Fetcher MCP is a server that enables AI assistants to retrieve web page content using Playwright's headless browser capabilities. It excels at handling dynamic JavaScript-heavy websites by executing client-side code, and includes intelligent content extraction through Mozilla's Readability algorithm to remove ads and non-essential elements. The implementation offers both single URL and batch fetching tools, supports HTML or Markdown output formats, and automatically blocks unnecessary resources like images and stylesheets to improve performance. This tool is particularly valuable for research workflows, content analysis, and information gathering that requires processing of modern web applications.
Retrieve web page content from a specified URL. Parameters: url (string, required), timeout (number, optional, default is 30000), waitUntil (string, optional, default is 'load'), extractContent (boolean, optional, default is true), maxLength (number, optional), returnHtml (boolean, optional, default is false), waitForNavigation (boolean, optional, default is false), navigationTimeout (number, optional, default is 10000), disableMedia (boolean, optional, default is true), debug (boolean, optional)
Batch retrieve web page content from multiple URLs in parallel. Parameters: urls (array of strings, required), other parameters are the same as fetch_url.
No reviews yet. Be the first to review!
Sign in to join the conversation
