e2e
End-to-end test framework where an AI agent drives web and mobile apps
What it is
A TypeScript testing framework for browser and iOS/Android apps in which test steps can be written as plain-language goals that an AI agent carries out, alongside ordinary locators and assertions. Once an agent action passes verification, it is recorded and replayed on later runs, so the model is only called again after the app's UI changes. Web runs go through Playwright engines, mobile runs through simulators, emulators, or hosted device clouds, and results can be posted as pull request comments. Frontend and mobile teams write and maintain UI tests with it, using their own API key, subscription, or local model.