mcp-server-fetch-typescript MCP Server

Name: mcp-server-fetch-typescript MCP Server
Author: MCP-Mirror

@MCP-Mirror

代码仓库

关于 mcp-server-fetch-typescript MCP Server

Mirror of

#mcp-server #web-fetching #data-extraction

基本信息

分类

浏览器自动化

传输方式

stdio

发布者

MCP-Mirror

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "tatn_mcp-server-fetch-typescript": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "npx",
        "-y",
        "mcp-server-fetch-typescript"
      ]
    }
  }
}

工具

Retrieve raw text content directly from URLs

Fetch fully rendered HTML content

Convert web content to Markdown format

Extract and convert main content

概览

What is mcp-server-fetch-typescript?

A Model Context Protocol server that provides web content fetching and conversion capabilities, with support for various formats and rendering methods. It is intended for tasks ranging from simple data extraction to sophisticated web scraping.

How to use mcp-server-fetch-typescript?

Install globally via npm install -g mcp-server-fetch-typescript or as a project dependency. Configure the server in your Claude Desktop config file (claude_desktop_config.json) using either npx -y mcp-server-fetch-typescript or a local build after cloning the repository. Debug with the MCP inspector.

Key features of mcp-server-fetch-typescript

get_raw_text – retrieve raw text from URLs without browser rendering
get_rendered_html – fetch fully rendered HTML with Playwright headless browser
get_markdown – convert web content to well-formatted Markdown
get_markdown_summary – extract main content as clean Markdown, removing navigation

Use cases of mcp-server-fetch-typescript

Extracting raw data from JSON, XML, CSV, or plain text files
Scraping content from modern JavaScript-rendered single-page applications
Archiving web articles and blog posts as Markdown
Generating clean summaries of web page main content

FAQ from mcp-server-fetch-typescript

What tools does the server provide?

It provides four tools: get_raw_text, get_rendered_html, get_markdown, and get_markdown_summary – each handling different web content retrieval and conversion needs.

Does this server require any external dependencies to run?

Yes, get_rendered_html uses Playwright for headless browser rendering; the README does not list explicit runtime requirements for Playwright but notes it is used.

How do I install the server?

You can install it globally with npm install -g mcp-server-fetch-typescript or as a project dependency with npm install mcp-server-fetch-typescript.

Where do I configure it for use with Claude Desktop?

On MacOS add config to ~/Library/Application Support/Claude/claude_desktop_config.json; on Windows to %APPDATA%/Claude/claude_desktop_config.json.