Anthropic launches code review tool to check flood of AI-generated code

Anthropic has launched Code Review, an artificial intelligence tool to review code changes in pull requests. The product aims to catch bugs and logical errors in code generated by AI tools before the changes enter a software codebase. The tool launched Monday in research preview for Claude for Teams and Claude for Enterprise customers. It integrates with GitHub to automatically analyze pull requests and add comments that explain potential issues along with suggested fixes. Cat Wu, Anthropic head of product, said the product addresses a review bottleneck that has grown as Claude Code has increased code output and pull request volume. Developer leads can enable Code Review to run by default for every engineer on a team. The system focuses on logical errors rather than style. It relies on multiple agents that examine the codebase from different perspectives in parallel, after which a final agent aggregates findings, removes duplicates and ranks issues. Severity appears through color labels, with red marking the highest priority. The tool includes light security analysis and allows customization of additional checks based on internal practices. Pricing is token-based and varies with code complexity. Wu estimated that each review would cost 15 to 25 dollars on average. She said the product targets large enterprise users such as Uber, Salesforce and Accenture.