Official Anthropic documentation for the Claude Security plugin for Claude Code. Multi-agent vulnerability scan inside a Claude Code session: architecture map, threat model, hunt, independent review, then optional patch suggestion. Distinct from managed Claude Security Enterprise product and from security-guidance, /security-review, and PR Code Review. Requires Claude Code v2.1.154+, paid plan with dynamic workflows, Python 3.9.6+. Install: /plugin install claude-security@claude-plugins-official then /reload-plugins. Jobs: Scan codebase, Scan changes (branch/PR/commit), Suggest patches. Patches never auto-applied — git apply from CLAUDE-SECURITY-/patches/. Scans count against plan usage. Results in timestamped CLAUDE-SECURITY- directory with RESULTS.md/jsonl and revision stamp.