tea-automate

Use when implementing test automation for features. The most commonly used TEA workflow.

Trigger with: "automate tests", "test automation", "write tests", "implement tests"

Instructions

Agent: TEA (Murat) Input: Test design document, story files, existing codebase Output: Automated test files

Step 1: Load Context

Read the relevant test design document. Read the story file if automating for a specific story. Detect the test framework from _bmad/config.yaml or project files (package.json, etc.).

Step 2: Select Tests to Automate

From the test design, select tests by priority:

1. P0 (Critical) — automate first, no exceptions
2. P1 (High) — automate in current sprint
3. P2 (Medium) — automate if time permits
4. P3 (Low) — defer or manual test

Step 3: Implement Tests

For each test scenario:

1. Set up test structure — describe block, test fixtures, data factories
2. Implement Given — preconditions, test data, mocks/stubs
3. Implement When — actions, API calls, user interactions
4. Implement Then — assertions, expected outcomes, cleanup
5. Run and verify — test passes with correct implementation, fails without it

Follow test architecture patterns:

• Page Object Model for e2e tests
• Data factories for test data
• Fixtures for reusable setup
• Proper test isolation (no shared state between tests)

Step 4: Verify Coverage

Check that automated tests cover the risk areas identified in the test design. Report any coverage gaps.

Step 5: Save and Report

Save test files alongside the implementation code. Report: tests created, coverage achieved, any gaps remaining.