How to Install Firecrawl MCP in Gemini CLI
Advanced web scraping with JavaScript rendering, PDF support, and autonomous research agents
What Firecrawl MCP does
Firecrawl is the most powerful web scraping MCP server — handling JavaScript-rendered pages, PDFs, sitemaps, and multi-page crawls that simpler tools can't manage. Features: - Scrape any URL to clean markdown, even JS-heavy SPAs - Crawl entire websites with depth control and URL filtering - Extract structured data using custom schemas - Map website structure and discover all pages - Deep Research agent — autonomously researches topics across multiple sources - PDF text extraction with accurate layout preservation - Sitemap parsing and discovery - Smart rate limiting and anti-bot handling - Batch scraping with concurrent processing Why Firecrawl over Fetch/Puppeteer? - Handles headless JS rendering automatically - Built-in residential proxies for blocked sites - Returns clean, LLM-optimized markdown (not raw HTML) - The Deep Research tool can tackle hour-long research tasks autonomously Free tier: 500 credits/month. Plans from $16/month.
Installation steps
- 1Copy the config below
- 2Open ~/.gemini/settings.json
- 3Paste inside "mcpServers": {}
- 4Restart Gemini CLI
Configuration
File location: ~/.gemini/settings.json
{
"firecrawlMcp": {
"command": "npx",
"args": [
"-y",
"firecrawl-mcp"
],
"env": {
"FIRECRAWL_API_KEY": "fc-..."
}
}
}Install in other clients
For maintainers
If you maintain Firecrawl MCP, add this badge to your README to show it's verified on CuratedMCP:
[](https://curatedmcp.com/marketplace/firecrawl-mcp)
Related servers
Brave Search MCP
Real-time web and local search using the Brave Search API — no tracking, no ads
Puppeteer MCP Server
Browser automation for screenshots, scraping, and JavaScript-rendered page interaction
Fetch MCP Server
Fetch any URL and convert web content to clean markdown for AI consumption