Scrapezy MCP Server (@scrapezy/mcp) is a Model Context Protocol (MCP) server that enables AI models to extract structured data from websites.
Key Capabilities
The server provides one tool: extract_structured_data, which is used to extract structured data from websites. This tool takes a URL and a prompt as required parameters, and returns structured data extracted from the website based on the prompt. The prompt should clearly describe what data needs to be extracted from the website.
Usage
Installation
Scrapezy MCP Server can be automatically installed into Claude Desktop via Smithery.
Manual Installation
There are two ways to provide the Scrapezy API key:
-
Environment variable: export SCRAPEZY_API_KEY=your_api_key npx @scrapezy/mcp
-
Command-line argument: npx @scrapezy/mcp --api-key=your_api_key
Using with Claude Desktop
The server configuration needs to be added. The configuration file locations are as follows:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%/Claude/claude_desktop_config.json
Example Usage in Claude
You can invoke this tool in Claude using a prompt similar to this. Claude will use this MCP server to extract the requested structured data from the website.
Debugging
Because MCP servers communicate via stdio, debugging can be challenging. It is recommended to use MCP Inspector, which is provided as a package script and will provide a URL so you can access the debugging tool in a browser.
Use Cases
Suitable for scenarios that require using AI models to extract structured data from websites, such as having MCP-enabled clients like Claude obtain and organize data directly from web pages based on natural language prompts.
Limitations
The materials only explainthe above functions and configuration methods; other aspects (such as specific pricing, supported platform scope, limits on the amount of data extracted, etc.) are not yet provided.
