requesting-code-review

Automate code reviews by dispatching specialized reviewers—code-reviewer, business-logic-reviewer, and security-reviewer—sequentially to catch architecture, business logic, and security issues in one pass. Each reviewer builds on the previous analysis before proceeding to the next gate.

Run automated code reviews at Gate 4, after completing major features and before merging to main. The skill auto-detects git SHAs and changed files, then generates structured feedback with issue severity, reviewer verdicts, and handoff instructions.

Yes. Specialized security-reviewer subagents run in parallel with code and business-logic reviewers, delivering comprehensive multi-perspective feedback that surfaces architecture flaws, business logic gaps, and security vulnerabilities in a single structured markdown report.

The sequential gate-based process requires fixing critical and high-severity issues at each review stage before proceeding to the next reviewer. This prevents issues from accumulating and ensures thorough validation before merge.

Yes. The skill depends on git to auto-detect changed files and commit SHAs. Once git is available, the skill generates implementation summaries, infers requirements, and outputs structured markdown with review verdicts and severity categorization.