TL;DR: On October 15, Anthropic moved Code Review for Claude Code out of research preview. It’s available to Claude Team and Enterprise organizations with GitHub repositories, and most reviews now finish in about 6-10 minutes, down from about 20 in the preview. Admins pick Quick, Standard or Deep per repository. It’s billed separately through usage credits, and it isn’t available with Zero Data Retention, the HIPAA configuration or customer-managed encryption keys.

What happened

Anthropic announced that “Code Review is out of research preview and available to Claude Team and Enterprise organizations.” Once an admin turns it on, Claude reviews new pull requests in the selected GitHub repositories automatically.

According to Anthropic’s announcement and docs:

  • Faster reviews. “Most reviews now finish about 3x faster (~20 minutes down to 6-10 minutes) compared to during preview.”
  • Review depth per repository. Admins choose Quick, Standard (the default) or Deep for each repository.
  • How it works. Several agents read each change with the surrounding code, a verification step checks findings, and results arrive as inline comments ranked by severity, plus a summary and a check run. “A review never approves or blocks the pull request.”
  • Where it works. Repositories on GitHub, including GitHub Enterprise Server and, according to the docs, GitHub Enterprise Cloud with data residency (Anthropic’s help center still lists data-residency ghe.com tenants as unsupported).
  • Credits offer. “Eligible Claude Enterprise organizations can claim at least $10,000 in Code Review credits by January 15, 2027. Credits expire 30 days after you claim them. Terms apply.”

Why it matters

If your team ships more code with AI agents, review becomes the slow step. Code Review is Anthropic’s managed answer to that, and the GA version adds the control teams asked for: a light pass on routine repos and a deep one where mistakes are expensive.

The limits decide whether you can use it at all. Anthropic’s help center says Code Review usage is “billed separately through usage credits and doesn’t count against your plan’s included usage.” The docs say it isn’t available to organizations with Zero Data Retention or the HIPAA configuration, isn’t covered by Anthropic’s BAA, and isn’t yet available with customer-managed encryption keys (CMEK). It works with GitHub only; for GitLab, the docs list running Claude in your own CI pipeline instead.

What to do

  1. Check eligibility first. You need a Team or Enterprise plan, GitHub repos, and none of the excluded setups (ZDR, HIPAA, CMEK).
  2. Find the right person. An Owner of your Claude organization turns it on, and they also need permission to install GitHub Apps in your GitHub organization.
  3. Start with a few repos. Pick the repos where bugs cost most, set them to Deep, and leave routine ones on Quick or Standard.
  4. Watch the cost. Anthropic’s help center says each review averages $15–25, scaling with PR size, and usage comes from separate credits. Set a monthly spend cap and check spend after the first week before turning it on everywhere.
  5. Not on Team or Enterprise? Anthropic’s docs say anyone can run /code-review in a Claude Code session on any plan. See our Claude and Codex usage limits guide for how plan limits work.

Sources & method

Sources

Method. Last checked October 18, 2026. Details and quotes come from Anthropic’s announcement, docs and help center. We haven’t tested Code Review. No affiliate links on this page. How we work.

Spotted an error? See our corrections policy and tell us.