This is an asynchronous web crawler framework developed with Python, capable of transforming website data into LLM-friendly output formats such as Markdown, JSON, etc. It is fully open-source and free, greatly simplifying the writing of asynchronous crawlers. Compared to the paid Firecrawl, it has faster crawling speed, supports simultaneous capture of multiple URLs, page screenshot, keyword optimized extraction (based on LLM), and complex multi-page session management.