Automates web browsing, UI verification, and data extraction using a command-line browser agent.
This skill empowers Claude to navigate websites, interact with UI elements, and perform end-to-end testing directly within the development workflow. By leveraging the agent-browser CLI and accessibility-based snapshots, it allows for precise element targeting, state persistence for authenticated sessions, and visual verification through screenshots. It is particularly useful for verifying deployed interfaces, reading live documentation, scraping web data, and capturing baseline states before making frontend changes to prevent design drift.
Key Features
01Accessibility-based element interaction using reference-based selection
02356 GitHub stars
03Session state management to handle authentication and persistent logins
04Full-page and viewport screenshot capture for visual UI verification
05Headless and headed execution modes for both CI/CD and debugging
06Advanced waiting mechanisms including network idle and JS conditions
Use Cases
01Verifying UI deployments and performing automated end-to-end testing
02Capturing baseline screenshots to prevent visual regression during design work
03Scraping data or reading live documentation directly from the web
