Claude Code Review Is Generally Available for Team and Enterprise, With Quick, Standard and Deep Levels

Anthropic moved Code Review out of research preview on October 15. It is now generally available to organizations on Claude Team and Enterprise plans. Code Review is part of Claude Code: once an admin turns it on, Claude reviews pull requests in your GitHub repositories and comments on the lines it flags.

How it works

Several agents read the change and the surrounding code in parallel, looking for logic errors, security vulnerabilities and broken edge cases. A verification step checks each finding against the code to filter out false positives, and the rest are ranked by severity. The review posts inline comments, a summary and a check run. It never approves or blocks a pull request, so the decision to merge stays with the team. By default it looks for bugs that would break production, not formatting or missing tests, and teams can tune it with a REVIEW.md file.

What changed at general availability

Speed: Anthropic says most reviews now finish in 6 to 10 minutes, about three times faster than the roughly 20 minutes during the preview.
Review levels: Quick (fastest and cheapest), Standard (the default) or Deep (most thorough, slowest and most expensive), set for the organization and, if needed, for each repository.
When it runs: each repository can review once when a pull request opens, after every push, or only when someone comments "@claude review".

Price and availability

There is no flat price per review. Reviews are billed on token usage through usage credits, separately from the plan's included usage, and the cost depends on the review level, the size of the change and how often reviews run. Admins can set a monthly spend limit. Eligible Enterprise organizations can claim at least $10,000 in Code Review credits by January 15, 2027; the credits expire 30 days after they are claimed. Code Review works with GitHub, including GitHub Enterprise Server, and is not available to organizations with Zero Data Retention or the HIPAA configuration.

GitHub's option, the same day

Also on October 15, GitHub made AI Scan for pull requests generally available. It is narrower: it looks for security vulnerabilities in pull request changes, in languages and frameworks that CodeQL does not cover, such as PHP, Shell, Terraform and Dockerfiles. It is free for public repositories; private ones need GitHub Code Security or Advanced Security, and no Copilot license is required. Its findings are advisory and cannot block merges.

What this means if you're choosing a review tool

If you want a second reviewer for bugs and logic, Claude Code Review covers more ground. Pilot it on a few repositories with reviews set to run once or on request, then decide on "after every push", which multiplies the cost.

If your main concern is security and you already pay for GitHub's security products, AI Scan comes with what you have. Teams already on GitHub Copilot can weigh both: Claude Code vs GitHub Copilot.