Overview & Capability
Open-source Python framework that connects AI models to browser actions (clicking, filling forms, extracting data, taking screenshots) with human-level reliability.
Key Highlights
- Vision-guided element selection with bounding boxes
- Handles complex logins, captchas, and multi-step checkouts
- Works with Playwright and any LLM with vision capabilities
Robbie's Independent Take
Verified: 2026-10-11
Best evaluated within a bounded workspace using version control. Configure strict tool approvals and MCP permission gates before granting terminal or file execution access. Monitor token spend carefully when running multi-turn autonomous refactoring loops.
Alternatives & Replaces
- Fragile Selenium web scrapers