TDD Workflow

Principle: Red -> Green -> Refactor (STRICT order!)

PHASE 0: Test Planning (FIRST STEP)

Fill test cases table BEFORE any code:

## TEST CASES (fill BEFORE implementation!)

| # | Scenario | Input | Expected Output | Type | Status |
|---|----------|-------|-----------------|------|--------|
| 1 | Happy path | <input> | <expected> | Unit | pending |
| 2 | Edge case: empty | [] | <expected> | Unit | pending |
| 3 | Edge case: null | null | Error | Unit | pending |
| 4 | Integration | <context> | <expected> | Integration | pending |

Coverage categories:

• Happy path (main scenario)
• Edge cases (empty data, boundaries, null/undefined)
• Error cases (invalid input, exceptions)
• Integration points (if dependencies exist)

STOP: Do NOT proceed to Phase 1 until table is filled!

PHASE 1: RED -- Write Failing Tests

1. Create test file (feature.test.ts)
2. Write tests for ALL cases from Phase 0 table
3. Run tests -- they MUST FAIL:

   npm test -- --grep "feature"
   # Expected: FAIL
   

STOP: Do NOT write implementation until tests fail!

PHASE 2: GREEN -- Minimal Code

1. Write MINIMAL code to make tests pass
2. Do NOT optimize, do NOT add "for later"
3. Run tests:

   npm test -- --grep "feature"
   # Expected: PASS
   

4. Repeat for each test case

STOP: Do NOT refactor until ALL tests are green!

PHASE 3: REFACTOR

1. All tests green? -> Refactor
2. Remove duplication, improve readability
3. After EACH change: npm test (must stay green)

PHASE 4: VERIFY

## TDD VERIFICATION

| Category | Written | Passing | Skipped |
|----------|---------|---------|---------|
| Happy path | X | X | 0 |
| Edge cases | X | X | 0 |
| Error cases | X | X | 0 |
| Integration | X | X | 0 |
| **TOTAL** | X | X | 0 |

FORBIDDEN

• Writing code BEFORE test cases table
• Writing code BEFORE failing tests
• Refactoring with red tests
• Skipping checkpoints
• "I'll write tests later"