Overview
What is Firecrawl Server for MCP?
An Express API server for Firecrawl that provides endpoints for web crawling and scraping functionality.
How to use Firecrawl Server for MCP?
Install with Docker using docker compose up -d or without Docker using npm install then npm start. The server runs on port 3000 by default. Refer to the sample_requests directory for HTTP request examples.
Key features of Firecrawl Server for MCP
- Scrape a single URL
- Scrape multiple URLs in batch
- Crawl a website and scrape all pages
- Docker Compose support for easy deployment
- Persistent storage volume for downloaded files
Use cases of Firecrawl Server for MCP
- Extract content from a single web page via the
/scrapeendpoint - Gather data from multiple URLs in one request via
/scrape-batch - Crawl an entire website and collect all page content via
/crawl
FAQ from Firecrawl Server for MCP
How do I install Firecrawl Server for MCP without Docker?
Run npm install to install dependencies and npm start to start the server.
What API endpoints are available?
The server exposes GET / (home), POST /scrape, POST /scrape-batch, and POST /crawl.
What port does the server use?
It runs on port 3000 by default, or the port specified in the PORT environment variable.
How do I stop the Docker container?
Use docker compose stop to stop, or docker compose down to stop and remove the container.
Can I scale the service to multiple instances?
Yes, use docker compose up -d --scale firecrawl-server=3 to run multiple instances.
