ワンクリックで
doc-lint
Use when modifying markdown files, preparing PRs, or suspecting link/structure inconsistencies in documentation.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Use when modifying markdown files, preparing PRs, or suspecting link/structure inconsistencies in documentation.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when Ground Truth alignment finds memory contamination and the user explicitly requests scoped cleanup with /cleanse.
Use when registering, reviewing, merging, or rebuilding HXSK glossary term definitions after glossary-detect suggests a candidate or the user invokes /define.
Use when staged changes exist requiring qlty checks, logical split detection, or HXSK phase-scoped conventional commit creation.
Use when validating code for circular imports, layer violations, or design doc logic errors before merging architecture-level changes.
Use when .hxsk/.bootstrap-version is missing (fresh install) or exists (verify/update) to initialize HExoskeleton.
Use when shell scripts need linting/formatting before execution or commit to ensure code quality.
| description | Use when modifying markdown files, preparing PRs, or suspecting link/structure inconsistencies in documentation. |
| name | doc-lint |
| trigger | 문서 정합성 검사, 깨진 링크, INDEX 동기화, 카운트 불일치, 고아 파일, pre-commit, 문서 구조 검사, 중복 파일, 경로 참조 검증, 내용 검증, 의미론적 검사, doc-lint, doc lint, markdown check, broken link, orphan file, duplicate file, link check, index sync, count mismatch, ref check, structural check, content validation, 문서 검사, markdown 정합성 |
doc-lint.sh 재실행으로 PASS 확인bash .hxsk/scripts/doc-lint.sh
If all PASS, skip to Step 3. If any FAIL, proceed to Step 2.
Fix FAIL items in priority order:
Re-run doc-lint.sh after fixes to confirm PASS.
Dispatch 3 parallel agents for semantic checks:
Agent A: CLAUDE.md + AGENTS.md
Read CLAUDE.md and AGENTS.md. Compare every claim (hook events,
workflow descriptions, file paths, agent boundaries) against
actual project state. Report as [MISMATCH] or [OK].
Agent B: README.md + ARCHITECTURE.md
Read README.md and .hxsk/ARCHITECTURE.md. Verify feature lists,
component descriptions, directory structure diagrams against
actual state. Report as [MISMATCH] or [OK].
Agent C: INDEX files (skills, agents, research)
Read each INDEX.md. Verify listed items match actual files,
descriptions are accurate, and no items are missing.
Report as [MISMATCH] or [OK].
Collect agent results. For each [MISMATCH]:
| Rule | What it checks | Scope |
|---|---|---|
| LINK-01 | Relative link targets exist | All .md |
| INDEX-01 | INDEX.md lists vs actual files | skills/, agents/, research/ |
| COUNT-01 | README count numbers vs actual | README.md |
| REF-01 | Backtick path references in L1 docs | CLAUDE.md, AGENTS.md |
| ORPHAN-01 | Files not referenced anywhere | All .md (excl. memories, templates) |
| DUP-01 | Same filename in multiple locations | All .md (excl. symlinks) |
NO CONTENT VALIDATION WITHOUT STRUCTURAL PASS FIRST NO FIX APPLICATION WITHOUT PRIORITY ORDER ADHERENCE NO COMPLETION WITHOUT RE-RUN LINT CHECK NO EDIT WITHOUT MINIMAL FIX PROPOSAL NO INDEX UPDATE WITHOUT FILE EXISTENCE VERIFICATION