Consolidates documentation sprawl in repos where LLMs have generated drafts, research notes, and overlapping specs. The workflow is surgical: inventory every doc, mark lifecycle metadata (draft/canonical/integrated/superseded), merge duplicates into one canonical source per topic, then delete integrated files on schedule. Includes templates for READMEs, ADRs, API docs, and changelogs. The standout insight is treating stale docs as agent bugs, since outdated information causes AI assistants to reimplement existing features or fix already-solved problems. Best for teams drowning in AI-generated documentation that needs authoritative pruning without losing useful content to git history alone.
npx skills add https://github.com/vasilyu1983/ai-agents-public --skill docs-codebase