Verified project record

firecrawl/firecrawl-mcp-server

An MCP server that gives AI agents access to live web data through search, scraping, crawling, autonomous research, and browser interaction. It offers multiple deployment options, including a hosted keyless tier and local execution via a single command.

Project overview

Project type: MCP · AI Search
Use cases: Search & Research · Automation
Deployment: Refer to project documentation
License: MIT

Best for

Developers and AI engineers needing a comprehensive suite of web interaction tools for MCP clients.

Key capabilities

Scrapes single URLs into clean, structured JSON, markdown, or branding profiles with advanced options like PII redaction.
Searches the web and optionally extracts content from search results with feedback mechanisms.
Starts a crawl job to extract content from multiple related pages with internal polling.
Maps a website to discover all indexed URLs before scraping.
An asynchronous AI agent layer that independently browses the internet, searches, and extracts structured data based on a prompt.
Clicks, types, navigates, and extracts state from dynamic web pages.
Searches and inspects research papers and GitHub repositories.
Creates and manages recurring page monitors with scheduled scrapes and diffing.

Limitations and risks

Hosted MCP cannot read the local filesystem directly for parsing; requires an upload-ref flow or self-hosted instance.
Crawl responses can be very large and may exceed token limits.
Scrape tool only accepts a single URL, not a list.
Usage is subject to API rate limits which may cause errors.

Getting started

Setup is rated as easy, supporting Npx one-liner execution, a hosted keyless endpoint, and click-to-install badges for VS Code. A basic run can be initiated with 'env FIRECRAWL_API_KEY=fc-YOUR_API_KEY npx -y firecrawl-mcp'.

Evidence and sources

README: A Model Context Protocol (MCP) server that brings [Firecrawl](https://github.com/firecrawl/firecrawl) to MCP-compatible AI agents — search, scrape, and interact with the live web…
README: - Scrape any URL into clean, structured data - Interact with pages — click, navigate, and operate
README: - Search the web and get full page content - Scrape any URL into clean, structured data - Interact with pages — click, navigate, and operate - Deep research with autonomous agent…
GitHub project description: 🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
README: Scrape content from a single URL with advanced options.

Reviewed at: 2026-07-18T06:51:37Z