Claude Code Review reaches GA

Anthropic’s Claude Code Review is now generally available, according to a LinkedIn post from Morgan Lunt, who works on Claude Code. That makes it a more concrete option for companies assessing automated reviews of pull requests. The post doesn’t provide pricing or adoption figures, so buyers still don’t have those details to compare.

Lunt says Anthropic built the reviewer using evaluations against real incidents and tests of different agent setups. He also says the team focused on limiting minor comments and returning findings together, rather than drip-feeding them. Those are the right problems to pay attention to: a review tool needs to surface useful issues without adding a stream of noise. Lunt says the tool is now about three times faster and much less expensive, though the post doesn’t give figures or explain the comparison.

That’s a shipping milestone, not proof that the tool will fit every team’s review process. Companies will need to assess the quality of its findings on their own code and decide how it fits alongside existing checks and human review. This is the kind of thing we build into an AI command centre, where operators can keep track of new tools and evaluate where they may help. Read Lunt’s announcement.