Agent Browser Automation

소개

This skill empowers Claude to perform sophisticated web-based tasks by acting as a coordinator for the Agent Browser CLI. It enables Claude to navigate websites, fill forms, and extract data through a specialized subagent, significantly reducing context window consumption compared to traditional DOM-heavy methods. It is ideal for developers needing to bridge the gap between their terminal and the web, allowing for seamless automation of logins, multi-page scraping, and UI state verification.

주요 기능

• Efficient browser navigation via subagent delegation for reduced context usage
• Automated form filling, button clicks, and multi-step workflow execution
• Support for multi-session browser management and screenshot capture
• Structured data extraction from dynamic and JavaScript-heavy websites
• 0 GitHub stars
• Interactive element identification using snapshot-based numbering

사용 사례

• Performing multi-page web scraping where data needs to be aggregated and summarized
• Automating login sequences to verify dashboard states or fetch private data
• Collecting pricing or star-count data from multiple repository or product pages