Scans your codebase for custom utility and integration modules over 100 lines, figures out what they actually do by reading the code, then hunts down open source alternatives through MCP research tools. It evaluates candidates based on GitHub stars, maintenance activity, licenses, and CVE status, then scores each replacement as high, medium, or low confidence. You get a markdown report with evidence plus a JSON summary. Useful when you're inheriting a project with lots of homegrown helpers or trying to reduce maintenance burden by swapping custom code for established libraries. The goal-based analysis approach (not just pattern matching) is the right call here.
npx skills add https://github.com/levnikolaevich/claude-code-skills --skill ln-645-open-source-replacer