What problem does it solve? Shallow code reviews often miss real engineering risks like hidden regressions, weak test assertions, architectural boundary violations, and security gaps. This Skill performs evidence-driven, multi-lens reviews of complex pull requests, legacy modules, and test automation suites, producing structured findings with severity, evidence, risk, and actionable recommendations. ## Core Features & Use Cases - Multi-Lens Review Workflow: A five-phase process covering framing, context analysis, inspection across seven dimensions (correctness, maintainability, architecture, security, performance, testing, operability), structured findings, and a final recommendation. - Severity-Ranked Findings: Every finding includes severity (Blocker/High/Medium/Low/Note), evidence, risk, and the smallest safe recommended fix, ending with an approve/comment/request-changes verdict. - Test Automation Lens: A dedicated checklist for reviewing test suites, covering flakiness, assertion quality, test isolation, page object design, selectors, and CI diagnostics. - Use Case: A reviewer receives a multi-file PR adding login tests with hard waits and weak negative-path assertions. The Skill produces a report flagging the flakiness risk as High severity, citing the exact evidence, and recommending deterministic wait conditions before merge. ## Quick Start Ask the AI to critically review this pull request for correctness, security, and test quality risks using the advanced code review skill.
