test-generator（测试发生器）

Test Generator Skill

Auto-suggest tests when you write new code.

When I Activate

• ✅ New function created
• ✅ New component added
• ✅ User mentions testing or tests
• ✅ Test file missing for implementation
• ✅ User asks "can you test this?"

What I Generate

Quick Test Scaffolding

• Basic happy path tests
• Null/undefined edge cases
• Simple error scenarios
• Framework-appropriate syntax (Jest/Vitest/Pytest)

What I Don't Do

• Comprehensive test suites → Use @test-engineer sub-agent
• Integration tests → Use @test-engineer sub-agent
• E2E test design → Use @test-engineer sub-agent
• Test strategy planning → Use @test-engineer sub-agent

Relationship with @test-engineer Sub-Agent

Me (Skill): Quick test scaffolding @test-engineer (Sub-Agent): Comprehensive testing strategy

Workflow

1. You write a function
2. I auto-generate basic test structure
3. You want full suite → Invoke @test-engineer sub-agent
4. Sub-agent creates comprehensive tests

Examples

JavaScript Function

// You write:
function calculateDiscount(price, percentage) {
  if (price <= 0) throw new Error('Invalid price');
  return price * (percentage / 100);
}

// I auto-generate:
describe('calculateDiscount', () => {
  it('calculates discount correctly', () => {
    expect(calculateDiscount(100, 10)).toBe(10);
  });

  it('throws error for invalid price', () => {
    expect(() => calculateDiscount(0, 10)).toThrow('Invalid price');
  });

  it('handles zero percentage', () => {
    expect(calculateDiscount(100, 0)).toBe(0);
  });

  // TODO: Add more edge cases
  // Consider: negative percentages, decimal values, very large numbers
});

React Component

// You write:
function UserCard({ user, onEdit }) {
  return (
    <div className="user-card">
      <h2>{user.name}</h2>
      <button onClick={() => onEdit(user.id)}>Edit</button>
    </div>
  );
}

// I auto-generate:
import { render, screen, fireEvent } from '@testing-library/react';

describe('UserCard', () => {
  const mockUser = { id: 1, name: 'John Doe' };
  const mockOnEdit = jest.fn();

  it('renders user name', () => {
    render(<UserCard user={mockUser} onEdit={mockOnEdit} />);
    expect(screen.getByText('John Doe')).toBeInTheDocument();
  });

  it('calls onEdit with user id when button clicked', () => {
    render(<UserCard user={mockUser} onEdit={mockOnEdit} />);
    fireEvent.click(screen.getByText('Edit'));
    expect(mockOnEdit).toHaveBeenCalledWith(1);
  });

  // TODO: Add tests for edge cases
  // - Missing user data
  // - Undefined onEdit
  // - Long names (UI testing)
});

Python Function

# You write:
def fetch_user_data(user_id: int) -> dict:
    if user_id <= 0:
        raise ValueError("Invalid user ID")
    return db.query("SELECT * FROM users WHERE id = ?", [user_id])

# I auto-generate:
import pytest

def test_fetch_user_data_success():
    """Test successful user data retrieval"""
    result = fetch_user_data(1)
    assert isinstance(result, dict)
    assert 'id' in result

def test_fetch_user_data_invalid_id():
    """Test with invalid user ID"""
    with pytest.raises(ValueError, match="Invalid user ID"):
        fetch_user_data(0)

def test_fetch_user_data_negative_id():
    """Test with negative ID"""
    with pytest.raises(ValueError):
        fetch_user_data(-1)

# TODO: Add integration tests with database
# TODO: Test database connection failures

Framework Detection

I automatically detect your testing framework:

• JavaScript/TypeScript: Jest, Vitest, Mocha
• Python: pytest, unittest
• Java: JUnit
• Go: testing package

Detection based on:

• package.json dependencies
• requirements.txt
• Existing test files
• Import statements

Test Patterns

Unit Tests

// Function testing
test('adds numbers correctly', () => {
  expect(add(2, 3)).toBe(5);
});

Component Tests

// React component testing
test('button click triggers callback', () => {
  const onClick = jest.fn();
  render(<Button onClick={onClick} />);
  fireEvent.click(screen.getByRole('button'));
  expect(onClick).toHaveBeenCalled();
});

Edge Cases

// Boundary testing
test('handles empty input', () => {
  expect(processData([])).toEqual([]);
});

test('handles null input', () => {
  expect(processData(null)).toBeNull();
});

When to Use Sub-Agent

Invoke @test-engineer for:

• Complete test suites (20+ tests)
• Integration test strategy
• E2E test planning
• Test coverage goals
• Complex mocking scenarios

Example:

Me: "Generated 3 basic tests for calculateDiscount()"
You: "@test-engineer create comprehensive test suite with all edge cases"
Sub-agent: [Creates 25+ tests covering all scenarios]

Sandboxing Compatibility

Works without sandboxing: ✅ Yes Works with sandboxing: ✅ Yes

• Filesystem: Writes test files to project
• Network: None required
• Configuration: None required

Customization

Edit test templates:

cp -r ~/.claude/skills/development/test-generator \
      ~/.claude/skills/development/my-test-generator

# Edit SKILL.md to customize:
# - Test patterns
# - Framework preferences
# - Coverage expectations

Integration with Commands

/test-gen Command

/test-gen --file utils.js --framework jest --coverage 90

# Combines:
# 1. My quick scaffolding
# 2. @test-engineer comprehensive tests
# 3. Full test file generation

Tips

1. Let me scaffold first - Review before invoking sub-agent
2. Add TODOs - I include TODO comments for complex cases
3. Framework consistency - I match your project's testing style
4. Quick iteration - Regenerate if not satisfied

Related Tools

• @test-engineer: Comprehensive test suite creation
• code-reviewer skill: Flags code that needs testing
• /test-gen command: Full test generation workflow