Claude Code Adds Multi-Agent Code Review to Catch Bugs in Pull Requests

Claude

Mar 15, 2026 · Updated Jun 5, 2026

Claude Code now dispatches agents to review every pull request, catching bugs human skims often miss. Multiple agents run in parallel, verify findings, and rank issues by severity. Available as a research preview for Team and Enterprise plans.

Claude Code introduced Code Review, a multi-agent system that automatically examines pull requests when they open. Multiple agents analyze code in parallel to find bugs, cross-check findings to filter false positives, and rank issues by severity before posting results as a summary comment plus inline annotations.

This brings agentic coding from the editor into the review workflow. At Anthropic, Code Review now runs on nearly every PR, lifting substantive review coverage from 16% to 54%. It will not approve changes — that remains a human decision — but it closes the gap so teams can actually cover what is shipping.

Teams on Team and Enterprise plans can start using Code Review now. Every new PR gets a deep multi-agent review automatically, so engineering teams that struggle with review bottlenecks finally have a way to catch bugs at scale before merge.

View the full update on claude.com

Claude

@claudeaiMar 9

Introducing Code Review, a new feature for Claude Code. When a PR opens, Claude dispatches a team of agents to hunt for bugs. https://t.co/AL2J4efxPw

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from Anthropic →

Keep reading

Anthropic Launches Claude Code Ultrareview to Hunt Bugs With Cloud Agent Fleets

Anthropic introduced /ultrareview for Claude Code, a research preview command that dispatches a swarm of cloud-based agents to find bugs in critical code changes. This shifts code verification from single-model analysis to a high-intensity multi-agent search for vulnerabilities in authentication and data migrations.

ClaudeFeb 20

Claude Code Security Scans Codebases for Vulnerabilities and Suggests Patches

Claude Code Security launches as a new capability that scans codebases the way a human researcher would - reasoning about component interactions, not just matching known patterns. It finds vulnerabilities, suggests patches, and surfaces everything for human approval before applying.

WarpApr 4

Warp Enables Inline Code Review and Feedback for AI Agents

Warp now supports native code reviews and inline commenting for AI agents like Claude Code and Codex directly in the terminal. This creates a structured human-in-the-loop workflow that allows developers to provide granular feedback to agents without leaving their development environment.

OpenAIMar 16

OpenAI Codex Now Runs Parallel Subagents for Multi-Agent Workflows

OpenAI's Codex coding agent now supports subagent workflows, letting developers spawn specialized agents that run tasks in parallel. A single prompt can split a PR review across security, quality, and test agents simultaneously, keeping the main context clean.