agentsclimarketplace

Skills

Skill majiayu000/spellbook/skills/test-driven-development

Cross-runtime skills for Claude Code, Codex, and multi-agent workflows.

Install
npx -y skills add majiayu000/spellbook --skill test-driven-development

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Enforces TDD discipline with RED-GREEN-REFACTOR cycle. Use when writing new features, fixing bugs, or refactoring code. Ensures tests genuinely verify behavior.

SKILL.md

2.9 KB, as published. Nobody here has run it

Test-Driven Development (TDD)

From obra/superpowers

Core Principle

Write tests before implementation. Watch them fail. Write minimal code to pass.

THE IRON LAW: NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST

When to Apply TDD

  • New features
  • Bug fixes
  • Refactoring
  • Any behavior changes

The Red-Green-Refactor Cycle

RED: Write a Failing Test

  1. Write ONE minimal test demonstrating desired behavior
  2. Use clear, descriptive test names
  3. Use real code, avoid unnecessary mocks
  4. Test should fail for the RIGHT reason
# Verify RED
- Run the test
- Confirm it fails
- Confirm failure is NOT due to syntax errors
- Confirm you're not testing existing functionality

GREEN: Make It Pass

  1. Write the SIMPLEST code that makes the test pass
  2. Don't add extra features
  3. Don't optimize yet
  4. Just make it work
# Verify GREEN
- Run all tests
- Confirm new test passes
- Confirm no other tests broke

REFACTOR: Clean Up

  1. Improve code quality while keeping tests green
  2. Remove duplication
  3. Improve naming
  4. Extract helpers if needed
  5. Run tests after each change

Why This Order Matters

Tests written AFTER implementation:

  • Pass immediately (proves nothing)
  • You never see them fail
  • Cannot verify they test what matters
  • Manual testing is NOT a substitute

Common Rationalizations to REJECT

ExcuseReality
"Too simple to test"Simple code still breaks
"I'll write tests after"Post-implementation tests pass immediately
"Already manually tested"Manual testing lacks systematic rigor
"Deleting work is wasteful"Unverified code is technical debt
"Keep as reference"You'll adapt it; that's testing-after

Red Flags Requiring RESTART

If any of these occur, DELETE the code and start over:

  • Writing code before tests
  • Tests passing immediately on first run
  • Rationalizing "just this once"
  • Planning to "test later"
  • Keeping pre-written code "as reference"

Example Workflow

# 1. RED - Write failing test
def test_user_can_login_with_valid_credentials():
    user = create_user(email="[email protected]", password="secret")
    result = login(email="[email protected]", password="secret")
    assert result.success == True
    assert result.user == user

# 2. Run test - MUST FAIL
# 3. GREEN - Implement minimal code
def login(email, password):
    user = User.find_by_email(email)
    if user and user.check_password(password):
        return LoginResult(success=True, user=user)
    return LoginResult(success=False, user=None)

# 4. Run test - MUST PASS
# 5. REFACTOR - Clean up if needed

Summary

  1. Write test first
  2. Watch it fail
  3. Write minimal code to pass
  4. Refactor while green
  5. Repeat

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.