Claude Code's new security plugin scans your code before you commit
Anthropic has widened access to the Claude Security plugin for the Claude Code CLI. It checks your changes for vulnerabilities before a commit or runs a full scan across the whole codebase, straight from the terminal on the same Claude inference you're already running.
Per Anthropic's docs and the GitHub repo, the plugin looks for issues like injection flaws, unsafe deserialization, and unsafe DOM APIs. Once installed it runs automatically, with no separate command to invoke.
The security-guidance plugin has been in Claude's official plugin repo since November 2025, and the Claude Security public beta opened on April 30, 2026. This wider rollout follows Anthropic's real-time cybersecurity work on Opus and Sonnet.
Related stories
- Anthropic's automated reviews now flag 54% of PRs
- Anthropic adds real-time cybersecurity filtering to Opus and Sonnet
- One git call let a repo escape the Claude Code sandbox
- Stolen session keys are draining Claude Max accounts
- Breaking Claude Code Opus 5 Auto Mode
- Claude Opus broke four SAML libraries in a month
Comments
No comments yet. Be the first.
Join the conversation
Sign in with Google to leave a comment. Your name and avatar come from your Google profile, and the comment appears after moderation.