ddg_search
A powerful Model Context Protocol (MCP) server for web search and URL content extraction using DuckDuckGo.
claude mcp add --transport stdio oevortex-ddg_search npx -y @oevortex/ddg_search@latest
How to use
This MCP server provides fast, privacy-friendly web search capabilities by wrapping DuckDuckGo-based web search together with AI-assisted answer tools. It exposes multiple search tools that can be invoked via MCP clients: a Web Search Tool for standard web queries, an IAsk AI Search Tool for AI-assisted responses, and a Monica AI Search Tool for quick queries. The server is designed to be MCP-compliant, so you can add it to your MCP client configuration and have AI assistants issue structured tool calls to retrieve web results and AI-generated answers without exposing API keys. Typical usage involves starting the server via npx or from an installed global CLI, then configuring your MCP client to route queries to the ddg-search MCP server using the provided command and arguments. You can expect real web scraping results (not limited API data) with built-in rate limiting and rotating user agents for security and reliability.
How to install
Prerequisites:
- Node.js and npm (or a Node-enabled environment capable of running npx commands)
- Internet access to download packages
Install and run using one of these options:
Option A - Run with npx (no installation required):
npx -y @oevortex/ddg_search@latest
This downloads and runs the MCP server on demand.
Option B - Global installation (recommended for repeated use):
npm install -g @oevortex/ddg_search
Run globally:
ddg-search-mcp
Option C - Local development (clone and run from source):
git clone https://github.com/OEvortex/ddg_search.git
cd ddg_search
npm install
npm start
Or with Yarn:
yarn install
yarn start
Or with pnpm:
pnpm install
pnpm start
Additional notes
Tips and considerations:
- No API keys are required to use this MCP server; it relies on web scraping via DuckDuckGo HTML and AI tools (IAsk AI, Monica, Brave AI).
- The server implements rate limiting and rotates user agents to improve privacy and reduce blocking.
- To integrate with an MCP client, use the provided mcpServers configuration (command: npx, args: [-y, @oevortex/ddg_search@latest]). If you have the package installed globally, you can reference the CLI (ddg-search-mcp) instead.
- If you plan to run locally for development, you can start with npm start and then attach your MCP client to the running instance.
- Environment variables are not strictly required for basic operation, but you may wish to configure additional networking or caching settings at runtime if supported by your deployment environment.
Related MCP Servers
zen
Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage
MCP -Deepseek_R1
A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)
mcp-fhir
A Model Context Protocol implementation for FHIR
mcp
Inkdrop Model Context Protocol Server
mcp-appium-gestures
This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..
dubco -npm
The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.