What problem does it solve? Writing good unit and component tests is a craft most developers learn by trial and error: choosing the wrong queries, mocking at the wrong layer, sharing mutable state across tests, and asserting implementation details instead of behavior. This Skill provides concrete authoring guidance at the moment a test is being written, across four major stacks, so tests are behavior-focused, non-flaky, and maintainable from the start. ## Core Features & Use Cases - Stack-specific authoring guides: React/Testing Library (query priority, userEvent, MSW network mocking, renderHook, jest-axe), Python/pytest (fixtures, parametrize, patch-site rules, AsyncMock), Go (table-driven tests, subtests, interface mocking, benchmarks), and Vue (mount strategies, composables, Pinia, nextTick handling). - Baseline testing standards: the 80% coverage floor and its qualifications, RED-GREEN-REFACTOR loop, Arrange-Act-Assert structure, behavior-describing naming, and a diagnose-in-order protocol for failing tests. - Regression test patterns: four recurring bug patterns (dual-path shape drift, projection omission, error-state leakage, missing rollback) and how to write the test a found bug earned. - Use Case: During a TDD workflow, you ask how to test a custom React hook that fetches data — the Skill walks you through renderHook with a fresh QueryClient per test, MSW handlers, and waitFor assertions, avoiding the shared-instance flake trap. ## Quick Start Ask the assistant to help you write a unit test for a specific component, hook, function, or store in React, Python, Go, or Vue, describing the behavior you want to verify.
