QA Flows Overview
QA Flows let you generate automated test cases from natural language requirements and run them against your application. Instead of writing tests by hand, describe what you want to test and let the agent build executable test flows for you.
How QA Flows work
Section titled “How QA Flows work”- Create a QA Flow — Provide a URL and requirements describing the behavior you want to test.
- Generate test cases — The agent explores your application and generates test cases that verify the requirements.
- Run tests — Execute the test cases against your application. Each execution is called a QA Run.
- Review results — See which tests passed, which failed, and why. Each failure is categorized for quick triage.
Key features
Section titled “Key features”- Automatic test generation — Describe what to test in plain language; the agent writes the test cases.
- Self-healing selectors — When your UI changes, the agent automatically attempts to find updated selectors so tests keep passing.
- Failure categorization — Each failure is classified as locator, assertion, timeout, flow change, or environment, making triage faster.
- Run history — Track every QA Run with pass/fail rates, duration, and detailed per-test results.
- Connector integration — Trigger QA Flows automatically via Connectors on a schedule or from CI events.
Failure categories
Section titled “Failure categories”| Category | Meaning |
|---|---|
| Locator | A UI element could not be found — usually caused by selector changes after a deploy |
| Assertion | The expected result did not match the actual result |
| Timeout | The application was too slow or unresponsive during the test |
| Flow change | The application workflow changed since the test was generated |
| Environment | An infrastructure or connectivity issue prevented the test from running |