Fetch MCP Server

@egoist

代码仓库

基本信息

分类

浏览器自动化

传输方式

stdio

作者

egoist

关于 Fetch MCP Server

An MCP server for fetching URLs / Youtube video transcript.

Fetch MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "fetch-mcp-egoist": {
      "command": "npx",
      "args": [
        "-y",
        "fetch-mcp"
      ]
    }
  }
}

Fetch MCP Server 工具

Fetch a URL, support HTML, text, and image

Fetch transcript for a Youtube video URL

概览

What is fetch-mcp?

fetch-mcp is a Model Context Protocol (MCP) server that fetches URLs and YouTube video transcripts. It is sponsored by ChatWise, an all-in-one LLM chatbot with first-class MCP support.

How to use fetch-mcp?

Run the server via npx. For a stdio server use npx -y fetch-mcp, for SSE use npx -y fetch-mcp --sse, and for a streamable HTTP server use npx -y fetch-mcp --http (optionally with a custom endpoint path like npx -y fetch-mcp --http /my-mcp).

Key features of fetch-mcp

Fetches URLs and returns content as HTML or Markdown.
Fetches YouTube video transcripts.
Supports stdio, SSE, and HTTP transports.
Easy to run via npx without installation.

Use cases of fetch-mcp

Retrieve web page content in Markdown format for LLM processing.
Obtain YouTube video transcripts for analysis or summarization.
Integrate URL fetching into any MCP-compatible client or host.

常见问题

What tools does fetch-mcp provide?

Two tools: `fetch_url` (fetches a URL, returns HTML or Markdown) and `fetch_youtube_transcript` (fetches a YouTube video transcript).

What transports are supported?

stdio, SSE (`--sse`), and streamable HTTP (`--http`, optionally with a custom endpoint).

How do I run fetch-mcp?

Use `npx -y fetch-mcp` with optional flags for transport mode.

Is any installation required?

No, it runs directly via npx.

Does fetch-mcp store any data?

The README does not mention any data storage or persistence.

浏览器自动化分类下的更多 MCP 服务器

playwright-server MCP server

blackwhite084

189

Fastcrw

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary.

286

TinyFish

tinyfish

Give your AI agent the live web. One MCP server for Search, Fetch, Browser, and Agent: browse real sites, extract structured data, fill forms, and run multi-step workflows. Search and Fetch are free.