def operation():
    attempts.append(1)
    if len(attempts) < 3:
        raise ValueError("fail")
    return "success"

result = retry_operation(operation)

assert result == "success"
assert len(attempts) == 3

Clear name, tests real behavior, one thing
</Good>

<Bad>
```python
def test_retry_works():
    """Vague test with mock."""
    mock = Mock()
    mock.side_effect = [ValueError(), ValueError(), "success"]

    retry_operation(mock)

    assert mock.call_count == 3

Vague name, tests mock not code

Requirements:

• One behavior
• Clear name
• Real code (no mocks unless unavoidable)

Verify RED - Watch It Fail

MANDATORY. Never skip.

# Run ONLY the test you just wrote (use appropriate command for test type)
cd api && just test-unit "tests/unit/test_retry.py::test_retries_failed_operations_3_times -v"
# Or: just test-integration "tests/integration/..."
# Or: just test-e2e "tests/e2e_mocked/..."

Confirm:

• Test fails (not errors)
• Failure message is expected
• Fails because feature missing (not typos)

Test passes? You're testing existing behavior. Fix test.

Test errors? Fix error, re-run until it fails correctly.

GREEN - Minimal Code

Write simplest code to pass the test.

Don't add features, refactor other code, or "improve" beyond the test.

Verify GREEN - Watch It Pass

MANDATORY.

# 1. Run the specific test you wrote (use appropriate command for test type)
cd api && just test-unit "tests/unit/test_retry.py::test_retries_failed_operations_3_times -v"
# Or: just test-integration "tests/integration/..."
# Or: just test-e2e "tests/e2e_mocked/..."

# 2. Run lint + type checking (catches syntax/type issues immediately)
cd api && just lint-and-fix

Confirm:

• Test passes
• Lint and mypy pass
• Output pristine (no errors, warnings)

Test fails? Fix code, not test.

Lint/mypy fails? Fix immediately before continuing.

NOTE: Do NOT run the full test suite here. The test-runner skill handles comprehensive testing after TDD is complete.

REFACTOR - Clean Up

After green only:

• Remove duplication
• Improve names
• Extract helpers

Keep tests green. Don't add behavior.

Repeat

Next failing test for next feature.

Good Tests

Why Order Matters

"I'll write tests after to verify it works"

Tests written after code pass immediately. Passing immediately proves nothing:

• Might test wrong thing
• Might test implementation, not behavior
• Might miss edge cases you forgot
• You never saw it catch the bug

Test-first forces you to see the test fail, proving it actually tests something.

"I already manually tested all the edge cases"

Manual testing is ad-hoc. You think you tested everything but:

• No record of what you tested
• Can't re-run when code changes
• Easy to forget cases under pressure
• "It worked when I tried it" ≠ comprehensive

Automated tests are systematic. They run the same way every time.

"Deleting X hours of work is wasteful"

Sunk cost fallacy. The time is already gone. Your choice now:

• Delete and rewrite with TDD (X more hours, high confidence)
• Keep it and add tests after (30 min, low confidence, likely bugs)

The "waste" is keeping code you can't trust. Working code without real tests is technical debt.

"TDD is dogmatic, being pragmatic means adapting"

TDD IS pragmatic:

• Finds bugs before commit (faster than debugging after)
• Prevents regressions (tests catch breaks immediately)
• Documents behavior (tests show how to use code)
• Enables refactoring (change freely, tests catch breaks)

"Pragmatic" shortcuts = debugging in production = slower.

"Tests after achieve the same goals - it's spirit not ritual"

No. Tests-after answer "What does this do?" Tests-first answer "What should this do?"

Tests-after are biased by your implementation. You test what you built, not what's required. You verify remembered edge cases, not discovered ones.

Tests-first force edge case discovery before implementing. Tests-after verify you remembered everything (you didn't).

30 minutes of tests after ≠ TDD. You get coverage, lose proof tests work.

Common Rationalizations

Red Flags - STOP and Start Over

• Code before test
• Test after implementation
• Test passes immediately
• Can't explain why test failed
• Tests added "later"
• Rationalizing "just this once"
• "I already manually tested it"
• "Tests after achieve the same purpose"
• "It's about spirit not ritual"
• "Keep as reference" or "adapt existing code"
• "Already spent X hours, deleting is wasteful"
• "TDD is dogmatic, I'm being pragmatic"
• "This is different because..."

All of these mean: Delete code. Start over with TDD.

Example: Bug Fix

Bug: Empty email accepted

RED

def test_rejects_empty_email():
    """Form submission should reject empty email."""
    result = submit_form({"email": ""})

    assert result["error"] == "Email required"

Verify RED

$ cd api && just test-unit "tests/unit/test_form.py::test_rejects_empty_email -v"
FAILED: AssertionError: assert None == 'Email required'

GREEN

def submit_form(data: dict) -> dict:
    """Submit form data with validation."""
    email = data.get("email", "").strip()
    if not email:
        return {"error": "Email required"}
    # ...

Verify GREEN

$ cd api && just test-unit "tests/unit/test_form.py::test_rejects_empty_email -v"
PASSED

$ cd api && just lint-and-fix
✅ All linting checks passed!

REFACTOR Extract validation for multiple fields if needed.

Verification Checklist

Before marking TDD work complete:

• Every new function/method has a test
• Watched each test fail before implementing
• Each test failed for expected reason (feature missing, not typo)
• Wrote minimal code to pass each test
• Specific tests pass (just test-unit/integration/e2e "path/to/test.py::test_name")
• Lint and mypy pass (just lint-and-fix)
• Tests use real code (mocks only if unavoidable)
• Edge cases and errors covered

Can't check all boxes? You skipped TDD. Start over.

NOTE: This checklist is for the TDD cycle only. Use test-runner skill for comprehensive testing (full test suite, all mocked tests, etc.) before committing.

When Stuck

Debugging Integration

Bug found? Write failing test reproducing it. Follow TDD cycle. Test proves fix and prevents regression.

Never fix bugs without a test.

E2E / Frontend Tests (Playwright)

TDD applies to frontend too. The same Red-Green-Refactor cycle works with Playwright E2E tests.

When to Use E2E Tests

E2E TDD Cycle

RED - Write Failing E2E Test

// tests/e2e/specs/upload-recording.spec.ts
test('user can upload a recording', async ({ page }) => {
  const dashboard = new DashboardPage(page);
  await dashboard.goto();

  const recordingId = await dashboard.uploadRecording('test-audio.wav');

  expect(recordingId).toBeTruthy();
  await expect(page.getByText('Upload complete')).toBeVisible();
});

Verify RED

just e2e-smoke
# FAILED: Expected "Upload complete" to be visible

Confirm test fails because feature is missing, not due to typos or selector issues.

GREEN - Implement Until Test Passes

Implement the minimal code (backend + frontend) to make the test pass.

Verify GREEN

just e2e-smoke
# PASSED

REFACTOR - Clean up, then verify tests still pass.

E2E Test Commands

# Run smoke tests (fast, 1-2 files)
just e2e-smoke

# Run all E2E tests (slow, full coverage)
just e2e

# Run specific test file
cd tests/e2e && npx playwright test specs/upload.spec.ts

# Debug with UI
cd tests/e2e && npx playwright test --ui

E2E Test Structure

Tests live in tests/e2e/ with Page Object pattern:

tests/e2e/
├── pages/           # Page Objects (selectors + actions)
├── flows/           # Reusable multi-page flows
├── specs/           # Actual test files
└── fixtures/        # Test data

Common E2E Mistakes

E2E + Backend Integration

When implementing a full feature:

1. Start with E2E test - Defines the user-facing behavior
2. Watch it fail - Probably fails because backend doesn't exist
3. Drop to unit tests - Write pytest tests for backend logic
4. Implement backend - Until pytest passes
5. Implement frontend - Until E2E test passes
6. Run lint - cd api && just lint-and-fix

This ensures both backend logic AND user experience are tested.

NOTE: Use test-runner skill for comprehensive testing after TDD is complete.

Final Rule

Production code → test exists and failed first
Otherwise → not TDD

No exceptions without your human partner's permission.