---

name: tester description: Comprehensive testing skill for GabeDA application - designs test strategies (UAT, integration, smoke, unit), creates tests for frontend (React/Playwright) and backend (Django/pytest), executes tests, analyzes results, and generates detailed reports with findings. Stores reports in ai/testing/ and tests in appropriate project folders.

Tester Skill - GabeDA Testing Framework

This skill provides comprehensive testing capabilities for the GabeDA full-stack application (React frontend + Django backend). Design test strategies, create tests, execute them, analyze results, and generate actionable reports.

Purpose

Ensure quality and reliability of the GabeDA application through systematic testing across all layers:

• Frontend Testing: React components, user flows, E2E scenarios (Playwright)
• Backend Testing: Django API endpoints, models, business logic (pytest)
• Integration Testing: Full-stack workflows across frontend and backend
• UAT Testing: User acceptance scenarios matching real business use cases
• Smoke Testing: Critical path validation for rapid deployment verification

When to Use This Skill

Use this skill when:

• Creating tests for new features or existing functionality
• Executing test suites and collecting results
• Analyzing test failures and generating bug reports
• Designing test strategies for sprints or releases
• Validating deployments with smoke tests
• Performing UAT with business stakeholders
• Generating test reports for documentation or review

Testing Standards & Principles

Test Design Standard

Follow the AAA Pattern (Arrange-Act-Assert):

Copydef test_example():
    # Arrange - Set up test data and preconditions
    user = create_test_user()

    # Act - Perform the action being tested
    result = user.login(email, password)

    # Assert - Verify the expected outcome
    assert result.is_authenticated == True

Test Coverage Goals

• Critical paths: 100% coverage (login, company creation, data upload)
• Core features: ≥90% coverage (dashboard, analytics, reports)
• Edge cases: ≥70% coverage (error handling, validation)
• UI components: ≥80% component coverage

Test Naming Convention

Copy# Backend (pytest)
def test_<feature>_<scenario>_<expected_result>():
    # Example: test_login_with_valid_credentials_returns_tokens()
    pass

# Frontend (Playwright)
test('<feature> - <scenario> - <expected result>', async () => {
    // Example: test('Login - valid credentials - redirects to dashboard')
})

Project Structure

CopyGabeDA Project:
├── gabeda_frontend/
│   └── tests/                          # Frontend tests (Playwright)
│       ├── e2e/                        # End-to-end tests
│       ├── integration/                # Integration tests
│       └── smoke/                      # Smoke tests
├── gabeda_backend/
│   └── tests/                          # Backend tests (pytest)
│       ├── unit/                       # Unit tests
│       ├── integration/                # Integration tests
│       └── api/                        # API endpoint tests
└── khujta_ai_business/
    └── ai/testing/                     # Test reports & strategies
        ├── reports/                    # Test execution reports
        │   └── YYYY-MM-DD_HH-MM_<test-type>_report.md
        ├── strategies/                 # Test strategy documents
        └── findings/                   # Bug reports & findings

Test Types & When to Use

1. Unit Tests

Purpose: Test individual functions/methods in isolation When: After creating any new function, API endpoint, or component Location: Backend tests/unit/, Frontend component tests Example:

Copy# Backend: Test RUT validation
def test_validate_rut_with_valid_format_returns_cleaned():
    result = validate_rut_field('12.345.678-9')
    assert result == '123456789'

2. Integration Tests

Purpose: Test interactions between multiple components/services When: After integrating multiple modules or connecting frontend-backend Location: Both tests/integration/ Example: Test complete login flow (frontend → API → database → response)

3. E2E Tests (Playwright)

Purpose: Test complete user workflows from browser perspective When: For critical user journeys and multi-step workflows Location: gabeda_frontend/tests/e2e/ Example: User registers → creates company → uploads CSV → views dashboard

4. Smoke Tests

Purpose: Quick validation of critical functionality after deployment When: Before/after deployments, CI/CD pipeline Location: Both tests/smoke/ Duration: Must complete in <5 minutes Example: Can login? Can create company? Can upload file?

5. UAT (User Acceptance Tests)

Purpose: Validate business requirements with stakeholders When: Before major releases, sprint demos Location: ai/testing/strategies/uat_<feature>.md Format: Written scenarios with expected outcomes

Workflow

Creating Tests

Step 1: Analyze the Feature

• Understand the feature requirements
• Identify critical paths and edge cases
• Determine appropriate test types (unit, integration, E2E)

Step 2: Design Test Cases

• List scenarios to test (happy path, edge cases, error cases)
• Define test data requirements
• Plan assertions and expected outcomes

Step 3: Write Tests

Copy# Frontend (Playwright/Vitest)
# Location: gabeda_frontend/tests/<type>/

# Backend (pytest)
# Location: gabeda_backend/tests/<type>/

Step 4: Execute and Verify

• Run tests locally
• Fix any failures
• Ensure tests are deterministic (no flaky tests)

Executing Tests

Frontend Tests:

Copycd gabeda_frontend

# Run all tests
npm run test

# Run E2E tests
npm run test:e2e

# Run specific test file
npm run test tests/e2e/login.spec.ts

Backend Tests:

Copycd gabeda_backend

# Run all tests
./benv/Scripts/python -m pytest

# Run specific test file
./benv/Scripts/python -m pytest tests/unit/test_rut_validation.py

# Run with coverage
./benv/Scripts/python -m pytest --cov=apps --cov-report=html

Generating Reports

After test execution, generate a comprehensive report:

Report Structure (stored in ai/testing/reports/):

Copy# Test Report: <Test Type> - <Date>

## Summary
- Total Tests: X
- Passed: X
- Failed: X
- Skipped: X
- Duration: X minutes
- Coverage: X%

## Test Results by Category
[Breakdown by feature/module]

## Failed Tests
[For each failure:]
- Test Name
- Error Message
- Stack Trace
- Reproduction Steps
- Suggested Fix

## Coverage Analysis
[Areas with low coverage]

## Recommendations
[Next steps, areas needing attention]

Test Execution Commands

Quick Reference

Copy# Frontend - All tests
cd gabeda_frontend && npm run test

# Frontend - E2E only
cd gabeda_frontend && npm run test:e2e

# Frontend - Watch mode
cd gabeda_frontend && npm run test:watch

# Backend - All tests
cd gabeda_backend && ./benv/Scripts/python -m pytest

# Backend - With coverage
cd gabeda_backend && ./benv/Scripts/python -m pytest --cov --cov-report=term-missing

# Backend - Specific module
cd gabeda_backend && ./benv/Scripts/python -m pytest tests/unit/accounts/

# Backend - Stop on first failure
cd gabeda_backend && ./benv/Scripts/python -m pytest -x

# Backend - Verbose output
cd gabeda_backend && ./benv/Scripts/python -m pytest -v

Example: Creating a Test for the Dashboard Company Switcher

Scenario: Test that users with multiple companies can switch between them

Step 1: Test Design

CopyFeature: Company Switcher
Test Type: E2E (Playwright)
Location: gabeda_frontend/tests/e2e/company-switcher.spec.ts

Scenarios:
1. User with 1 company - switcher not visible
2. User with 2+ companies - switcher visible
3. Switching companies - dashboard updates
4. Selected company persists - on page reload

Step 2: Implementation (see references/test-examples.md for complete code)

Step 3: Execution

Copycd gabeda_frontend
npm run test:e2e tests/e2e/company-switcher.spec.ts

Step 4: Report Generation Create report in ai/testing/reports/2025-11-01_18-30_e2e-company-switcher_report.md with results

Best Practices

Test Data Management

• Use factories/fixtures for test data creation
• Clean up after tests (database transactions, file cleanup)
• Don't rely on production data - create isolated test data
• Use realistic data - matching production data shapes

Test Independence

• Each test must be independent - can run in any order
• No shared state between tests
• Reset database/state between test runs

Assertions

• Be specific - assert exact values, not just truthy/falsy
• One logical assertion per test - split complex scenarios
• Clear error messages - make failures easy to debug

Performance

• Keep unit tests fast (<100ms each)
• Parallelize when possible - run independent tests concurrently
• Mock external services - don't hit real APIs in tests

References

• references/test-examples.md - Complete test examples for all test types
• references/playwright-patterns.md - Playwright best practices and patterns
• references/pytest-patterns.md - Pytest fixtures and patterns
• references/test-data-factories.md - Test data creation strategies

Troubleshooting

Flaky Tests: Tests that pass/fail inconsistently

• Add explicit waits for async operations
• Use Playwright's built-in wait mechanisms
• Check for race conditions in test setup

Slow Tests: Tests taking too long

• Mock external API calls
• Use in-memory database for unit tests
• Parallelize test execution

Failed Assertions: Tests failing unexpectedly

• Check test data setup
• Verify environment variables
• Review recent code changes

Output Locations

Test Files:

• Frontend: C:\Projects\play\gabeda_frontend\tests\
• Backend: C:\Projects\play\gabeda_backend\tests\

Test Reports:

• All reports: C:\Projects\play\khujta_ai_business\ai\testing\reports\
• Test strategies: C:\Projects\play\khujta_ai_business\ai\testing\strategies\
• Bug findings: C:\Projects\play\khujta_ai_business\ai\testing\findings\

Skill Usage Examples

Example 1: Create E2E test for company dashboard

CopyUser: "Create an E2E test that logs in a test user, creates 2 companies,
       and verifies both companies show in the switcher"

Tester Skill:
1. Creates test file at gabeda_frontend/tests/e2e/test-dashboard-multi-company.spec.ts
2. Implements test with Playwright
3. Executes test and collects results
4. Generates report at ai/testing/reports/2025-11-01_19-00_e2e-dashboard_report.md

Example 2: Run smoke tests before deployment

CopyUser: "Run smoke tests for the frontend and backend"

Tester Skill:
1. Executes frontend smoke tests
2. Executes backend smoke tests
3. Collects all results
4. Generates combined report with pass/fail status
5. Recommends proceed/block deployment based on results

Example 3: Design UAT strategy for new feature

CopyUser: "Design UAT test cases for the file upload feature"

Tester Skill:
1. Reviews feature requirements
2. Creates UAT strategy document at ai/testing/strategies/uat_file_upload.md
3. Lists scenarios: valid CSV, invalid format, large files, etc.
4. Defines acceptance criteria for each scenario
5. Provides test data samples