Fetcher MCP provides a Model Context Protocol server that fetches and extracts web page content using a Playwright headless browser, enabling support for JavaScript-rendered dynamic content. The server offers tools including `fetch_urls` for concurrent multi-URL retrieval and content extraction via a built-in Readability algorithm, with flexible output in HTML or Markdown formats and configurable options for timeouts and resource blocking. It solves the problem of accessing and processing modern web applications and dynamically-generated content that traditional web scrapers cannot handle.
claude mcp add --transport stdio jae-jae-fetcher-mcp uvx fetcher-mcp