Captures web content from JavaScript-heavy, login-protected, and multi-page sites using the agent-browser CLI.
This skill empowers Claude to navigate and extract data from complex web environments where traditional static scrapers often fail. By leveraging the agent-browser CLI, it effectively handles client-side rendering for frameworks like React and Vue, manages authentication flows to access gated documentation, and performs automated multi-page crawls. It is an essential tool for data pipeline engineers who need to ingest high-quality content from dynamic sources for RAG systems, technical documentation, or competitive analysis while maintaining session states and respecting rate limits.
主な機能
01Client-side rendering support for modern SPAs (React, Vue, Angular)
02Integrated screenshot and JS console debugging for complex page structures
03Interactive element snapshots for precise content targeting and extraction
04Automated multi-page site crawling for documentation and tutorials
05Secure authentication handling with session state persistence and reuse
0669 GitHub stars
ユースケース
01Scraping data from dynamic, infinite-scroll, or lazy-loading web applications
02Compiling complete documentation sets from multi-page developer portals into clean text
03Extracting content from login-protected private wikis or internal company documentation
