review-commit

Automate code review by running an agent that analyzes working directory changes for critical, important, and minor issues, uses a dedicated fixer agent to resolve problems, and enforces an iterative review loop before creating a pull request.

The best way to catch security vulnerabilities and bugs is to run an automated code analysis on your working directory changes, which categorizes issues by severity and employs a fixer agent to resolve critical and important problems.

The iterative fixing process enforces a maximum of five review loops using dedicated agents to review code changes and address critical and important issues, generating an approval flag only upon successful resolution of those issues.

Yes, you can run an automated code review directly on your current working directory changes to analyze quality, security, and best practices, preparing your code for a pull request without requiring external dependencies.

Automated code review categorizes identified issues into critical, important, and minor severity levels, using dedicated agents to iteratively fix the critical and important problems before generating a final approval flag.

The iterative review loop runs a maximum of five times to fix critical and important issues using a dedicated fixer agent; if these issues remain unresolved, the automated code review will not generate the final approval flag.