Memory
The AI builds knowledge about your application as it works, so tests get more accurate over time.
What is Memory
As the AI explores your app, runs tests, and executes tasks, it learns what pages exist, how navigation works, what forms and interactions are available, and how features behave. All of this gets stored as Memory.
The practical effect: the AI gets better at your specific application the more it works with it. Test generation becomes more relevant, and execution becomes more reliable.
How Memory helps
When the AI already knows your application, it generates test cases that cover the right flows and edge cases instead of guessing. It navigates pages confidently instead of hunting for elements, which cuts down on flaky results. And it makes better calls about what counts as a pass or fail because it has context about your domain.
How Memory grows
Memory builds up naturally as you use qtrl. The most direct way to build it is through exploration tasks, where the AI deliberately browses your application to map it out. This is worth doing when you first set up a project or after major changes to your app.
Test execution and test generation also contribute. Every time the AI runs a test, it picks up information about how pages respond and how workflows actually behave. Even generating tests adds knowledge, since the AI explores your app while figuring out what to cover.
Tips
Run an exploration task before generating tests on a new project. The upfront investment pays off in better test quality from the start.
After major application updates, run another exploration to keep the AI current. If you redesigned your navigation or added a new section, the AI needs to see it before it can test it well.
Memory is project-specific. Each project maintains its own knowledge, scoped to the application you're working with.