Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.
84%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Project context — Centaur Spec (read first):
- Test runner: vitest. Run a workspace's tests with
pnpm turbo test --filter=<workspace>. From within an app/package dir,pnpm testworks directly. Append--run --coveragefor a one-shot coverage report.- Test file conventions: use
*.test.ts, co-located with source —vitest.config.tscollects only*.test.tsfrom the covered trees; a*.spec.tsunit test would silently never run (Playwright owns*.spec.tsintests/browser/, so the tiers never overlap).- Domain layer is pure (ADR-024):
packages/domain/andpackages/application/have NO I/O. Aggregate invariants and use-case happy paths are unit-testable with no mocks needed. The preferred pattern is real objects + readonly types, not mocks.- Adapter / integration tests: real implementations against ephemeral fixtures wherever possible (Postgres via Drizzle test schema, R2 via local mock or a CI-only bucket). Use mocks only at the system boundary.
- Domain language: assertions and test names use the names defined in
docs/domain-glossary.md(e.g.,Report,ReportVersion,Acl,Scope). Do not invent local aliases inside test files.- What counts as source, and what counts as its paired test:
scripts/guards.config.sh. Those two regexes are the machine-readable version of "this change needed a test". Keep them in step withvitest.config.ts's include globs: a tree the runner covers but the guards' policy file doesn't is silently un-guarded.- Decision records:
docs/adr/. Respect the ones covering the area you're touching; a test that contradicts a binding decision is a finding, not a green.- TDD enforcement layer: the
.husky/pre-pushTDD pairing guard blocks a push whose source changes carry no test changes (bypassPUSH_WITHOUT_TESTS=1, logged), and.github/workflows/tdd-pairing.ymlruns the same rule over the PR (hatch: thetdd-exemptlabel, which goes green with a notice). This skill provides the procedure; the hook and the workflow provide the enforcement.The rest of this file is Matt Pocock's upstream skill — universal advice; the project context above adapts it to our stack.
Core principle: Tests should verify behavior through public interfaces, not implementation details. Code can change entirely; tests shouldn't.
Good tests are integration-style: they exercise real code paths through public APIs. They describe what the system does, not how it does it. A good test reads like a specification - "user can checkout with valid cart" tells you exactly what capability exists. These tests survive refactors because they don't care about internal structure.
Bad tests are coupled to implementation. They mock internal collaborators, test private methods, or verify through external means (like querying a database directly instead of using the interface). The warning sign: your test breaks when you refactor, but behavior hasn't changed. If you rename an internal function and tests fail, those tests were testing implementation, not behavior.
See tests.md for examples and mocking.md for mocking guidelines.
DO NOT write all tests first, then all implementation. This is "horizontal slicing" - treating RED as "write all tests" and GREEN as "write all code."
This produces crap tests:
Correct approach: Vertical slices via tracer bullets. One test → one implementation → repeat. Each test responds to what you learned from the previous cycle. Because you just wrote the code, you know exactly what behavior matters and how to verify it.
WRONG (horizontal):
RED: test1, test2, test3, test4, test5
GREEN: impl1, impl2, impl3, impl4, impl5
RIGHT (vertical):
RED→GREEN: test1→impl1
RED→GREEN: test2→impl2
RED→GREEN: test3→impl3
...When exploring the codebase, use the project's domain glossary so that test names and interface vocabulary match the project's language, and respect ADRs in the area you're touching.
Before writing any code:
Ask: "What should the public interface look like? Which behaviors are most important to test?"
You can't test everything. Confirm with the user exactly which behaviors matter most. Focus testing effort on critical paths and complex logic, not every possible edge case.
Write ONE test that confirms ONE thing about the system:
RED: Write test for first behavior → test fails
GREEN: Write minimal code to pass → test passesThis is your tracer bullet - proves the path works end-to-end.
For each remaining behavior:
RED: Write next test → fails
GREEN: Minimal code to pass → passesRules:
After all tests pass, look for refactor candidates:
Never refactor while RED. Get to GREEN first.
[ ] Test describes behavior, not implementation
[ ] Test uses public interface only
[ ] Test would survive internal refactor
[ ] Code is minimal for this test
[ ] No speculative features addedAdapted from engineering/tdd in mattpocock/skills — MIT, see .agents/skills/LICENSE-mattpocock-skills.md. The project-context prelude above is this repo's; the four sidecars are verbatim; deep-modules.md's two diagrams are redrawn as fenced diagram-language blocks, since craft rule §10 forbids character art.
agranado2k/ai-report-platform
caeb0da
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.