What problem does it solve?
It removes the manual effort of clicking, typing, navigating, and screenshotting web pages by letting an LLM drive a real browser workflow deterministically from a captured page snapshot.
Core Features & Use Cases
- Browser interaction automation: Navigate pages, click and type into UI elements, select options, drag/hover, and handle dialogs.
- Evidence and outputs: Create screenshots and PDFs, evaluate page JavaScript, and inspect console/network/DevTools signals.
- Stateful testing workflows: Manage sessions, persist profiles, save and restore storage state (cookies/localStorage/sessionStorage), and support concurrent named browser sessions.
- Use case: Validate a login and onboarding flow by opening the site, filling credentials, submitting, capturing a screenshot of the post-login dashboard, and saving storage state for reuse in subsequent tests.
Quick Start
Open a browser session and navigate to the target URL, then take a snapshot to identify element references for the next steps.