n\nPresent compiled summary to user before triage.\n\n### Phase 3: Triage (Interactive)\n\nGo through items one-by-one using AskUserQuestion:\n\n```\nFor each item, present options:\n\n- Fix now      → Execute the fix immediately\n- Task         → Create task with context notes\n- Inbox        → Add to inbox for later review\n- Punt         → Skip, not worth tracking\n- Discuss      → Need more context before deciding\n```\n\n**Triage principles:**\n- One item at a time (don't batch)\n- User decides disposition\n- Execute decisions immediately\n- Add context notes to tasks\u002Finbox\n\n### Phase 4: Execution\n\nFor each decision type:\n\n**Fix now:**\n```bash\n# Delete dead files\nrm path\u002Fto\u002Fdead-file.js\n\n# Quick edits\n# Use Edit tool for small fixes\n\n# Commit cleanup changes at end\ngit add -A && git commit -m \"chore: Audit cleanup - [summary]\"\n```\n\n**Task:**\n```bash\nkspec task add --title \"Description\" --priority N --slug slug-name --tag relevant-tag\nkspec task note @slug \"Context: file location, what needs doing, why\"\n```\n\n**Inbox:**\n```bash\nkspec inbox add \"Description of item for later\" --tag tag1 --tag tag2\n```\n\n### Phase 5: Summary\n\nAt end of triage, report:\n\n```\n## Audit Summary\n\n### Fixed Now\n- Item 1\n- Item 2\n\n### Tasks Created (N)\n| Task | Priority | Description |\n|------|----------|-------------|\n| @slug | PN | Brief description |\n\n### Inbox Added (N)\n- Item description\n\n### Punted (N)\n- Reason for each\n\n### Remaining\n- Any items not yet triaged\n```\n\nCommit any direct fixes made during audit.\n\n## Key Principles\n\n- **Parallel discovery** - Explore agents run concurrently for speed\n- **Severity-first** - Critical items get triaged first\n- **Interactive triage** - User controls every decision\n- **Immediate execution** - Don't batch, act on each decision\n- **Full tracking** - Everything either fixed or tracked\n- **Clean commit** - End with committed state\n\n## Quick Reference\n\n```bash\n# Get session context first\nkspec session start\n\n# After triage, commit fixes\ngit add -A && git commit -m \"chore: Audit cleanup\"\n\n# Create task with context\nkspec task add --title \"...\" --priority N --slug slug --tag tag\nkspec task note @slug \"Context...\"\n\n# Add to inbox\nkspec inbox add \"Description\" --tag tag\n\n# Check what was created\nkspec task list --status pending\nkspec inbox list\n```\n\n## Exploration Prompts\n\nUse these prompts for the parallel Explore agents:\n\n**Docs:**\n> Review documentation files for cruft in preparation for release. Look for: outdated content, stale TODOs, inconsistencies between docs, wrong command syntax, features documented that don't exist. Check README.md, AGENTS.md, CLAUDE.md, and other .md files.\n\n**Code:**\n> Review src\u002F directory for code cruft. Look for: dead code, unused exports, unresolved TODO\u002FFIXME comments, duplicate functions, commented-out code, debug artifacts, unused imports. Note file paths and line numbers.\n\n**Config:**\n> Review configuration files for cruft. Look for: unused dependencies, misplaced deps (devDeps vs deps), broken npm scripts, stale tsconfig\u002Foxlint\u002Foxfmt options, version inconsistencies. Check package.json, tsconfig.json, and config files.\n\n**Tests:**\n> Review tests\u002F for cruft. Look for: skipped tests (.skip\u002F.todo), dead mock files, duplicate test names, flaky patterns, missing coverage, outdated fixtures. Verify temp directory isolation pattern.\n\n**Specs:**\n> Review .kspec\u002F for spec and task cruft. Look for: stale spec items, orphaned tasks, incorrect statuses, empty stubs, inbox items that should be promoted, tasks with empty slugs.\n\n## Tips\n\n- Run tests before and after to catch regressions\n- Check git status frequently during fixes\n- For complex fixes, create tasks rather than fixing inline\n- Group related fixes in a single commit\n- If unsure about a fix, punt and discuss later\n","\u003C!-- kspec-managed -->\n\n# Codebase Audit\n\nSystematic review of the entire codebase to identify cruft, outdated content, and issues before release. Uses parallel exploration for discovery, then interactive triage for decisions.\n\n## When to Use\n\n- Pre-release cleanup\n- Periodic codebase health check\n- After major refactoring\n- When onboarding to understand technical debt\n\n## Workflow\n\n### Phase 1: Discovery (Parallel Exploration)\n\nLaunch 5 parallel Explore agents to review different areas:\n\n```\nTask tool (subagent_type: Explore) - run all 5 in parallel:\n\n1. DOCS: Review *.md files for outdated content, stale TODOs,\n   inconsistencies, wrong command syntax, aspirational features\n   documented as current\n\n2. CODE: Review src\u002F for dead code, unresolved TODOs\u002FFIXMEs,\n   duplicate functions, commented-out code, debug artifacts,\n   unused imports\u002Fexports\n\n3. CONFIG: Review package.json, tsconfig, etc. for unused deps,\n   misplaced deps (dev vs prod), broken scripts, stale configs,\n   version drift\n\n4. TESTS: Review tests\u002F for skipped tests, dead mocks, duplicate\n   test names, flaky patterns, missing coverage, outdated fixtures\n\n5. SPECS: Review .kspec\u002F for stale spec items, orphaned tasks,\n   status mismatches, empty stubs, inbox items that should be\n   promoted, accumulating session files\n```\n\nEach agent should report findings with:\n- File and location\n- What the issue is\n- Why it's cruft (outdated, redundant, broken, etc.)\n- Recommended action\n\n### Phase 2: Compilation\n\nOrganize all findings by severity:\n\n| Severity | Criteria | Examples |\n|----------|----------|----------|\n| **CRITICAL** | Blocks release, breaks functionality | Runtime deps in devDeps, security vulns, broken required scripts |\n| **HIGH** | Should fix, user-facing or misleading | Wrong docs, incorrect command syntax, misleading task statuses |\n| **MEDIUM** | Recommended, technical debt | Unresolved TODOs, duplicate code, test improvements |\n| **LOW** | Nice to have, minor cleanup | Doc cross-refs, test name clarity, helper consolidation |\n\nPresent compiled summary to user before triage.\n\n### Phase 3: Triage (Interactive)\n\nGo through items one-by-one using AskUserQuestion:\n\n```\nFor each item, present options:\n\n- Fix now      → Execute the fix immediately\n- Task         → Create task with context notes\n- Inbox        → Add to inbox for later review\n- Punt         → Skip, not worth tracking\n- Discuss      → Need more context before deciding\n```\n\n**Triage principles:**\n- One item at a time (don't batch)\n- User decides disposition\n- Execute decisions immediately\n- Add context notes to tasks\u002Finbox\n\n### Phase 4: Execution\n\nFor each decision type:\n\n**Fix now:**\n```bash\n# Delete dead files\nrm path\u002Fto\u002Fdead-file.js\n\n# Quick edits\n# Use Edit tool for small fixes\n\n# Commit cleanup changes at end\ngit add -A && git commit -m \"chore: Audit cleanup - [summary]\"\n```\n\n**Task:**\n```bash\nkspec task add --title \"Description\" --priority N --slug slug-name --tag relevant-tag\nkspec task note @slug \"Context: file location, what needs doing, why\"\n```\n\n**Inbox:**\n```bash\nkspec inbox add \"Description of item for later\" --tag tag1 --tag tag2\n```\n\n### Phase 5: Summary\n\nAt end of triage, report:\n\n```\n## Audit Summary\n\n### Fixed Now\n- Item 1\n- Item 2\n\n### Tasks Created (N)\n| Task | Priority | Description |\n|------|----------|-------------|\n| @slug | PN | Brief description |\n\n### Inbox Added (N)\n- Item description\n\n### Punted (N)\n- Reason for each\n\n### Remaining\n- Any items not yet triaged\n```\n\nCommit any direct fixes made during audit.\n\n## Key Principles\n\n- **Parallel discovery** - Explore agents run concurrently for speed\n- **Severity-first** - Critical items get triaged first\n- **Interactive triage** - User controls every decision\n- **Immediate execution** - Don't batch, act on each decision\n- **Full tracking** - Everything either fixed or tracked\n- **Clean commit** - End with committed state\n\n## Quick Reference\n\n```bash\n# Get session context first\nkspec session start\n\n# After triage, commit fixes\ngit add -A && git commit -m \"chore: Audit cleanup\"\n\n# Create task with context\nkspec task add --title \"...\" --priority N --slug slug --tag tag\nkspec task note @slug \"Context...\"\n\n# Add to inbox\nkspec inbox add \"Description\" --tag tag\n\n# Check what was created\nkspec task list --status pending\nkspec inbox list\n```\n\n## Exploration Prompts\n\nUse these prompts for the parallel Explore agents:\n\n**Docs:**\n> Review documentation files for cruft in preparation for release. Look for: outdated content, stale TODOs, inconsistencies between docs, wrong command syntax, features documented that don't exist. Check README.md, AGENTS.md, CLAUDE.md, and other .md files.\n\n**Code:**\n> Review src\u002F directory for code cruft. Look for: dead code, unused exports, unresolved TODO\u002FFIXME comments, duplicate functions, commented-out code, debug artifacts, unused imports. Note file paths and line numbers.\n\n**Config:**\n> Review configuration files for cruft. Look for: unused dependencies, misplaced deps (devDeps vs deps), broken npm scripts, stale tsconfig\u002Foxlint\u002Foxfmt options, version inconsistencies. Check package.json, tsconfig.json, and config files.\n\n**Tests:**\n> Review tests\u002F for cruft. Look for: skipped tests (.skip\u002F.todo), dead mock files, duplicate test names, flaky patterns, missing coverage, outdated fixtures. Verify temp directory isolation pattern.\n\n**Specs:**\n> Review .kspec\u002F for spec and task cruft. Look for: stale spec items, orphaned tasks, incorrect statuses, empty stubs, inbox items that should be promoted, tasks with empty slugs.\n\n## Tips\n\n- Run tests before and after to catch regressions\n- Check git status frequently during fixes\n- For complex fixes, create tasks rather than fixing inline\n- Group related fixes in a single commit\n- If unsure about a fix, punt and discuss later\n",{"name":5,"description":7},"main","c4b0de3035f8512bba7eb0a9b4e39ea2fbd3f08e",".claude\u002Fskills\u002Faudit",2,"https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec","https:\u002F\u002Fgithub.com\u002Flepahc","2026-07-22 18:57:40",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add lepahc\u002Fkynetic-spec --skill audit",{"tool":43,"command":44},"bunx","bunx skills add lepahc\u002Fkynetic-spec --skill audit",{"tool":46,"command":47},"pnpm","pnpm dlx skills add lepahc\u002Fkynetic-spec --skill audit","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Faudit\u002FSKILL.md",true,[],[52,70,75,83,94,102],{"skillNo":53,"name":54,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":67,"publishedAt":25},"01M6C7KCZE43MDZ74FV3A5TDCQ","writing-specs","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Fwriting-specs\u002FSKILL.md",[57,62],{"categoryNo":58,"categoryName":59,"taxonomyType":60,"taxonomyTypeDesc":61},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":63,"categoryName":64,"taxonomyType":65,"taxonomyTypeDesc":66},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[68,69],"writing","writing specs",{"skillNo":71,"name":19,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":25},"01M6C7KAK9CBJ87PX3BCW7B7YG","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Ftriage\u002FSKILL.md",[],[19],{"skillNo":76,"name":77,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":25},"01M6C7K89VN6FJCT8M86K56EE8","triage-inbox","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Ftriage-inbox\u002FSKILL.md",[],[19,81,82],"inbox","triage inbox",{"skillNo":84,"name":85,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":91,"publishedAt":25},"01M6C7K6F2342AC1TJCN9PM0ZB","triage-automation","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Ftriage-automation\u002FSKILL.md",[88],{"categoryNo":89,"categoryName":90,"taxonomyType":60,"taxonomyTypeDesc":61},"01JBSEEDCAT000000000000023","Skill与智能体开发",[19,92,93],"automation","triage automation",{"skillNo":95,"name":96,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":25},"01M6C7K3TFQ88QQBW3PXYDHQG2","task-work","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Ftask-work\u002FSKILL.md",[],[100,101],"work","task work",{"skillNo":103,"name":104,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":25},"01M6C7K168CH3XJCC3PDGA9H6C","review","https:\u002F\u002Fgithub.com\u002Flepahc\u002Fkynetic-spec\u002Fblob\u002Fmain\u002Ftemplates\u002Fskills\u002Freview\u002FSKILL.md",[],[104]]