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

Anthropic has launched Code Review, a new AI-powered tool designed to address the growing bottleneck created by AI-generated code. The product, available in research preview for Claude for Teams and Claude for Enterprise customers, automatically analyzes pull requests in GitHub, leaving comments that explain potential logical errors and suggest fixes. This directly responds to enterprise leaders' concerns about efficiently reviewing the flood of code produced by Claude Code, which has seen its run-rate revenue surpass $2.5 billion and caused a significant increase in pull requests.

Code Review employs a multi-agent architecture where different AI agents examine code from various perspectives in parallel. A final agent aggregates and ranks the findings, removing duplicates and prioritizing the most critical issues. The tool focuses exclusively on logical errors over style to provide immediately actionable feedback, labeling severity with a color system: red for highest priority, yellow for potential problems, and purple for issues tied to preexisting code. While it provides a light security analysis, Anthropic directs users needing deeper scrutiny to its separate Claude Code Security product. The launch comes as Anthropic's enterprise subscriptions have quadrupled since the start of the year, making tools that manage AI productivity at scale crucial for its core business with clients like Uber, Salesforce, and Accenture.