Claude Code Security Review is Anthropic’s automated security review for Claude Code, announced in August 2025 in Automate security reviews with Claude Code. The name is the one under which the announcement links the features’ documentation; the post itself speaks of automated security reviews in two forms: a /security-review command that a developer runs in Claude Code from the terminal, and a GitHub Action that runs on pull requests. In both, Claude is asked to identify security concerns in code and can then be asked to fix them.
It addresses the concern that as developers rely on AI to ship faster and build more complex systems, code security becomes more critical. Anthropic positions it as a way to integrate Security Code Review into existing workflows so that vulnerabilities are caught before they reach production.
Capabilities
The /security-review command runs an ad-hoc analysis before code is committed: Claude searches the codebase for potential vulnerabilities and gives detailed explanations of what it finds. It uses a specialized security-focused prompt that checks for common vulnerability patterns, including SQL injection risks, cross-site scripting, authentication and authorization flaws, insecure data handling and dependency vulnerabilities. Claude Code can then be asked to implement a fix for each issue. The command is available by updating Claude Code to its latest version, and it can be customized.
The GitHub Action triggers automatically when a pull request is opened, reviews the changes for security vulnerabilities, applies customizable rules to filter out false positives and known issues, and posts inline comments on the pull request describing each concern with recommended fixes.
Adoption & Ecosystem
Anthropic announced both features as available to all Claude Code users, and describes the action as integrating with an existing CI/CD pipeline and being customizable to a team’s security policies. It reports using the action on its own code, including Claude Code itself, and gives two examples in which it caught a vulnerability before the pull request was merged or the code shipped: a remote code execution vulnerability exploitable through DNS rebinding in an internal tool that started a local HTTP server, and an SSRF vulnerability in a proxy built to manage internal credentials.