behavior-focused-tests

name: behavior-focused-tests description: Review tests to ensure they validate behavior and outcomes rather than implementation details, making them more maintainable and resilient to refactoring

Behavior-Focused Test Review

Role: worker. This skill reviews test files and produces actionable recommendations. It does not modify files.

Review tests to ensure they follow behavior-driven principles and remain implementation agnostic.

Constraints

1. Read-only. Do not modify test files. Report findings only.
2. Be concise. A few high-impact findings beat an exhaustive list.
3. Follow the output format. Use the structured format below for every review.

Core Principles

Test WHAT, not HOW:

• Tests should verify outcomes and behavior
• Tests should not depend on internal implementation details
• Tests should survive refactoring when behavior remains unchanged

Review Checklist

When reviewing test code, check for these anti-patterns:

1. Testing Private/Internal Implementation

❌ Bad: Testing private methods, internal state, or implementation details ✅ Good: Testing public API and observable behavior

2. Over-Mocking

❌ Bad: Mocking everything, verifying mock call counts/arguments ✅ Good: Mock external dependencies only, assert on outcomes

3. Brittle Selectors/Identifiers

❌ Bad: CSS class names, DOM structure, internal IDs ✅ Good: Semantic attributes (data-testid, aria-labels), user-facing text

4. Testing Framework Internals

❌ Bad: Testing component lifecycle, framework-specific details ✅ Good: Testing user-visible behavior and state changes

5. Coupling to Code Structure

❌ Bad: Tests that break when code is reorganized but behavior unchanged ✅ Good: Tests that describe requirements and expected behavior

Review Process

1. Identify the Test Subject: What is being tested?

2. Analyze Test Dependencies:

   • What does the test import/require?
   • What is being mocked?
   • Are mocks necessary or could real implementations work?

3. Check Assertions:

   • Do assertions verify outcomes or implementation?
   • Are we asserting "that X happened" or "how X happened"?
   • Could this assertion pass with different implementation?

4. Evaluate Brittleness:

   • Would refactoring break this test?
   • Does the test know too much about internals?
   • Is it testing through the public interface?

5. Provide Specific Feedback:

   • Point to exact lines with issues
   • Explain why it's implementation-focused
   • Suggest behavior-focused alternatives
   • Show concrete refactored examples

Examples

Example 1: Over-Mocking

❌ Implementation-Focused:

test('getUserData calls API with correct params', () => {
  const mockFetch = jest.fn();
  getUserData(mockFetch, 123);
  expect(mockFetch).toHaveBeenCalledWith('/api/users/123');
});

✅ Behavior-Focused:

test('getUserData returns user information', async () => {
  const userData = await getUserData(123);
  expect(userData).toMatchObject({
    id: 123,
    name: expect.any(String),
    email: expect.any(String)
  });
});

Example 2: Testing Internal State

❌ Implementation-Focused:

test('Counter increments internal state', () => {
  const counter = new Counter();
  counter.increment();
  expect(counter._value).toBe(1); // Testing private field
});

✅ Behavior-Focused:

test('Counter displays incremented value', () => {
  const counter = new Counter();
  counter.increment();
  expect(counter.getValue()).toBe(1); // Testing public API
});

Example 3: UI Testing

❌ Implementation-Focused:

test('button has correct class', () => {
  render(<LoginButton />);
  expect(screen.getByRole('button')).toHaveClass('btn-primary');
});

✅ Behavior-Focused:

test('clicking login button navigates to dashboard', async () => {
  const user = userEvent.setup();
  render(<LoginButton />);
  await user.click(screen.getByRole('button', { name: /login/i }));
  expect(screen.getByText(/dashboard/i)).toBeInTheDocument();
});

Red Flags

Look for these patterns that often indicate implementation coupling:

• Accessing private fields/methods (e.g., _internal, #private)
• Mocking with call count verification (toHaveBeenCalledTimes)
• Testing in terms of "calls X then Y then Z"
• CSS class assertions unrelated to behavior
• Testing framework lifecycle methods
• Relying on code structure (file paths, module structure)
• Snapshot tests of implementation details

Output Format

For each test file reviewed, provide:

1. Summary: Overall assessment of behavior vs implementation focus
2. Issues Found: Specific lines with problems
3. Recommendations: Concrete suggestions with code examples
4. Refactored Example: Show at least one fully refactored test

Arguments

Accept file paths as arguments. If no arguments are given, ask which files to review.

Instructions

When this skill is invoked:

1. Parse arguments. Use provided file paths or ask user which test file(s) to review.
2. Read the test file(s)
3. Analyze each test case against the checklist
4. Identify implementation-focused patterns
5. Provide detailed, actionable feedback
6. Offer to refactor specific tests
7. Explain the "why" behind each recommendation

Remember: The goal is tests that document requirements and survive refactoring.