Anthropic Unveils AI-Powered Code Review Tool for Claude Code

Anthropic has introduced Code Review, an AI-powered code review tool for users of its Claude Code platform, in a move to address increasing bottlenecks in reviewing AI-generated code. The tool is currently available in research preview to Claude for Teams and Claude for Enterprise customers and is intended for large-scale enterprise deployment.

Code Review is designed to automatically analyze pull requests—formal requests to merge code changes—primarily focusing on detecting logical errors rather than stylistic issues. According to Cat Wu, Anthropic’s head of product, the system highlights problems by severity using a color-coded approach: red for critical issues, yellow for potential problems, and purple for preexisting concerns in the codebase.

"We decided we're going to focus purely on logic errors. This way we're catching the highest priority things to fix," Wu said. By concentrating on logic flaws, Anthropic aims to provide developers with actionable feedback that reduces the risk of introducing significant bugs during new feature development.

The tool integrates directly with GitHub, one of the world’s leading code repository platforms. Code Review uses a multi-agent AI architecture, with several autonomous agents reviewing code portions in parallel. A final agent aggregates and prioritizes the findings, removes duplicates, and presents developers with the most pressing issues to address.

A light security analysis is built into Code Review, while more comprehensive audits are available via the separate Claude Code Security product. Pricing follows a token-based model, with each review estimated to cost between $15 and $25, depending on code complexity.

Anthropic reports strong adoption within its enterprise business. Subscriptions to the company’s AI products have quadrupled since the start of the year, and Claude Code’s projected annual revenue has surpassed $2.5 billion. Companies such as Uber, Salesforce, and Accenture are already leveraging Claude Code in development workflows; now, team leaders can enable Code Review as the default for all their developers.

The launch comes as Anthropic filed two lawsuits against the U.S. Department of Defense regarding a supply chain risk designation—a dispute that could have implications for the company’s enterprise ambitions.

By targeting large enterprises, Anthropic hopes to address industry demands for faster, safer deployment of new software features and reduce the friction in modern software development cycles. As AI-generated code becomes more prevalent, automated code review tools are seen as essential in maintaining code quality and efficiency in large organizations.

Source: dataconomy.com