---

name: sdlc-testing description: Comprehensive testing guidance. Use when writing tests, debugging, or validating functionality. Covers unit, integration, and system testing.

SDLC Testing Skill

When to Use This Skill

• Writing tests for new features
• Debugging issues
• Validating functionality
• Creating test plans
• Quality assurance

Testing Strategy

1. Unit Testing

Purpose: Test individual functions in isolation

Guidelines:

• Test one thing at a time
• Mock external dependencies
• Aim for 80%+ coverage
• Test happy path and edge cases

Example Test Structure: \\python def test_keyword_extraction(): # Arrange prompt = "Find mental health sponsors in NYC"

Copy# Act
keywords = extract_keywords(prompt)

# Assert
assert len(keywords['primary']) > 0
assert 'mental health' in keywords['primary']
assert 'NYC' in keywords['location']

\\

2. Integration Testing

Purpose: Test how components work together

What to Test:

• API endpoints (if applicable)
• Database operations
• External service calls
• Data flow between components

Example: \\python def test_complete_workflow(): # Test Phase 1 -> Phase 2 integration keywords = keyword_extractor.extract(prompt) sponsors = web_researcher.search(keywords) assert len(sponsors) > 0 \\

3. System Testing

Purpose: Test entire system end-to-end

What to Test:

• Complete workflows
• Performance (response time)
• Error handling
• Data validation

4. Edge Case Testing

Always test:

• Empty inputs: What if prompt is empty?
• Null values: What if no sponsors found?
• Large datasets: Can it handle 1000 sponsors?
• Invalid data: What if keywords are malformed?
• Network failures: What if web search fails?

Test Categories

Positive Tests (Happy Path)

• Valid inputs produce expected outputs
• Workflow completes successfully

Negative Tests (Error Cases)

• Invalid inputs are rejected gracefully
• Errors are logged properly
• System doesn't crash

Boundary Tests

• Minimum values (0 sponsors)
• Maximum values (500 sponsors)
• Edge of valid range

Test Template

\\python def test_sponsor_search(): """ Test complete sponsor search functionality

CopyGiven: A valid search prompt
When: The workflow executes
Then: Return validated sponsor list
\"\"\"
# Arrange
prompt = "Find mental health sponsors in NYC"
expected_min_sponsors = 10

# Act
results = search_sponsors(prompt)

# Assert
assert len(results) >= expected_min_sponsors
assert all('name' in sponsor for sponsor in results)
assert all('type' in sponsor for sponsor in results)
assert all('relevance_score' in sponsor for sponsor in results)

# Verify data quality
for sponsor in results:
    assert sponsor['relevance_score'] >= 5
    assert sponsor['name'] != ''
    assert sponsor['type'] in ['Corporation', 'Foundation', 'NGO', 'Individual']

\\

Testing Checklist

Before marking feature complete:

• Unit tests written and passing
• Integration tests passing
• Edge cases tested
• Error handling tested
• Performance is acceptable
• Manual testing completed
• No regressions in existing features
• Test coverage =80%

Example Usage

Input: "Create tests for the CSV export feature"

Output:

• Unit tests for export function
• Integration tests with file system
• Edge case tests (empty data, special characters)
• Performance tests (large datasets)