Captures web content from JavaScript-rendered pages and authenticated sites using the agent-browser CLI.
This skill empowers Claude to extract data from complex web environments that traditional scrapers cannot access, such as Single Page Applications (SPAs) and gated content. By leveraging the agent-browser CLI, it performs multi-step interactions like logging into private wikis, handling infinite scrolls, and navigating documentation trees to provide a clean, markdown-ready snapshot of required data for development or research.
主要功能
01Authentication flow handling for login-protected documentation
02Multi-page crawling for documentation trees and tutorial series
0369 GitHub stars
04Session state management to save and reuse authentication cookies
05Full JavaScript rendering for SPAs like React, Vue, and Angular
06Interactive element snapshots for precise content targeting
使用场景
01Scraping content from modern web apps that require client-side hydration
02Automating the collection of data from infinite-scroll or dynamic news feeds
03Extracting documentation from private or gated developer portals
