Provides a self-hostable, Vercel-compatible Model Context Protocol (MCP) server to integrate Firecrawl's web content extraction and browser automation capabilities.
Sponsored
This tool acts as an MCP server for Firecrawl, allowing users to self-host and deploy a robust proxy for Firecrawl's powerful web scraping and browser automation features. Built with strict TypeScript, Effect-TS, and the Bun runtime, it offers a highly structured and extensible platform for AI models to interact with web content via a JSON-RPC 2.0 API. It's designed for seamless integration with platforms like Claude Desktop and LM Studio, providing fine-grained control over Firecrawl's functionalities, enhanced observability, and adherence to SOLID principles for maintainability.
Key Features
01Configurable cloud service and safe modes
02JSON-RPC 2.0 API for web content extraction and browser automation
032 GitHub stars
04Self-hostable and Vercel-compatible deployment
05Strict TypeScript, Effect-TS, and Bun runtime architecture
06Built-in observability with Prometheus metrics and structured logging
Use Cases
01Integrate Firecrawl's web scraping and browser automation into AI agents or LLMs via an MCP server.
02Deploy a customizable Firecrawl proxy service on Vercel for controlled access and enhanced features.
03Extend Firecrawl's capabilities with new tools and custom logic using a modular architecture.