UseScraper MCP Server

What is UseScraper MCP Server?

UseScraper MCP Server is a TypeScript-based MCP server that provides web scraping capabilities through the UseScraper API. It exposes a single tool for extracting content from web pages in text, HTML, or Markdown format. It is designed for developers integrating web scraping into AI assistants or workflows.

How to use UseScraper MCP Server?

Install via Smithery (npx -y @smithery/cli install usescraper-server --client claude) or manually (clone repo, npm install, npm run build). Configure Claude Desktop by adding the server to its config file with the USESCRAPER_API_KEY environment variable. Invoke the scrape tool with the required url parameter and optional format, advanced_proxy, or extract_object arguments.

Key features of UseScraper MCP Server

• Single scrape tool for web page content extraction
• Output formats: text, HTML, and Markdown
• Optional advanced proxy to circumvent bot detection
• Optional extract_object for structured data extraction
• Leverages UseScraper API for reliable scraping

Use cases of UseScraper MCP Server

• Extract article or blog content from any URL
• Scrape product data for price monitoring
• Convert web pages to Markdown for reading or processing
• Automate data collection from multiple URLs

FAQ from UseScraper MCP Server

What parameters does the scrape tool support?

The tool requires a url parameter. Optional parameters: format (text, html, markdown; default markdown), advanced_proxy (boolean; default false), and extract_object (object for structured extraction).

How do I configure UseScraper MCP Server with Claude Desktop?

Add a server entry to claude_desktop_config.json with the command node and the path to the built index.js file. Set the USESCRAPER_API_KEY environment variable to your API key.

What are the runtime dependencies?

Node.js and npm are required to build the server. A valid UseScraper API key is needed for web scraping.

How does the server communicate?

MCP servers communicate over stdio. For debugging, the README recommends using the MCP Inspector via npm run inspector.

Can I use the server programmatically?

Yes, the server exposes the scrape tool through the MCP interface, which can be called by any MCP client with the appropriate arguments.