name: run-tests description: Run unit and integration tests user_invocable: true

Run Tests

Execute tests for the Feirb solution. There are three test layers:

Test Layers

Running Tests

1. Unit + Integration tests (fast, no infrastructure needed)

# All .NET tests
dotnet test Feirb.sln

# Single project
dotnet test tests/Feirb.Api.Tests

# Filter by class or method name
dotnet test --filter "ClassificationServiceTests"
dotnet test --filter "Execute_AlreadyRunning_SkipsExecutionAsync"

2. Full stack tests (Bruno + Playwright in Docker)

Requires Docker or Podman. Builds API image, starts Postgres + GreenMail, runs Bruno API tests and Playwright E2E tests sequentially, then tears everything down.

tests/run-tests.sh

This spins up via tests/docker-compose.test.yml:

• postgres (seeded with FEIRB_SEED_DATA)
• greenmail (SMTP/IMAP mock with preloaded emails)
• api (published .NET app on port 7273)
• bruno (runs bru run --env docker)
• playwright (runs npx playwright test in Chromium)

3. Bruno tests only (against running local API)

If the API is already running via Aspire / dev-harness:

cd tests/bruno
npx @usebruno/cli run --env local

The local environment points to https://localhost:7272.

4. Playwright tests only (against running local API)

cd tests/playwright
npx playwright test

Set BASE_URL environment variable if the API is not on the default port.

5. Interactive browser verification (against running local API)

Use /test-ui <path> <what to verify> for manual browser verification via Playwright MCP — visual checks, form interaction, and DOM assertions. Useful for debugging test failures or verifying features not covered by automated tests.

Reporting Results

• Report number of tests passed, failed, skipped per layer
• If tests fail: show failing test names, error messages, and identify relevant source code
• For Bruno/Playwright failures in Docker: check tests/run-tests.sh exit codes and container logs

Bruno Test Collections

Playwright Specs