Firecrawl MCP Server

This is a Model Context Protocol (MCP) server implementation that integrates with Firecrawl for web scraping. It offers features like scraping, crawling, searching, extracting, deep research, and batch scraping with JS rendering. It supports URL discovery, web search with content extraction, automatic retries with exponential backoff, and efficient batch processing with rate limiting.

The server monitors credit usage for cloud APIs, provides a comprehensive logging system, and supports both cloud and self-hosted FireCrawl instances. It also offers mobile/desktop viewport support and smart content filtering with tag inclusion/exclusion.

Installation is available via npx, manual installation, or Smithery. Configuration is managed through environment variables for API keys, retry settings, and credit monitoring. Available tools include firecrawl_scrape, firecrawl_batch_scrape, firecrawl_check_batch_status, firecrawl_search, firecrawl_crawl, firecrawl_extract, firecrawl_deep_research, and firecrawl_generate_llmstxt. The server provides robust error handling and detailed logging.