code-reviewer

Automate pre-merge PR reviews by orchestrating seven specialized agents that inspect changed files in parallel for architecture, security, performance, and logic. This coordinates deep code-quality checks to reduce manual review time and improve merge safety.

Multi-agent code review works by deploying seven specialized agents to assess quality, architecture, security, and performance in parallel. Each agent focuses on a specific domain, consolidating results into a prioritized, actionable findings report.

Run a comprehensive codebase health check by invoking the skill with the target changed path. Seven agents execute parallel reviews to audit implemented features, producing a structured report prioritized by Critical, Warning, and Info.

You can get a prioritized findings report from a parallel code review that includes a self-challenge step and an integration workflow. The structured report categorizes issues by Critical, Warning, and Info levels for actionable merge safety.

The best way to perform quality assurance checks on pull requests is automating deep reviews with seven specialized agents. This parallel approach assesses readability, test coverage, and simplicity, reducing manual effort and ensuring architectural adherence.

Automated code review assesses performance and logic vulnerabilities using dedicated agents. The Performance and Logic agents run in parallel alongside others to inspect changed files, identifying bottlenecks and flaws before generating a consolidated report.