search-fetch-server MCP Server

Name: search-fetch-server MCP Server
Author: MCP-Mirror

@MCP-Mirror

代码仓库

关于 search-fetch-server MCP Server

Mirror of

#search-fetch-server #mcp #notes-system

基本信息

分类

搜索

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": {
    "search-fetch-server": {
      "command": "node",
      "args": [
        "/path/to/search-fetch-server/build/index.js"
      ]
    }
  }
}

工具

Create new text notes

Fetch content from a URL and optionally convert to markdown using Puppeteer

Perform a DuckDuckGo search

概览

What is search-fetch-server?

A Model Context Protocol server built with TypeScript that provides resources for notes, tools for creating notes, fetching URLs, and performing DuckDuckGo searches, and prompts for summarizing notes. It demonstrates core MCP concepts for developers.

How to use search-fetch-server?

Install dependencies with npm install, build with npm run build, then add the server configuration to your Claude Desktop config file (claude_desktop_config.json) with the command pointing to the built index.js. For development, use npm run watch for auto-rebuild or npm run inspector for debugging via the MCP Inspector.

Key features of search-fetch-server

Resources: Text notes accessible via note:// URIs with metadata
Tool create_note: Creates new text notes with title and content
Tool fetch_url: Fetches URL content, optionally converting to markdown using Puppeteer
Tool duckduckgo_search: Performs DuckDuckGo searches and returns JSON results
Prompt summarize_notes: Generates a structured summary of all stored notes

Use cases of search-fetch-server

Creating and managing text notes through MCP
Fetching web content and optionally converting it to markdown
Searching the web via DuckDuckGo and retrieving results
Summarizing stored notes using an LLM

FAQ from search-fetch-server

What runtime dependencies does search-fetch-server require?

Node.js and npm. Install dependencies with npm install and build with npm run build.

How does the `fetch_url` tool work?

It takes a URL and an optional boolean use_puppeteer. If true, it uses Puppeteer with waitUntil: 'networkidle2' and a 30-second timeout; otherwise it uses axios.