security-guidance
Security warnings on edits plus LLM diff review
What it is
Three layers: instant pattern warnings on Edit/Write for about 25 risky constructs, an LLM review of the diff when Claude finishes a turn, and an agentic commit reviewer for injection, XSS, SSRF and hardcoded secrets.
Good for
- Catch unsafe deserialization
- Review a diff for vulnerabilities
- Flag hardcoded secrets
Install or run
/plugin install security-guidance@claude-plugins-officialCommand as given in the project README. Read the repository before you run anything on your machine.
Worth knowingPattern checks produce warnings, not guarantees; LLM review adds cost per turn.
Sources
- https://github.com/anthropics/claude-plugins-official/tree/main/plugins/security-guidance (project README, read 13 Nov 2026)
- https://api.github.com/repos/anthropics/claude-plugins-official (GitHub API, read 13 Nov 2026)
The summary is ours, written from the README. Spotted an error? See how we work.