Automation Testing

knowledge skill

The agent writes and executes automated test scripts to verify software functionality against specified requirements. It produces a test execution report detailing passed and failed assertions.

Worked examples

• {"input": {"target_url": "https://store.example.com/login", "test_scenarios": ["Enter valid credentials and click submit", "Verify redirect to /dashboard"], "framework": "Playwright"}, "output": {"test_script": "await page.fill('#user', 'admin'); await page.click('#submit'); await expect(page).toHaveURL(/.*dashboard/);", "execution_results": {"login_success": "passed"}, "error_logs": []}}
• {"input": {"target_url": "https://api.example.com/v1/users", "test_scenarios": ["Send GET request without auth token", "Verify 401 Unauthorized response"], "framework": "Pytest"}, "output": {"test_script": "def test_auth(): response = requests.get(url); assert response.status_code == 401", "execution_results": {"auth_check": "failed"}, "error_logs": ["AssertionError: 200 != 401"]}}

Input

• target_url: string: The endpoint or page to be tested
• test_scenarios: array: List of user actions and expected outcomes
• framework: string: The testing library to use (e.g., Playwright, Selenium)

Output

• test_script: string: The executable code for the automation
• execution_results: object: Map of test cases to pass/fail status
• error_logs: array: Stack traces for failed assertions

Details

• Skill type: knowledge skill
• Safety level: safe_public_research
• Version: 1.0.0