What problem does it solve? Code reviews often miss subtle defects because a single reviewer cannot cover correctness, performance, API usability, test coverage, and documentation consistency at once, and unverified findings waste time on false positives. ## Core Features & Use Cases - Parallel Multi-Lens Review: Runs independent reviewer agents concurrently across five lenses: correctness, performance, API usability, test coverage, and spec/doc synchronization. - Adversarial Claim Verification: Every shortlisted finding is challenged by three independent skeptical verifiers; claims refuted by at least two votes are removed before reporting. - Iterative Fix Loop: Validated findings are demonstrated with failing tests, fixed minimally, and the loop restarts until a round produces no surviving findings, followed by full test suite and benchmark checks. - Use Case: After finishing a bug fix in a C# service, invoke this Skill to have agents trace classification branches for false positives, check hot-path allocations, verify negative test coverage, and confirm docs match the new behavior before requesting external review. ## Quick Start Ask the AI to review the implementation changes in the current diff using the adversarial code review workflow before submitting for external review.
