Runs after your implementation tasks are marked Done and calculates a quantitative Code Quality Score based on metrics like cyclomatic complexity, function size, and nesting depth. It checks for DRY/KISS/YAGNI violations, architecture boundary breaks, and uses MCP Ref to validate against external best practices for performance and optimality. Issues are categorized with prefixes like SEC, PERF, ARCH, and MNT, each carrying point penalties that subtract from a base score of 100. The two-layer detection system means it won't flag threshold violations without understanding if they're justified by design. Useful when you want objective quality metrics before code review rather than relying on gut feel alone.
npx skills add https://github.com/levnikolaevich/claude-code-skills --skill ln-511-code-quality-checker