Skill Detail

Gate pull requests with targeted diff-aware AI security review using Claude Code Security Review

Run a Claude Code powered security review pass on trusted pull requests so suspicious auth, secret, injection, and unsafe logic changes surface before merge.

Security & VerificationClaude Code
Security & Verification Claude Code Security Reviewed
⭐ 4.3k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill gate-pull-requests-with-targeted-diff-aware-ai-security-review-using-claude-code-security-review Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
GitHub Actions, Claude API access, pull request workflows on trusted repositories
Install & setup
Add anthropics/claude-code-security-review to a GitHub Actions workflow that runs on trusted pull_request events, then configure the Claude API secret and any optional scan or filtering inputs.
Author
Anthropic
Publisher
Organization
Last updated
Apr 18, 2026
Quick brief

This skill is for repositories that want a dedicated security-focused AI review pass inside GitHub Actions. It covers the workflow of triggering on trusted pull requests, scanning only the changed code, tuning false-positive filtering, and posting findings back to the pull request for human review.

How it works

What this skill actually does

Invoke this instead of using Claude Code manually when you need a repeatable, CI-native security checkpoint rather than an informal one-off review. It fits best when maintainers want security findings attached to the PR discussion and governed by normal merge policy.

The scope boundary is tight: this is not a general Claude Code listing and not a generic SAST product card. It is specifically the GitHub Action workflow for AI-assisted security review of pull request diffs.