[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ASM2TQH93S566755B2T4HX":3,"skill-same:01M7ASM2TQH93S566755B2T4HX":77},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":29,"commitSha":30,"skillDirectory":31,"forksCount":32,"repositoryHtmlUrl":33,"repositoryHomepageUrl":6,"ownerHtmlUrl":34,"ownerBlogUrl":6,"ownerDescription":35,"updatedAt":36,"install":37,"resources":50},"01M7ASM2TQH93S566755B2T4HX","docs-drift",null,"Detect documentation drift in a repo's docs\u002F folder — ADRs, glossaries, API specs (OpenAPI), rendered artifacts, READMEs, code comments, runbooks, and caller-migration notes — that has fallen out of sync with the code a PR changes, and flag documentation-completeness gaps (missing CHANGELOG\u002FREADME\u002F docstring updates). Use when reviewing a PR to verify docs still describe the CURRENT model: stale comments describing a removed mechanism, docs asserting a superseded design, a rendered file lagging its source, ADR index rows that don't reflect an amendment or supersede, env\u002Fconfig docs that name the wrong app or variable, missing runbooks for ops-facing changes, or missing caller migration notes for breaking APIs. Use even when the PR doesn't obviously touch docs — code changes silently invalidate docs elsewhere in the tree.","ozzycodes2\u002Fclaudellum","ozzycodes2","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21285094?v=4","01M7ASJ57KTP519F96S4QGM7C5","Osniel Quintana","01M7ASJ589JS8S075AA92Y08S4",1,"MIT","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: docs-drift\ndescription: >\n  Detect documentation drift in a repo's docs\u002F folder — ADRs, glossaries, API\n  specs (OpenAPI), rendered artifacts, READMEs, code comments, runbooks, and\n  caller-migration notes — that has fallen out of sync with the code a PR\n  changes, and flag documentation-completeness gaps (missing CHANGELOG\u002FREADME\u002F\n  docstring updates). Use when reviewing a PR to verify docs still describe the\n  CURRENT model: stale comments describing a removed mechanism, docs asserting a\n  superseded design, a rendered file lagging its source, ADR index rows that\n  don't reflect an amendment or supersede, env\u002Fconfig docs that name the wrong\n  app or variable, missing runbooks for ops-facing changes, or missing caller\n  migration notes for breaking APIs. Use even when the PR doesn't obviously\n  touch docs — code changes silently invalidate docs elsewhere in the tree.\nallowed-tools: Bash Read Grep Glob\nmetadata:\n  author: Osniel Quintana-Salgado\n  facet-of: review\n---\n\n# Documentation Drift & Completeness Check\n\nTwo facets, one skill:\n\n- **Drift** — docs that now *contradict the code* (or each other) after the\n  PR's changes. The hard part: the drift lives in a file the PR did\n  not touch, so a diff-only reviewer never sees it.\n- **Completeness** — expected doc updates the PR *should have made but didn't*\n  (CHANGELOG entry, README\u002FAPI doc for a new surface, docstrings on new public\n  items, light DX: ADR\u002Frunbook\u002Fcaller migration).\n\nThe generative insight (from real multi-round PR reviews): most doc findings\nsurface one commit at a time because each is a *cross-file* claim — a sentence\nin `docs\u002F` asserting a code fact that a later commit quietly falsified. Catch\nthem by extracting every doc claim that references a code fact, then verifying\neach claim against the post-change code in ONE pass.\n\n## Inputs\n\n- **PR URL** — full GitHub\u002FEnterprise URL.\n- **OUTPUT_DIR** (optional) — defaults to a private `mktemp -d`.\n- **Analysis JSON** — `\u003COUTPUT_DIR>\u002Fpr-analysis-\u003CPR_NUMBER>.json` from the\n  `analyze` skill, if present (gives changed files + languages).\n\n## Output\n\nWrite findings to `\u003COUTPUT_DIR>\u002Fdocs-drift-\u003CPR_NUMBER>.json`. Use the template\nin `assets\u002Fdocs-drift.output.json` verbatim as the shape. Findings that map to\na specific doc line carry `path` + `line` so the orchestrator posts them\ninline; cross-file\u002Farchitectural findings omit both and go in the section body.\n\nA stale line you can rewrite outright also carries `suggested_change` — the\ncorrected text for lines `start_line`..`line`, which posts as a committable\nGitHub suggestion. `doc_says` already quotes the line verbatim, so a\nstale-comment, env-mismatch or adr-index finding usually qualifies. A\nrendered-lag finding never does: the fix is regenerating the artifact, not\npatching the rendered line.\n\nBuild the JSON with `jq -n` and `--arg`\u002F`--argjson` so doc text is escaped —\nnever heredoc untrusted strings.\n\n## Workflow\n\nProgress:\n- [ ] Step 1 — Inventory the docs surface (`scripts\u002Fdocs_inventory.py`)\n- [ ] Step 2 — Extract code-referencing claims from docs\n- [ ] Step 3 — Verify each claim against the changed code (plan → validate)\n- [ ] Step 4 — ADR-specific cross-checks (index ↔ files ↔ status)\n- [ ] Step 5 — Rendered-artifact staleness (source ↔ generated)\n- [ ] Step 6 — Completeness gaps (+ light DX fold)\n- [ ] Step 7 — Score + emit JSON\n- [ ] Step 8 — Validate every finding, drop the confirmations, then write it out\n\n### Step 0: Parse arguments\n\n```bash\nPR_URL=\"$1\"\nOUTPUT_DIR=\"${2:-$(mktemp -d -t pr-review-XXXXXX)}\"\nPR_NUMBER=$(echo \"$PR_URL\" | grep -oE 'pull\u002F[0-9]+' | grep -oE '[0-9]+')\nREPO_ROOT=\"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || echo .)\"\n```\n\n### Step 1: Inventory the docs surface\n\n```bash\npython3 scripts\u002Fdocs_inventory.py --repo-root \"$REPO_ROOT\" > \"$OUTPUT_DIR\u002Fdocs-inventory.json\"\nif [ ! -s \"$OUTPUT_DIR\u002Fpr.diff\" ]; then   # standalone; \u002Freview:review captures these\n  gh pr diff \"$PR_URL\" > \"$OUTPUT_DIR\u002Fpr.diff\"\n  gh pr diff \"$PR_URL\" --name-only > \"$OUTPUT_DIR\u002Fpr.files\"\nfi\n```\n\n`pr.diff` and `pr.files` are shared with the passes running beside this one.\nRead them; never write to them.\n\n### Step 2: Extract claims\n\nRead `references\u002Fdrift-patterns.md` — authoritative catalog of drift\nclasses. Work through it; do not improvise categories.\n\nA **claim** is any doc sentence, table row, code comment, or config comment\nasserting a *checkable code fact*. Collect claims from every doc the inventory\nfound — not only docs in the PR diff.\n\n### Step 3: Verify each claim against the changed code\n\nFor every claim, locate the code fact in the post-change tree and decide\n`intact | drifted | uncertain`. Prefer grepping the repo at HEAD over memory.\nA claim is **drifted** when the asserted fact is absent, renamed, or replaced.\nRecord doc line AND contradicting code location in the finding body.\n\n### Step 4: ADR cross-checks\n\nIf `docs\u002Fadrs\u002F` exists, verify ADR bookkeeping the inventory extracted (status\nreflected in index, supersede links bidirectional, model changes retire old docs).\n\n### Step 5: Rendered-artifact staleness\n\nFor each source→rendered pair: if source changed but rendered did not (or still\ncontains old strings), flag it. Prefer \"regenerate\" when CI\u002Fhooks own the artifact.\n\n### Step 6: Completeness gaps\n\n- New public surface (endpoint, exported type, CLI flag, env var) with no\n  matching README\u002FAPI-doc\u002F`dot-env` entry.\n- Behavior change with no CHANGELOG entry (only if the repo keeps one).\n- New public items missing docstrings (only to the repo's existing bar).\n\n#### DX docs fold (light touch)\n\nWhen the PR changes behavior that operators or callers must know about, also check:\n\n- **ADRs** — behavior\u002Farchitecture shift without ADR add\u002Famend\u002Fsupersede when the repo keeps ADRs\n- **Runbooks \u002F ops notes** — deploy, rollback, feature-flag, or on-call steps when ops behavior changes\n- **Caller migration** — breaking internal\u002Fpublic APIs without examples, codemods, or deprecation notes for callers\n\nKeep this light: only flag when the gap would mislead a teammate or on-call. Do not invent a docs policy the repo does not hold.\n\n### Step 7: Score & emit\n\nSeverity: **high** = a doc actively misleads; **medium** = stale but not\ndangerous; **low** = cosmetic\u002Findex bookkeeping. Emit JSON per the asset template.\n\n### Step 8: Validation loop\n\nFor each finding: (1) quotes the doc line, (2) names contradicting code or missing\nartifact, (3) severity matches rubric. Drop unverifiable claims.\n\n**Then drop every finding whose remedy is a confirmation.** `fix` works this\nlist to completion, so a finding with nothing to edit becomes a deferral that\ncosts a routing decision and a report line and changes no file. Drop it when:\n\n- the remedy's verb is *confirm*, *verify*, *check*, *ensure*, *consider*,\n  *be aware*, or *no action needed*;\n- it restates that the doc and the code agree;\n- it reports an absence the repo has no policy about (no CHANGELOG in a repo\n  that keeps none, no ADR for a change the ADR index does not cover);\n- it is cosmetic drift in a generated artifact that the next regeneration fixes.\n\nKeep the finding when a reader of the current doc would act wrongly.\n\nThen write `docs-drift-\u003CPR_NUMBER>.json`.\n\n## mac-ui repos (Web Components)\n\nWhen `analyze` set `uses_mac_ui: true`, resolve docs live. Read\n`references\u002Fmac-ui-docs.md` for the fetch recipe.\n\n## Gotchas\n\n- Drift usually lives OUTSIDE the diff. Inventory the whole `docs\u002F` tree.\n- Rendered artifacts: fix is regeneration, not hand-edit.\n- Code comments count as docs.\n- Don't flag drift from memory — mark `uncertain` when greps are inconclusive.\n\n## Files\n\n- `references\u002Fdrift-patterns.md` — drift-class catalog (read in Step 2).\n- `references\u002Fmac-ui-docs.md` — how to fetch mac-ui's live docs.\n- `assets\u002Fdocs-drift.output.json` — output JSON template.\n- `scripts\u002Fdocs_inventory.py` — docs\u002FADR\u002Frendered inventory (`--help`).\n- `evals\u002Fevals.json` — trigger + output evals.\n","# Documentation Drift & Completeness Check\n\nTwo facets, one skill:\n\n- **Drift** — docs that now *contradict the code* (or each other) after the\n  PR's changes. The hard part: the drift lives in a file the PR did\n  not touch, so a diff-only reviewer never sees it.\n- **Completeness** — expected doc updates the PR *should have made but didn't*\n  (CHANGELOG entry, README\u002FAPI doc for a new surface, docstrings on new public\n  items, light DX: ADR\u002Frunbook\u002Fcaller migration).\n\nThe generative insight (from real multi-round PR reviews): most doc findings\nsurface one commit at a time because each is a *cross-file* claim — a sentence\nin `docs\u002F` asserting a code fact that a later commit quietly falsified. Catch\nthem by extracting every doc claim that references a code fact, then verifying\neach claim against the post-change code in ONE pass.\n\n## Inputs\n\n- **PR URL** — full GitHub\u002FEnterprise URL.\n- **OUTPUT_DIR** (optional) — defaults to a private `mktemp -d`.\n- **Analysis JSON** — `\u003COUTPUT_DIR>\u002Fpr-analysis-\u003CPR_NUMBER>.json` from the\n  `analyze` skill, if present (gives changed files + languages).\n\n## Output\n\nWrite findings to `\u003COUTPUT_DIR>\u002Fdocs-drift-\u003CPR_NUMBER>.json`. Use the template\nin `assets\u002Fdocs-drift.output.json` verbatim as the shape. Findings that map to\na specific doc line carry `path` + `line` so the orchestrator posts them\ninline; cross-file\u002Farchitectural findings omit both and go in the section body.\n\nA stale line you can rewrite outright also carries `suggested_change` — the\ncorrected text for lines `start_line`..`line`, which posts as a committable\nGitHub suggestion. `doc_says` already quotes the line verbatim, so a\nstale-comment, env-mismatch or adr-index finding usually qualifies. A\nrendered-lag finding never does: the fix is regenerating the artifact, not\npatching the rendered line.\n\nBuild the JSON with `jq -n` and `--arg`\u002F`--argjson` so doc text is escaped —\nnever heredoc untrusted strings.\n\n## Workflow\n\nProgress:\n- [ ] Step 1 — Inventory the docs surface (`scripts\u002Fdocs_inventory.py`)\n- [ ] Step 2 — Extract code-referencing claims from docs\n- [ ] Step 3 — Verify each claim against the changed code (plan → validate)\n- [ ] Step 4 — ADR-specific cross-checks (index ↔ files ↔ status)\n- [ ] Step 5 — Rendered-artifact staleness (source ↔ generated)\n- [ ] Step 6 — Completeness gaps (+ light DX fold)\n- [ ] Step 7 — Score + emit JSON\n- [ ] Step 8 — Validate every finding, drop the confirmations, then write it out\n\n### Step 0: Parse arguments\n\n```bash\nPR_URL=\"$1\"\nOUTPUT_DIR=\"${2:-$(mktemp -d -t pr-review-XXXXXX)}\"\nPR_NUMBER=$(echo \"$PR_URL\" | grep -oE 'pull\u002F[0-9]+' | grep -oE '[0-9]+')\nREPO_ROOT=\"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || echo .)\"\n```\n\n### Step 1: Inventory the docs surface\n\n```bash\npython3 scripts\u002Fdocs_inventory.py --repo-root \"$REPO_ROOT\" > \"$OUTPUT_DIR\u002Fdocs-inventory.json\"\nif [ ! -s \"$OUTPUT_DIR\u002Fpr.diff\" ]; then   # standalone; \u002Freview:review captures these\n  gh pr diff \"$PR_URL\" > \"$OUTPUT_DIR\u002Fpr.diff\"\n  gh pr diff \"$PR_URL\" --name-only > \"$OUTPUT_DIR\u002Fpr.files\"\nfi\n```\n\n`pr.diff` and `pr.files` are shared with the passes running beside this one.\nRead them; never write to them.\n\n### Step 2: Extract claims\n\nRead `references\u002Fdrift-patterns.md` — authoritative catalog of drift\nclasses. Work through it; do not improvise categories.\n\nA **claim** is any doc sentence, table row, code comment, or config comment\nasserting a *checkable code fact*. Collect claims from every doc the inventory\nfound — not only docs in the PR diff.\n\n### Step 3: Verify each claim against the changed code\n\nFor every claim, locate the code fact in the post-change tree and decide\n`intact | drifted | uncertain`. Prefer grepping the repo at HEAD over memory.\nA claim is **drifted** when the asserted fact is absent, renamed, or replaced.\nRecord doc line AND contradicting code location in the finding body.\n\n### Step 4: ADR cross-checks\n\nIf `docs\u002Fadrs\u002F` exists, verify ADR bookkeeping the inventory extracted (status\nreflected in index, supersede links bidirectional, model changes retire old docs).\n\n### Step 5: Rendered-artifact staleness\n\nFor each source→rendered pair: if source changed but rendered did not (or still\ncontains old strings), flag it. Prefer \"regenerate\" when CI\u002Fhooks own the artifact.\n\n### Step 6: Completeness gaps\n\n- New public surface (endpoint, exported type, CLI flag, env var) with no\n  matching README\u002FAPI-doc\u002F`dot-env` entry.\n- Behavior change with no CHANGELOG entry (only if the repo keeps one).\n- New public items missing docstrings (only to the repo's existing bar).\n\n#### DX docs fold (light touch)\n\nWhen the PR changes behavior that operators or callers must know about, also check:\n\n- **ADRs** — behavior\u002Farchitecture shift without ADR add\u002Famend\u002Fsupersede when the repo keeps ADRs\n- **Runbooks \u002F ops notes** — deploy, rollback, feature-flag, or on-call steps when ops behavior changes\n- **Caller migration** — breaking internal\u002Fpublic APIs without examples, codemods, or deprecation notes for callers\n\nKeep this light: only flag when the gap would mislead a teammate or on-call. Do not invent a docs policy the repo does not hold.\n\n### Step 7: Score & emit\n\nSeverity: **high** = a doc actively misleads; **medium** = stale but not\ndangerous; **low** = cosmetic\u002Findex bookkeeping. Emit JSON per the asset template.\n\n### Step 8: Validation loop\n\nFor each finding: (1) quotes the doc line, (2) names contradicting code or missing\nartifact, (3) severity matches rubric. Drop unverifiable claims.\n\n**Then drop every finding whose remedy is a confirmation.** `fix` works this\nlist to completion, so a finding with nothing to edit becomes a deferral that\ncosts a routing decision and a report line and changes no file. Drop it when:\n\n- the remedy's verb is *confirm*, *verify*, *check*, *ensure*, *consider*,\n  *be aware*, or *no action needed*;\n- it restates that the doc and the code agree;\n- it reports an absence the repo has no policy about (no CHANGELOG in a repo\n  that keeps none, no ADR for a change the ADR index does not cover);\n- it is cosmetic drift in a generated artifact that the next regeneration fixes.\n\nKeep the finding when a reader of the current doc would act wrongly.\n\nThen write `docs-drift-\u003CPR_NUMBER>.json`.\n\n## mac-ui repos (Web Components)\n\nWhen `analyze` set `uses_mac_ui: true`, resolve docs live. Read\n`references\u002Fmac-ui-docs.md` for the fetch recipe.\n\n## Gotchas\n\n- Drift usually lives OUTSIDE the diff. Inventory the whole `docs\u002F` tree.\n- Rendered artifacts: fix is regeneration, not hand-edit.\n- Code comments count as docs.\n- Don't flag drift from memory — mark `uncertain` when greps are inconclusive.\n\n## Files\n\n- `references\u002Fdrift-patterns.md` — drift-class catalog (read in Step 2).\n- `references\u002Fmac-ui-docs.md` — how to fetch mac-ui's live docs.\n- `assets\u002Fdocs-drift.output.json` — output JSON template.\n- `scripts\u002Fdocs_inventory.py` — docs\u002FADR\u002Frendered inventory (`--help`).\n- `evals\u002Fevals.json` — trigger + output evals.\n",{"name":5,"metadata":24,"description":27,"allowed-tools":28},{"author":25,"facet-of":26},"Osniel Quintana-Salgado","review","Detect documentation drift in a repo's docs\u002F folder — ADRs, glossaries, API specs (OpenAPI), rendered artifacts, READMEs, code comments, runbooks, and caller-migration notes — that has fallen out of sync with the code a PR changes, and flag documentation-completeness gaps (missing CHANGELOG\u002FREADME\u002F docstring updates). Use when reviewing a PR to verify docs still describe the CURRENT model: stale comments describing a removed mechanism, docs asserting a superseded design, a rendered file lagging its source, ADR index rows that don't reflect an amendment or supersede, env\u002Fconfig docs that name the wrong app or variable, missing runbooks for ops-facing changes, or missing caller migration notes for breaking APIs. Use even when the PR doesn't obviously touch docs — code changes silently invalidate docs elsewhere in the tree.\n","Bash Read Grep Glob","main","a3f19bcc313571bff79dca391a83f9c4d38bb335","review\u002Fskills\u002Fdocs-drift",0,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum","https:\u002F\u002Fgithub.com\u002Fozzycodes2","Software Engineer | Engineering Manager | AI whisperer | ☕Enjoyer","2026-11-14 04:50:11",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add ozzycodes2\u002Fclaudellum --skill docs-drift",{"tool":43,"command":44},"bunx","bunx skills add ozzycodes2\u002Fclaudellum --skill docs-drift",{"tool":46,"command":47},"pnpm","pnpm dlx skills add ozzycodes2\u002Fclaudellum --skill docs-drift","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002FSKILL.md",true,[51,57,63,67,71],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"assets\u002Fdocs-drift.output.json","ASSET","资产文件",1185,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002Fassets\u002Fdocs-drift.output.json",{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"evals\u002Fevals.json","REFERENCE","参考文档",1938,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002Fevals\u002Fevals.json",{"path":64,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Fdrift-patterns.md",3677,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002Freferences\u002Fdrift-patterns.md",{"path":68,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":69,"sourceHtmlUrl":70},"references\u002Fmac-ui-docs.md",1650,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002Freferences\u002Fmac-ui-docs.md",{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"scripts\u002Fdocs_inventory.py","SCRIPT","脚本",6086,"https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Freview\u002Fskills\u002Fdocs-drift\u002Fscripts\u002Fdocs_inventory.py",[78,85,92,99,106,113],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7ASQB3CTKMKDNAXD9DCF1YC","writing-design-docs","Use when the user wants to write a problem-statement design doc, says \"let's plan a feature properly\" \u002F \"write a design doc\" \u002F \"I'm thinking about how to do X\" for a non-trivial X, or before committing to a non-obvious technical direction","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Ftechnical-writing\u002Fskills\u002Fwriting-design-docs\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M7ASQ6JD4QRV5ZKK2QME401Y","write-doc","Use when the user asks to write or draft a README, ADR, runbook, tutorial, how-to, explanation, or reference doc, or a caller passes \"type ADR\" or \"type README\". Picks the type, fills a template, adds diagrams. RFCs and design docs belong to writing-design-docs.","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Ftechnical-writing\u002Fskills\u002Fwrite-doc\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7ASQ1VWNVP1RE7M5D4G2QMG","review-doc","Use when the user asks to review, critique, score, grade, or \"how good is this doc\" for a Markdown doc or pasted text. Read-only. Scores it on the 20 Practical Prose dimensions and emits a findings table. Does not edit; fix-doc applies the findings next.","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Ftechnical-writing\u002Fskills\u002Freview-doc\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":19},"01M7ASPWST208JZNR7HF7J6KMV","fix-doc","Use when the user asks to fix, edit for prose quality, polish, tighten, humanize, or de-slop a Markdown doc or pasted text, or says \"make this read less AI\". Applies review-doc findings, generates flagged diagrams, lists substantive issues for the author, then runs the AI-tell pass last. Modifies the target.","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Ftechnical-writing\u002Fskills\u002Ffix-doc\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":19},"01M7ASPS6Y9ZGK9KCDHNPDR75A","create-slides","Use when the user asks for a slide deck, presentation, talk, or slides from a document, report, PR, notes, or topic. Builds one self-contained Machinify-branded HTML deck in a minimalist style, with headline-plus-visual slides, speaker notes, and every diagram drawn by architecture-design:create-diagram.","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Ftechnical-writing\u002Fskills\u002Fcreate-slides\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":19},"01M7ASPAKFY045KHR92Y4QFTWV","start-session","Open a Claude Code working session in a fresh, isolated git worktree, so the user can start a task with a single \u002Fstart-session \u003Cname>. Use at the START of a session when the user says \"start session\", \"start a worktree\", \"spin up a worktree for ABC-123\", \"begin work on \u003Cticket>\", or passes a name\u002Fticket to work under. Creates and ENTERS a worktree via the EnterWorktree tool (never `git worktree add`) under .claude\u002Fworktrees\u002F, named exactly after the argument, on a local branch `worktree-\u003Cargument>`. Reports what the worktree inherited (`.worktreeinclude` copies, symlinked node_modules) and whether its `.env` still points at the main checkout's stack. If the argument is a JIRA key (e.g. ABC-123, PROJ-1505), also loads that story with the Atlassian CLI (`acli`) when available and summarizes the implementation ask. Also renames the session to the argument (the same effect as `\u002Frename \u003CNAME>`) so it's easy to find in `\u002Fresume`. Then stops and waits — it sets up the workspace and context, it does not implement.","https:\u002F\u002Fgithub.com\u002Fozzycodes2\u002Fclaudellum\u002Fblob\u002Fmain\u002Fsession-mgmt\u002Fskills\u002Fstart-session\u002FSKILL.md",[],[]]