Bridges Claude to the LionScraper browser extension so your AI can scrape lists, articles, links, images, and other structured data from live web pages. Ships as both Node and Python packages with three interfaces: an MCP stdio server for editors like Cursor, a CLI daemon for terminal workflows, and a loopback HTTP API at port 13808. The real extraction runs in the Chrome or Edge extension using full JavaScript execution. Falls back to simple HTTP fetch if the extension isn't connected. Install the extension from the Chrome or Edge store, match the bridge port in settings, then ask Claude to scrape in natural language. Useful when you need to extract data from complex pages that require rendering or interaction beyond static HTML parsing.
claude mcp add --transport stdio io.github.dowant-lionscraper-python uvx lionscraper-python