SerpFire MCP Server

A powerful Model Context Protocol (MCP) server that combines Serper.dev (Google Search) and Firecrawl (Web Scraping) to provide AI agents with comprehensive research capabilities.

Features

• 🔍 Google Search via Serper.dev API
• 🌐 Web Scraping via Firecrawl API
• 🤖 Smart Research - Automatically search and scrape top results
• 🚀 Easy Integration with Cursor, Claude Desktop, and other MCP clients

Installation

npm install -g @zrald/serpfire

Prerequisites

You need API keys from:

• Serper.dev - Google Search API
• Firecrawl.dev - Web Scraping API

Configuration

For Cursor / Claude Desktop

Add to your MCP settings file (e.g., mcp.json):

{
  "mcpServers": {
    "serpfire": {
      "command": "npx",
      "args": ["-y", "@zrald/serpfire"],
      "env": {
        "SERPER_API_KEY": "your_serper_api_key_here",
        "FIRECRAWL_API_KEY": "your_firecrawl_api_key_here"
      }
    }
  }
}

Environment Variables

Alternatively, set environment variables:

export SERPER_API_KEY="your_serper_api_key"
export FIRECRAWL_API_KEY="your_firecrawl_api_key"

Available Tools

1. serper_search

Search Google and get structured results.

Arguments:

• query (string, required): Search query
• num (number, optional): Number of results (default: 10)

2. firecrawl_scrape

Scrape a single web page.

Arguments:

• url (string, required): URL to scrape
• formats (array, optional): Output formats (default: ["markdown"])

3. firecrawl_crawl

Crawl an entire website.

Arguments:

• url (string, required): Base URL to crawl
• limit (number, optional): Max pages to crawl

4. research_topic ⭐ (Recommended)

Automatically search a topic and scrape top results.

Arguments:

• topic (string, required): Topic to research
• max_sources (number, optional): Number of sources to scrape (default: 3)

How it works:

1. Searches Google for the topic
2. Extracts top N URLs
3. Scrapes all URLs in parallel
4. Returns synthesized research

Usage Examples

Once configured, ask your AI assistant:

• "Use serpfire to research the latest AI developments"
• "Search for MCP servers and scrape the top 3 results"
• "Find information about Firecrawl API features"

License

ISC

Author

Created by Zrald

Links

• GitHub Repository
• npm Package

💝 Support Development

If you find this project helpful, consider supporting development:

Crypto Donations

• BNB (Binance Smart Chain): 0xB8E0b6D4BaaFd1ac4De9245A760cB8F09bB7D084
• Bitcoin (BTC): bc1q77k0ju6ta3sp0vm3phm6dek432rzg7cqwf43z6
• Ethereum (ETH): 0xB8E0b6D4BaaFd1ac4De9245A760cB8F09bB7D084
• Polygon (MATIC): 0xB8E0b6D4BaaFd1ac4De9245A760cB8F09bB7D084
• Dogecoin (DOGE): DGbDQEgJLnNR2yEmWFrusFd7jLya4aoZMA

Your support helps maintain and improve this open-source project! 🙏