Anthropic Introduces Code Review Feature to Claude Code
Key Points
- Anthropic adds Code Review to Claude Code, an AI‑powered coding assistant.
- Feature automatically scans pull requests, flags bugs, and provides a high‑signal overview comment plus inline notes.
- Multi‑agent system scales dynamically, handling larger changes with more agents.
- Typical review completes in about 20 minutes.
- Pricing is based on token usage, averaging between $15 and $25 per review.
- Admins can set monthly caps, repository restrictions, and monitor spend via an analytics dashboard.
- Internal testing showed a significant increase in substantive review comments.
- Code Review is now available to Claude for Teams and Enterprise subscribers in a research preview.
Anthropic has added a new Code Review capability to its Claude Code AI coding assistant. The feature automatically analyzes pull requests, flags bugs, and supplies actionable feedback through a high‑signal overview comment and inline notes. It scales its multi‑agent review process based on the size and complexity of the change, typically completing a review in about 20 minutes. While the tool costs more than lightweight alternatives, Anthropic offers caps and dashboards to help manage expenses. Early internal testing shows a surge in substantive review comments, and the feature is now rolling out to Claude for Teams and Enterprise subscribers in a research preview.
New Code Review Feature for Claude Code
Anthropic’s AI coding assistant, Claude Code, now includes a feature called Code Review designed to help developers locate and fix bugs more quickly. When a pull request (PR) is opened, the system dispatches a team of agents that search for bugs in parallel, verify findings to filter out false positives, and rank issues by severity. The output appears as a single high‑signal overview comment on the PR, accompanied by inline comments for specific problems.
Dynamic Multi‑Agent Scaling
The review system adjusts its resources based on the PR’s size and complexity. Larger or more intricate changes receive additional agents and deeper analysis, while smaller updates get a lighter review. Anthropic reports that the average PR review completes in about 20 minutes.
Cost Structure and Management Tools
Code Review is priced on token usage and is more expensive than lightweight options such as the Claude Code GitHub Action. Average costs fall between $15 and $25 per review, scaling with PR size and complexity. To help administrators control spending, Anthropic provides monthly organization caps, repository‑level restrictions, and an analytics dashboard that tracks reviewed PRs, acceptance rates, and total spend.
Internal Success and Market Rollout
The feature was initially built to streamline Anthropic’s internal operations after the company observed a 200% increase in code generated per engineer over the past year. Internal testing showed a significant rise in substantive review comments, and the system is now used on nearly every PR within Anthropic. Following this success, Code Review is being released to Claude for Teams and Enterprise subscribers in a research preview.
Business Impact
Claude Code’s revenue has reportedly surpassed $2.5 billion since launch, more than doubling since early 2026, with enterprise customers now representing over half of total revenue. The addition of Code Review is expected to further boost adoption among business users seeking robust, AI‑driven code quality tools.