Anthropic Launches AI Code Review Tool for Pull Requests

Anthropic on Monday introduced Code Review, a new AI-powered tool integrated into its Claude Code platform, designed to automate the analysis of pull requests and identify bugs in code generated by AI assistants. The feature, available in research preview for Claude for Teams and Enterprise customers, targets large enterprises grappling with surging volumes of code submissions.

Pull requests, which allow developers to propose code changes for review before integration, have proliferated as AI tools like Claude Code accelerate software development. According to Anthropic, this has created bottlenecks for human reviewers, prompting the company to deploy a multiagent system that examines codebases in parallel. Multiple AI agents scrutinize the code from various perspectives, focusing primarily on logical errors rather than stylistic issues. A verification process filters out false positives, and a final agent aggregates findings, removes duplicates, and ranks them by severity using color labels: red for critical problems, yellow for potential issues, and purple for preexisting bugs in the codebase.

The tool integrates directly with GitHub, where administrators can install it via...