This is a solid code analysis tool that goes beyond basic linting to catch the stuff that slips through: code smells like long functions and magic numbers, security risks like SQL injection patterns and hardcoded secrets, and statistical outliers using Z-score and IQR methods. It's read-only and generates reports with severity levels, so you can run it during code reviews or as part of CI without worrying about it changing anything. The built-in false positive awareness is a nice touch since pattern matching tools tend to be noisy. Best used when you're reviewing a PR, preparing to refactor a messy codebase, or setting up automated security scans. It won't replace dedicated security tools, but it's a good first pass.
npx -y skills add supercent-io/skills-template --skill pattern-detection --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
hoodini/ai-agents-skills
addyosmani/agent-skills
giuseppe-trisciuoglio/developer-kit
agamm/claude-code-owasp