What problem does it solve? Test suites often contain tests that pass but verify nothing, flaky tests that erode trust, and duplicated or poorly named tests that are hard to maintain. This Skill reviews existing test code and pinpoints exactly which tests give false confidence or cause maintenance pain, with concrete fixes. ## Core Features & Use Cases - Severity-ranked audit: Classifies findings as Critical, High, Medium, or Low, covering missing assertions, swallowed exceptions, tautological assertions, flakiness indicators (Thread.Sleep, DateTime.Now, unseeded randomness), order-dependent shared state, over-mocking, magic values, and duplicate tests. - Multi-language support: Works with .NET (xUnit/NUnit/MSTest), Python/pytest, TypeScript/Jest, Java/JUnit, Go, Ruby/RSpec, Rust, C++, and PowerShell/Pester via language-specific extension guidance. - Honest calibration: Distinguishes real anti-patterns from idiomatic patterns (Go table-driven tests, pytest bare asserts) and leads with positive observations when tests are well written. - Use Case: Point it at a flaky CI suite and receive a report showing which tests use wall-clock sleeps or shared static state, with before/after fixes for each Critical finding. ## Quick Start Ask the AI to audit the tests in your project for anti-patterns and report the most critical issues with fixes.
