What problem does it solve? Verifying that generated or existing code actually works requires a complete, executable test strategy — most projects end up with scattered happy-path tests that miss integration failures, contract drift, and performance regressions. ## Core Features & Use Cases - Full test pyramid generation: Produces unit, integration (testcontainers/docker-compose), contract (Pact/OpenAPI schema), E2E (Playwright/Cypress), and performance (k6) tests under a dedicated tests/ tree, traced to BRD acceptance criteria via a traceability matrix. - Oracle ownership and TDD pairing: Writes failing acceptance scaffolds first, then guards the tests/ tree against weakening by other agents, flagging skipped or loosened assertions as critical findings. - Execution with real numbers: Runs every suite it writes and reports tests_run/tests_passing/tests_failing, plus CI workflow configuration with coverage thresholds and flaky-test handling. - Use Case: After backend and frontend code is generated, invoke this skill to produce a test plan, all test suites, CI test workflow, and an execution report proving which acceptance criteria pass. ## Quick Start Ask the agent to write and run a complete test suite covering unit, integration, and e2e flows for the services in this project.
