Unit, bloc, widget, and integration testing conventions — bloc_test, mocktail, http_mock_adapter, test naming, and what to test. Use when writing or fixing any test, mocking a dependency, or deciding what tests a change requires.
Use when writing tests, deciding what to test, reviewing test coverage, or verifying that a spec's acceptance criteria are actually covered. Also when a test is failing and the fix is unclear.
Coverage-driven, discovery-first product testing that operates a real app across risk-driven sessions and multi-session campaigns, maintains complementary product-surface and risk-dimension coverage, uncovers unreported bugs, and independently verifies fixes without changing production code. Use for Quick Exploration, Delta Discovery, systematic Full Discovery Campaigns, Final Broad Testing, Fix Verification / Regression after Builder implementation, Maestro-assisted flows, hardware/device coverage planning, or proactive testing beyond existing unit/UI suites.
Run, write, and debug tests — unit, integration, and end-to-end — in whatever framework the project already uses (Jest, Vitest, PHPUnit, Pytest, Go test, cargo test, RSpec, JUnit). Invoke when the user asks to test something, run the suite, add or fix tests, check coverage, or investigate a failing test. Skip when they want a code review rather than tests, or are asking how a test framework works in the abstract.
Stratégies de test, TDD, tests unitaires, d'intégration, E2E, et couverture de code. Use when writing or fixing unit, integration or E2E tests, choosing a test strategy, setting up TDD, mocks, fixtures or coverage, or when the user says test, tester, couverture, TDD, jest, vitest ou playwright.
Write comprehensive, maintainable tests following TDD and AAA pattern. Use when writing unit tests, integration tests, setting up fixtures, mocking dependencies, or improving test coverage. Covers Python (pytest), TypeScript (Jest), Go (testing), and Rust (built-in + proptest). Do NOT use for mutation testing specifics (use mutation-testing skill).
Write unit tests, integration tests, and end-to-end tests. Use when implementing test suites, improving coverage, or ensuring code quality through testing.
Use when implementing any feature or bugfix and a test-driven approach is appropriate (red-green-refactor, seams-first), or when the user asks to find testing gaps, mutation test, or identify surviving mutants. Unit and integration tests only; browser journeys go to e2e-testing, test infrastructure to observability-quality, completion claims to gates.
Test scenarios, coverage analysis, test quality standards, and the mandatory test failure stop protocol. Load this skill for any task requiring test coverage.
Use when defining a test strategy, choosing layers, raising meaningful coverage, or fixing flaky suites — write tests with the feature, failing test first for bugs, QA audit before launch.
Write and review tests at all levels — Unit, Integration/Feature, and E2E — following Martin Fowler's testing pyramid. Use this skill when creating, modifying, or reviewing any test. Prevents test bloat, fragile assertions, and misplaced test levels. Adapts to any language/framework by following the decision flowchart and layer guidelines.
Use when writing, running, or fixing automated tests, including tests for new code, regression tests, one test or a whole suite, failing or flaky tests, snapshots and golden files, mocks, and coverage. Runs tests non-interactively, shows each new test can fail, keeps tests deterministic, and never deletes, skips, or weakens a test to get a green run.
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md or AGENTS.md?
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
What is a skill?
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Can I search my own team's files too?
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.