Firecrawl
by Mendable
The Web Data API for AI — turn entire websites into LLM-ready markdown or structured data
browser Node.js Beginner Self-hostable Verified Featured
⭐ 86.7k stars 📅 Updated: 1w ago
Description
Firecrawl is a comprehensive web data platform that turns entire websites into LLM-ready markdown or structured data. Its MCP server exposes powerful scraping, crawling, site mapping, search, and AI agent capabilities directly to AI coding assistants. Features include structured JSON extraction with schema validation, browser automation for JavaScript-heavy pages, PDF and DOCX parsing, full-page screenshots, and change tracking for monitoring content updates. The AI-powered extraction agent can navigate multi-step workflows to gather data automatically. Firecrawl handles the complexity of modern web pages — dynamic rendering, anti-bot measures, and pagination — so your AI gets clean, structured content every time.
✅ Best for
AI workflows that need clean, structured web data from any website without dealing with scraping complexity
⏭️ Skip if
You only need to read simple static pages — a basic fetch tool would suffice
💡 Use cases
- Scraping web pages and converting them to clean markdown for AI consumption
- Crawling entire websites to build comprehensive knowledge bases
- Extracting structured data from web pages using JSON schemas
- Monitoring websites for content changes and updates
👍 Pros
- ✓ Handles JavaScript rendering, anti-bot measures, and complex page structures automatically
- ✓ Supports structured JSON extraction with schema validation
- ✓ AI agent mode for multi-step automated data gathering workflows
- ✓ Massive community adoption (86k+ stars) with active development
👎 Cons
- ✗ Requires API key — free tier has rate limits, production use needs a paid plan
- ✗ AGPL-3.0 license may be restrictive for some commercial use cases
- ✗ Self-hosting the full platform requires significant infrastructure
🔧 Exposed tools (6 tools)
| Tool | Category | Description |
|---|---|---|
| agent | automation | AI agent that navigates multi-step workflows to gather data automatically |
| map | discovery | Generate a sitemap of all discoverable URLs on a website |
| search | discovery | Search the web and return results as markdown content |
| scrape | extraction | Scrape a single URL and convert to markdown or structured data |
| crawl | extraction | Crawl an entire website following links up to a specified depth |
| batch | extraction | Scrape multiple URLs in parallel for efficient bulk data extraction |
⚡ Installation
Prerequisites:
- • Node.js v18+
- • Firecrawl API key from firecrawl.dev
- • API key required
Check Claude Code documentation to configure this MCP server.
💡 Tips & tricks
Use the structured extraction feature with a JSON schema to get precisely formatted data from web pages. The crawl tool with site mapping is ideal for building documentation knowledge bases. Start with the hosted API to avoid self-hosting complexity.
Quick info
- Author
- Mendable
- License
- AGPL-3.0
- Runtime
- Node.js 18+
- Transport
- stdio
- Category
- browser
- Difficulty
- Beginner
- Self-hostable
- ✅
- Auth
- —
- Docker
- —
- Version
- latest
- Updated
- Feb 28, 2026
Client compatibility
- ✅ Claude Code
- ✅ Cursor
- ❓ VS Code Copilot
- ❓ Gemini CLI
- ❓ Windsurf
- ❓ Cline
- ❓ JetBrains AI
- ❓ Warp
Platforms
🍎 macOS 🐧 Linux 🪟 Windows