scraperis mcp

Local 2025-08-31 23:17:39 0

Enables extracting data from websites using natural language prompts, allowing users to specify exactly what content they want in plain English and returning structured JSON data.

A Model Context Protocol (MCP) integration for Scraper.is - A powerful web scraping tool for AI assistants.

This package allows AI assistants to scrape web content through the MCP protocol, enabling them to access up-to-date information from the web.

Features

? Web Scraping: Extract content from any website
? Screenshots: Capture visual representations of web pages
? Multiple Formats: Get content in markdown, HTML, or JSON
? Progress Updates: Real-time progress reporting during scraping operations
? MCP Integration: Seamless integration with MCP-compatible AI assistants

Installation

npm install -g scraperis-mcp

Or with yarn:

yarn global add scraperis-mcp

Prerequisites

You need a Scraper.is API key to use this package.

Getting Your API Key

Sign up or log in at scraper.is
Navigate to the API Keys section in your dashboard: https://www.scraper.is/dashboard/apikeys
Create a new API key or copy your existing key
Store this key securely as you'll need it to use this package

Usage

Environment Setup

Create a .env file with your Scraper.is API key:

SCRAPERIS_API_KEY=your_api_key_here

Claude Desktop Integration

To use this package with Claude Desktop:

Install the package globally:
```
npm install -g scraperis-mcp
```

Add the following configuration to your claude_desktop_config.json file:

{
  "mcpServers": {
    "scraperis_scraper": {
      "command": "scraperis-mcp",
      "args": [],
      "env": {
        "SCRAPERIS_API_KEY": "your-api-key-here",
        "DEBUG": "*"
      }
    }
  }
}

Replace your-api-key-here with your actual Scraper.is API key.
Restart Claude Desktop to apply the changes.

Running with MCP Inspector

For development and testing, you can use the MCP Inspector:

npx @modelcontextprotocol/inspector scraperis-mcp

Integration with AI Assistants

This package is designed to be used with AI assistants that support the Model Context Protocol (MCP). When properly configured, the AI assistant can use the following tools:

Scrape Tool

The scrape tool allows the AI to extract content from websites. It supports various formats:

markdown: Returns the content in markdown format
html: Returns the content in HTML format
screenshot: Returns a screenshot of the webpage
json: Returns structured data in JSON format

Example prompt for the AI:

Can you scrape the latest news from techcrunch.com and summarize it for me?

API Reference

Tools

scrape

Scrapes content from a webpage based on a prompt.

Parameters:

prompt (string): The prompt describing what to scrape, including the URL
format (string): The format to return the content in (markdown, html, screenshot, json, quick)

Example:

{
  "prompt": "Get me the top 10 products from producthunt.com",
  "format": "markdown"
}

Development

Setup

Clone the repository:

git clone https://github.com/Ai-Quill/scraperis-mcp.git
cd scraperis-mcp

Install dependencies:
```
npm install
```
Build the project:
```
npm run build
```

Scripts

npm run build: Build the project
npm run watch: Watch for changes and rebuild
npm run dev: Run with MCP Inspector for development
npm run test: Run tests
npm run lint: Run ESLint

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgements

[
  {
    "description": "Extract data from websites using natural language prompts. The prompt should include the website URL and what data you want to extract. For example: 'Get me the top 10 products from producthunt.com' or 'Extract all article titles and authors from techcrunch.com/news'",
    "inputSchema": {
      "properties": {
        "parse": {
          "anyOf": [
            {
              "type": "boolean"
            },
            {
              "type": "null"
            }
          ],
          "default": {
            "path": "yarn.lock",
            "type": "blob"
          },
          "description": "Should result be parsed. If result should not be parsed then html will be stripped and converted to markdown file",
          "title": "Parse"
        },
        "prompt": {
          "description": "Natural language prompt describing what to extract and from where. For example: 'Get me the top 10 products from producthunt.com' or 'Find all articles about AI from techcrunch.com'",
          "title": "Prompt",
          "type": "string"
        }
      },
      "required": [
        "prompt"
      ],
      "title": "scrape_urlArguments",
      "type": "object"
    },
    "name": "scraperis_scraper"
  }
]