agent_test_execution

Agent Test Execution Rules

Reference: Testing standards documentation (repository-specific)

Purpose

Ensures agents automatically run tests after implementing code changes and fix any failures before proceeding. This prevents bugs from being discovered only during manual testing.

Scope

This document covers:

• Mandatory test execution after code changes
• Test execution commands and workflows
• Failure handling and remediation
• Test result verification

This document does NOT cover:

• Test quality standards (see repository test quality rules)
• Test writing patterns (see repository testing standards)
• Test-first workflow (see test_first_workflow.mdc)

Trigger Patterns

Agents MUST run tests immediately after:

• Implementing new features or functions
• Modifying existing code
• Adding or updating tests
• Changing database schemas or migrations
• Updating API endpoints or contracts
• Modifying UI components
• Before marking any implementation as complete

Mandatory Test Execution

Step 1: Run Relevant Test Suite

After ANY code change, agents MUST run the appropriate test suite:

For backend/domain changes:

# Type check first
npm run type-check

# Run unit tests
npm test

# Run integration tests
npm run test:integration

For frontend/UI changes:

# Type check
npm run type-check

# Run unit tests
npm test

# Run E2E tests (Playwright)
npm run test:e2e

For database/schema changes:

# Run migrations
npm run migrate

# Run integration tests (includes DB operations)
npm run test:integration

# Check schema advisors (if available)
npm run check:advisors

For all changes:

# Lint check
npm run lint

Step 2: Verify Test Results

After running tests, agents MUST:

1. Check exit codes: All test commands must exit with code 0
2. Review test output: Verify all tests passed (not just "some tests passed")
3. Check coverage: Verify coverage meets requirements (if coverage check is enabled)
4. Review failures: If any tests fail, identify root cause