Firecrawl

Firecrawl MCP Server enables automated web scraping, site crawling, and data extraction through a scalable MCP interface. It combines powerful scraping, searching, crawling, and mapping capabilities with the Firecrawl API, making it easy to integrate web data workflows into your applications and pipelines.

How to use

Interact with the server through an MCP client to perform scraping, crawling, mapping, and data extraction tasks. You can run predefined actions like scraping a page, crawling a site, generating a site map, or extracting structured data from multiple URLs. Your MCP client will send commands to the server and receive structured results.

How to install

Prerequisites: ensure you have Node.js installed on your system. Use a current LTS version for compatibility.

Install the MCP server globally so you can run the command from anywhere.

# Global installation
npm install -g @modelcontextprotocol/mcp-server-firecrawl

# Local project installation
npm install @modelcontextprotocol/mcp-server-firecrawl

Configuration and usage notes

Set your API key to authorize requests to the Firecrawl backend. You can export it in your shell session or place it in a .env file for development. The key is required to access the Firecrawl features.

# Unix/Linux/macOS
export FIRECRAWL_API_KEY=your-api-key

# Windows Command Prompt
set FIRECRAWL_API_KEY=your-api-key

# Windows PowerShell
$env:FIRECRAWL_API_KEY = "your-api-key"

Start the server

Run the MCP server from your terminal. The server starts a local process that listens for MCP client commands.

mcp-server-firecrawl

Usage examples

Below are representative usage patterns you can run via your MCP client to perform common tasks. Each example shows the action name and arguments you would pass to the server.

{
  "name": "scrape_url",
  "arguments": {
    "url": "https://example.com",
    "formats": ["markdown"],
    "onlyMainContent": true
  }
}

{
  "name": "crawl",
  "arguments": {
    "url": "https://example.com",
    "maxDepth": 2,
    "limit": 100
  }
}

{
  "name": "map",
  "arguments": {
    "url": "https://example.com",
    "includeSubdomains": true,
    "limit": 1000
  }
}

{
  "name": "extract",
  "arguments": {
    "urls": ["https://example.com/product1", "https://example.com/product2"],
    "prompt": "Extract product details",
    "schema": {
      "name": "string",
      "price": "number",
      "description": "string"
    }
  }
}

Security and best practices

Protect your API key and enforce domain allowlisting to prevent unauthorized use. Enable rate limiting and provide clear error messages to clients.

Troubleshooting

If you encounter issues, verify the API key is set correctly, ensure the server is reachable by your MCP client, and check for any rate limit or network errors in the client logs.

Examples and notes

Common actions you can perform include web scraping, site crawling, site mapping, and data extraction using structured prompts and schemas. Use the provided command patterns to integrate these capabilities into your workflows.

Available tools

scrape

Perform web scraping with configurable options such as device emulation, ad blocking, content filtering, structured data extraction, and multiple output formats.

crawl

Traverse websites with depth control, path filtering, rate limiting, progress tracking, and sitemap integration.

map

Generate site structure maps with subdomain support, search filtering, link analysis, and visual hierarchy.

extract

Extract structured data from multiple URLs with schema validation and batch processing.