Claude Code now dispatches a team of agents on every PR to catch bugs that skims miss. Available in research preview for Team and Enterprise.

Claude

Anthropic is launching Code Review for Claude Code, a multi-agent system that automatically reviews pull requests on GitHub. When a PR is opened, a team of agents works in parallel to find bugs, verify findings to reduce false positives, and rank issues by severity. The system scales with PR size and complexity, averaging about 20 minutes per review. Internally at Anthropic, it increased the rate of PRs receiving substantive review comments from 16% to 54%. On large PRs, 84% receive findings averaging 7.5 issues, with less than 1% of findings marked incorrect by engineers. Reviews are billed on token usage, averaging $15–25 per PR. The feature is now in research preview for Team and Enterprise plans, with admin controls for monthly spend caps, repository-level enablement, and an analytics dashboard.