What problem does it solve? Code reviews that only skim a diff miss logic errors, dead code, and design problems. This Skill enforces a systematic six-phase review process that always runs the project's linters, formatters, and test suite before analyzing logic, design, complexity, naming, performance, and documentation ā while never modifying the code under review. ## Core Features & Use Cases - Mandatory automated verification: Detects project tooling from config files and CI, runs lint, type checks, format checks, and the full test suite, and reports every issue without fixing anything. - Structured design and logic analysis: Checks DRY, SOLID, YAGNI, complexity, naming, dead code, performance patterns (N+1, O(nĀ²)), and documentation using dedicated reference guides. - Labeled advisory report: Delivers findings with Conventional Comments labels (issue, suggestion, nitpick, question, praise) and an advisory verdict of APPROVE, COMMENT, or REQUEST CHANGES. - Use Case: Before merging a pull request that touches payment logic, ask for a review. The Skill runs the test suite, flags a sensitive payment path, recommends a separate security-review agent, and hands test-quality assessment off to the test-driven-development skill. ## Quick Start Review the changes in this pull request: run the linter and tests, check the logic and design, and give me a labeled findings report with a verdict.
