SHIPPING
Claude Code Security Review
Choose the review surface first: the session command, the PR GitHub Action or the separate Claude Security product. They run at different times and cover different scopes; findings still need investigation.
On this page
Review surface → finding → validated change
01/security-review is distinct from the Claude Security product.
02Claude Security’s current product documentation describes Enterprise public beta.
03A clean automated report does not certify the application.
Choose the review surface
The command reviews within a Claude Code session; the official Action analyzes PR changes and posts findings. Current Claude Security documentation describes continuous codebase scanning and tracking, now public beta for Enterprise.
Turn findings into reviewable claims
The Action describes contextual analysis and false-positive filtering. For each finding, require the affected path, trigger, impact and remediation rationale. Investigate and validate a proposed fix before treating it as resolved.
Keep independent checks
Official guidance says automated review complements existing security practices and manual review. No findings is not evidence that an application is secure; use tests and specialist review appropriate to the code and deployment.
A few more details
Is /security-review the same as Claude Security?
No. The command is an on-demand session feature; Claude Security is a separate codebase-scanning product with its own availability.
Does a clean report certify the app?
No. Automated analysis can miss issues and needs validation alongside other checks.
Sources & maker notes
This guide is based on the cited documentation and maker sources. vibeforthat has not run an independent installation or end-to-end product test for this article.