testland/test-isolation-patterns
Pure reference catalog of test-isolation and fixture-lifecycle patterns - the four-phase test pattern (Meszaros), fixture scope (per-test / per-describe / shared / global), the Fresh-Fixture vs Shared-Fixture trade-off (Fowler), parallel-safety patterns, and cleanup discipline (afterEach / afterAll / tagged-cleanup), plus a pattern-selection guide and a worked leaking-state diagnosis. The database-isolation strategies (transaction-rollback / database-per-worker / template-database) and network / external-service stubbing live in references/. This is the architecture-tier reference, not a file-level fixture-coupling style rule. Use when designing fixture scope and isolation strategy, auditing fixture coupling or retry/wait policy, or moving a suite to parallel execution.
75
94%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
network-service-isolation.mdreferences/
Network and external-service isolation
Deep reference for test-isolation-patterns SKILL.md. Consult when a test touches a service the team does not control (third-party HTTP APIs, external endpoints). Tests should not depend on external services they don't control; three patterns cover the choice.
Stub (canned response)
Use when the test doesn't care about the network itself. Reach for a stub library - nock, WireMock, Mountebank - or the msw-handlers / wiremock-stubs / mountebank-imposters skills in qa-test-data.
Contract test
Use when the test cares whether the service contract holds. Pact or schemathesis verify the contract rather than a canned body.
Real network call in a controlled environment
Use for a smoke / canary test in a staging tier with a dedicated test partition, where exercising the live service is the point of the test.
Anti-patterns
| Anti-pattern | Why it fails |
|---|---|
| Unit tests calling the real external API | Tests fail when the API is down; tests pass when the API silently changes |
| Stubs that drift from production response shape | Tests pass with stubs that don't match reality |
| One global stub for the whole suite | Tests cross-couple through the stub configuration |
| Contract test with no contract refresh | Stub goes stale; tests pass while production breaks |
See also the msw-handlers, wiremock-stubs, and mountebank-imposters skills in qa-test-data for stub implementation, and the SKILL's pattern-selection guide for when each pattern applies.