What is library-skills?
Library-skills is a Python CLI tool (with a TypeScript twin) that scans your project's installed packages in site-packages or node_modules, discovers bundled AI agent skills from libraries like FastAPI or Streamlit, and symlinks them into .agents/skills for easy access. Run `uvx library-skills` or `npx library-skills` to list available skills, install with `install` or the default sync, and keep them updated automatically as you upgrade deps—no more agent library failed to init errors from outdated patterns. It solves AI agents hallucinating deprecated library usage by pulling official, version-synced instructions directly from agent skills library github repos.
Why is it gaining traction?
Unlike generic prompts or static docs, it delivers library authors' own up-to-date skills embedded in packages, symlinked so they refresh on updates—zero manual chasing claude skills library github examples. The interactive CLI with scan, list, install, and remove commands feels instant, supports Claude Code via `--claude` for .claude/skills, and filters to top-level deps only. Devs love dodging agent library failed to init instrument intellij headaches in mixed Python/JS projects.
Who should use this?
Backend devs building FastAPI apps with Claude or Cursor, wanting agents to spit correct code without jdwp-like init fails. Frontend TS folks using agent library typescript in Node projects tired of rewriting boilerplate. Teams on github library project with Streamlit or anthropic skills library github integrations needing reliable, maintained agent prompts.
Verdict
Grab it for early AI agent workflows—solid docs, full test coverage, and MIT license make it low-risk despite 65 stars and 1.0% credibility score signaling beta maturity. Pair with libraries already shipping skills for quick wins, but watch for broader adoption.
(198 words)
