Agent Browser Automation

0•

Automates web browser interactions using a specialized CLI designed for AI agents to navigate, scrape, and test websites.

The agent-browser skill empowers Claude to perform complex web-based tasks by controlling Chrome or Chromium directly via the Chrome DevTools Protocol (CDP). It simplifies web automation for AI agents through a unique interactive snapshot system that assigns shorthand references to page elements, enabling reliable clicking, typing, and form submission. With robust support for authentication persistence, encrypted credential storage, and viewport emulation, it is a comprehensive solution for web scraping, end-to-end testing, and browser-based workflow automation.

主要功能

01Comprehensive capture tools for full-page, annotated, or visual-diff screenshots and PDFs
02Interactive element referencing (@e1, @e2) for precise AI control during snapshots
03Built-in device emulation and viewport scaling for mobile and responsive testing
04Multiple authentication strategies including state persistence, session management, and encrypted vaults
05Advanced waiting mechanisms for network idle, URL patterns, text changes, and downloads
060 GitHub stars

使用场景

01Performing end-to-end testing and visual regression checks for web developers
02Automated data extraction from complex JavaScript-heavy web applications
03Automating repetitive administrative tasks in authenticated dashboards and internal tools