Agent Browser CLI

•
38
•

Automates web browsing, form filling, and data extraction using a command-line interface optimized for AI agents.

Agent Browser is a specialized CLI-based browser automation skill that leverages Vercel's agent-browser to provide AI agents with a powerful interface for web interaction. By using an accessibility-tree-based reference system, it simplifies complex tasks like filling forms, clicking elements, and capturing screenshots without the overhead of heavy DOM manipulation. It supports parallel browser sessions, semantic locators, and structured JSON output, making it an ideal choice for developers who need to integrate real-world web navigation into their AI coding workflows while maintaining a lean, terminal-focused development environment.

主要功能

01Parallel browser management via named sessions for complex workflows.
0238 GitHub stars
03Comprehensive navigation controls including back, forward, and reload.
04Semantic locators to find elements by role, text, label, or placeholder.
05Ref-based element interaction using accessibility snapshots for precision.
06Full-page and viewport screenshot and PDF generation.

使用场景

01End-to-end testing of user flows like login, registration, and checkout.
02Automated documentation generation with full-page visual screenshots.
03Automated web scraping and data collection from dynamic, JavaScript-heavy sites.