plan-executor:run-reviewer-team

Automated multi-agent code review orchestration executes Claude, Codex, Gemini, and security reviewers in parallel to evaluate code updates simultaneously. This coordinates concurrent assessments within a specified plan context to ensure comprehensive quality assurance.

To triage and deduplicate issues found by multiple code reviewers, you can use structured issue triaging to collect, deduplicate, and categorize findings. This summarizes critical issues, such as fixes required and rejections, for quick resolution.

You can evaluate code changes using prior review history and recent updates by coordinating multiple AI reviewers to assess file modifications within a specified plan context. This ensures thorough assessment by considering previous review iterations alongside recent changes.

For automated code review orchestration, you need to provide the plan context, execution outputs, and file modifications. Supplying these inputs allows the multi-agent system to evaluate your current code changes and generate a comprehensive review report.

Automated review triage works with specific AI models like Claude, Codex, and Gemini, alongside dedicated security reviewers. It launches these reviewers in parallel to evaluate code changes and then collects their findings for structured categorization.