Converts complex web pages into clean, LLM-friendly markdown for seamless data extraction and processing.
This skill integrates the high-performance Crawl4AI library directly into Claude Code, enabling the transformation of any URL into structured, readable markdown. It is specifically optimized for AI workflows, stripping away web clutter to provide high-quality text that enhances RAG pipelines, documentation analysis, and automated research. By leveraging Playwright and uv, it handles dynamic content efficiently while maintaining an isolated environment for reliable scraping within your development workflow.
主な機能
01Extracts hyperlinks and image references with configurable flags
02Automates browser management and environment setup using Playwright and uv
03Converts HTML to clean, structured markdown optimized for LLMs
04Supports headless and non-headless modes for debugging and production
050 GitHub stars
06Includes built-in diagnostic tools to verify installation and connection health
ユースケース
01Scraping online documentation to provide technical context for AI coding tasks
02Automating competitive research and data collection from news sites or blogs
03Gathering clean text data from websites for RAG (Retrieval-Augmented Generation) systems
