When it comes to coding, peer feedback is critical to catching bugs early, maintaining consistency across the codebase, and improving overall software quality.
The rise of “vibe coding,” which uses AI tools to quickly generate large amounts of code by taking instructions given in plain language, has changed the way developers work. Although these tools have accelerated development, they have also introduced new bugs, security risks, and poorly understood code.
Anthropic’s solution is an AI reviewer designed to catch bugs before they are added to the software’s codebase. A new product called Code Review launched Monday at Claude Code.
“We have seen Claude Code grow significantly, especially within the enterprise. One of the questions we continue to receive from business leaders is: Now that Claude Code is submitting numerous pull requests, how do we ensure that these requests are reviewed in an efficient manner?” Cat Wu, head of product at Anthropic, told TechCrunch.
A pull request is a mechanism that developers use to submit code changes for review before those changes are applied to the software. Wu said Claude Code has dramatically increased code output, which has led to an increase in pull request reviews, which creates a bottleneck in shipping code.
“Code Review is our answer to that,” Wu said.
Anthropic’s launch of Code Review (arriving first to Claude for Teams and Claude for Enterprise customers in research preview) is a pivotal moment for the company.
Tech Crunch Event
San Francisco, California
|
October 13-15, 2026
On Monday, Anthropic filed two lawsuits against the Department of Defense in response to the agency’s designation of Anthropic as a supply chain risk. The dispute will force Anthropic to focus more on its booming enterprise business, which has seen its subscriber base quadruple since the beginning of the year. Claude Code’s running revenue has exceeded $2.5 billion since launch, according to the company.
“This product is aimed at large enterprise users, so companies like Uber, Salesforce, and Accenture are already using Claude Code and now they want help with the huge volume of pull requests that this product helps generate,” Wu said.
She added that developer leads can enable code reviews to run by default for all engineers on their team. When activated, it integrates with GitHub to automatically analyze pull requests and leave comments describing potential issues and suggested fixes directly in the code.
Wu said he focuses on correcting logical errors rather than style.
“This is very important because many developers have seen AI automation feedback before and get annoyed if it isn’t immediately actionable,” Wu said. “We decided to focus purely on logic errors. This would allow us to capture the top priorities for fixing.”
AI explains its reasoning step by step, outlining what it thinks the problem is, why it might be a problem, and how it could potentially be solved. The system uses color to indicate the severity of the problem. The highest severity is in red, potential issues worth reviewing are in yellow, and issues related to existing code or historical bugs are in purple.
It does this quickly and efficiently by relying on multiple agents working in parallel, each agent examining the codebase from a different perspective or dimension, Wu said. The final agent aggregates and ranks the results, removing duplicates and prioritizing the most important ones.
The tool provides simple security analysis and allows engineering leaders to customize additional checks based on internal best practices. Wu said Anthropic’s recently released Claude Code Security provides more in-depth security analysis.
The multi-agent architecture means this can be a resource-intensive product, Wu said. Like other AI services, pricing is token-based and costs vary depending on code complexity. Wu estimated that each review would cost an average of $15 to $25. She added that this is a premium experience and one that is needed as AI tools generate more and more code.
“(Code review) comes from a huge amount of market traction,” Wu said. “As engineers develop using Claude Code, they struggle to create new features (decrease) and the demand for code reviews becomes much higher. So we expect this will help companies build faster than before and with far fewer bugs than before.”
