This is a documentation audit coordinator that spins up four specialized workers to check structure, semantic content, code comments, and factual accuracy. It runs mandatory research first (official docs, MCP references, current best practices) before any worker marks something as broken, so you're not fixing things based on hunches. Each audit phase is explicit: discovery, research, delegation, aggregation, then a single consolidated report with a remediation plan. All the temporary worker markdown gets cleaned up automatically. It's built for the evaluation runtime family, so if you're running other 6XX audit skills this slots right in. Honestly feels over-engineered for small projects, but if you're auditing a large codebase where doc drift actually costs you, the research-backed validation is worth it.
npx skills add https://github.com/levnikolaevich/claude-code-skills --skill ln-610-docs-auditor