testinglisted

# Purpose Write tests, run test suites, and debug test failures across the project's repositories (`.claude/project.json` → `repos`). Typical layout: a device/edge repo `<device>` (Python/pytest), the main app `apps/<mainApp>` (TypeScript/Jest + RTL, Playwright E2E), and an infrastructure repo (deployment config via template render + dry-run); placeholders come from `project.json → mainApp` / `device`. # Rules (never violate) 1. Verify the source module exists (Glob/Read) before writing a test — never test a module you cannot locate. 2. Always run a test after writing it; never return a test that was not executed and passing. 3. Test behavior, not implementation; no vacuous assertions (`expect(true).toBe(true)`). 4. A single test file stays under 200 lines — split by feature area if larger. 5. Check prerequisites before integration/E2E runs: `make test-all` needs a database, `npm run test:e2e` needs a running app. 6. Never delete a flaky test or run `npm install`/`pip install` without user approval. # Resources - Read `resources/writing-test-patterns.md` when writing tests: philosophy (AAA, TDD), the testing pyramid, the framework/location/naming table, and code patterns for pytest, Jest, RTL, Playwright, and deployment config dry-runs. - Read `resources/running-and-debugging.md` when running suites or debugging failures: per-repo commands, the 4-step debug procedure, self-check, common mistakes, escalation, and failure modes. # How to use ## What it does A hands-on t