Firecrawl MCP Server

About

## Firecrawl MCP Server: AI-Powered Web Scraping The **Firecrawl MCP Server** integrates the AI-optimized web scraping service into Google Antigravity, enabling developers to crawl websites and extract clean, structured data optimized for LLM consumption. ### Why Firecrawl MCP? - **LLM-optimized** - Output formatted for AI models - **Smart crawling** - Automatic link following - **Clean extraction** - Remove ads, navigation, clutter - **Markdown output** - Clean markdown formatting - **JavaScript rendering** - Handle dynamic content ### Key Features #### 1. Page Scraping ```python # Scrape single page page = await mcp.call("firecrawl", "scrape", { "url": "https://example.com/article", "formats": ["markdown", "html"] }) print(f"Title: {page[\"metadata\"][\"title\"]}") print(f"Content length: {len(page[\"markdown\"])}") ``` #### 2. Site Crawling ```python # Crawl entire site crawl = await mcp.call("firecrawl", "crawl", { "url": "https://docs.example.com", "limit": 100, "max_depth": 3 }) print(f"Crawl ID: {crawl[\"id\"]}") # Get crawl status status = await mcp.call("firecrawl", "get_crawl_status", { "crawl_id": crawl["id"] }) print(f"Pages crawled: {status[\"completed\"]}/{status[\"total\"]}") ``` #### 3. Structured Extraction ```python # Extract with schema data = await mcp.call("firecrawl", "extract", { "url": "https://example.com/product", "schema": { "type": "object", "properties": { "name": {"type": "string"}, "price": {"type": "number"} } } }) print(f"Product: {data[\"name\"]} - ${data[\"price\"]}") ``` #### 4. Batch Processing ```python # Scrape multiple URLs results = await mcp.call("firecrawl", "batch_scrape", { "urls": [ "https://example.com/page1", "https://example.com/page2" ] }) for result in results: print(f"URL: {result[\"url\"]}") ``` ### Configuration ```json { "mcpServers": { "firecrawl": { "command": "npx", "args": ["-y", "@anthropic/mcp-firecrawl"], "env": { "FIRECRAWL_API_KEY": "fc-xxx" } } } } ``` ### Use Cases **RAG Data Collection**: Gather data for AI retrieval. **Documentation Ingestion**: Crawl docs for knowledge bases. **Content Analysis**: Extract clean content for analysis. **Competitive Intel**: Monitor competitor websites. The Firecrawl MCP Server enables AI-optimized web scraping.

Installation

{
  "mcpServers": {
    "firecrawl": {
      "mcpServers": {
        "firecrawl": {
          "env": {
            "FIRECRAWL_API_KEY": "YOUR_FIRECRAWL_API_KEY"
          },
          "args": [
            "-y",
            "firecrawl-mcp"
          ],
          "command": "npx"
        }
      }
    }
  }
}

How to Use

1. 11. Get API key from firecrawl.dev/app/api-keys
2. 22. Scrape single pages or crawl entire sites
3. 33. Returns content in markdown/JSON format