About Fetch MCP Server
Fetch MCP Server is a Node.js MCP server for AI applications that need web content and search results. It uses Playwright to fetch pages and can convert their HTML to Markdown, with SSE communication.
Key features
- 01Fetch multiple URLs with Playwright
- 02Search the web and retrieve SERP results
- 03Convert HTML content to Markdown
- 04Manage resources with templates
- 05Communicate through SSE
Use cases
- 01Fetch several webpages for AI processing
- 02Search for pages and retrieve search results
- 03Convert webpage content into Markdown
MCP support
Exposes `fetchUrls` and `search` tools to MCP clients. It supports SSE connections through `/sse` and messages through `/messages`.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEFetch MCP Server
A Model Context Protocol (MCP) server implementation with web content fetching capabilities, enabling server-sent events (SSE) based communication.
Overview
Fetch MCP Server provides a Model Context Protocol implementation that allows for real-time communication using Server-Sent Events (SSE). It features web content fetching and search capabilities, designed to work with AI applications and provides a standardized way of managing and communicating with models and resources.
Features
- Model Context Protocol (MCP) implementation
- Server-Sent Events (SSE) for real-time communication
- Web content fetching with Playwright
- Google search capability with SERP results
- HTML to Markdown conversion
- Resource templating and management
Prerequisites
- Node.js (v14 or higher recommended)
- Yarn package manager
- Kubernetes cluster (for deployment)
Installation
Clone the repository and install dependencies:
git clone [email protected]:pnparadise/fetch-mcp-server.git
cd fetch-mcp-server
yarn install
Usage
Starting the Server
To start the server in development mode:
yarn dev
This will run the server with the MCP inspector for debugging.
To start the server in production mode:
yarn start
By default, the server runs on port 3001. You can override this by setting the PORT environment variable:
PORT=5000 yarn start
Endpoints
/sse- SSE endpoint for establishing persistent connections/messages- Endpoint for sending messages to the MCP server
Available Tools
fetchUrls- Fetch multiple URLs and convert their content to markdownsearch- Search and retrieve content from web pages with SERP results
Project Structure
src/index.ts- Main entry point of the applicationsrc/mcp-server.ts- Implementation of the MCP server and tool definitionssrc/sse-server.ts- SSE server implementation using Expressk8s/- Kubernetes deployment configurations
Dependencies
- @modelcontextprotocol/sdk - SDK for Model Context Protocol
- Express - Web framework for Node.js
- Playwright - Browser automation library
- Turndown - HTML to Markdown converter
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.