ddg_search
A powerful Model Context Protocol (MCP) server for web search and URL content extraction using DuckDuckGo.
Documentation
Quick Start
# Search from the command line
npx -y @oevortex/ddg_search@latest "your search query"
# Or install globally
npm install -g @oevortex/ddg_search
ddg "your search query"Usage
CLI
# Basic web search (DuckDuckGo)
ddg "quantum computing"
# Get more results
ddg "latest news" -n 10
# AI-generated answer (IAsk)
ddg "explain recursion" -m ai
# AI-generated answer (Monica)
ddg "hello world" -m ai -b monica
# IAsk with options
ddg "gravity" -m ai --iask-mode academic --detail-level comprehensiveFlags
| Flag | Description | Default |
|---|---|---|
| `-m, --mode` | `web` or `ai` | `web` |
| `-b, --backend` | `iask` or `monica` (AI mode only) | `iask` |
| `-n, --num-results` | Number of results (1-20, web mode) | `3` |
| `--iask-mode` | `question`, `academic`, `forums`, `wiki`, `thinking` | `thinking` |
| `--detail-level` | `concise`, `detailed`, `comprehensive` | — |
| `-h, --help` | Show help | |
| `-v, --version` | Show version |
MCP Server
Start the MCP server for use with Claude Desktop, OpenCode, or other MCP clients:
ddg --serverClaude Desktop config:
{
"mcpServers": {
"ddg-search": {
"command": "npx",
"args": ["-y", "@oevortex/ddg_search@latest"]
}
}
}Global install config:
{
"mcpServers": {
"ddg-search": {
"command": "ddg"
}
}
}Programmatic API
import { searchDuckDuckGo } from '@oevortex/ddg_search/src/utils/search.js';
import { searchIAsk } from '@oevortex/ddg_search/src/utils/search_iask.js';
import { searchMonica } from '@oevortex/ddg_search/src/utils/search_monica.js';
// DuckDuckGo
const results = await searchDuckDuckGo("query", 5);
// IAsk AI
const answer = await searchIAsk("query", "thinking", null);
// Monica AI
const answer = await searchMonica("query");MCP Tool Schema
Single unified tool `web-search`:
| Parameter | Type | Default | Description | |
|---|---|---|---|---|
| `query` | string | required | Search query | |
| `mode` | `"web"` \ | `"ai"` | `"web"` | DuckDuckGo results or AI answer |
| `backend` | `"iask"` \ | `"monica"` | `"iask"` | AI backend (mode=ai only) |
| `numResults` | integer | `3` | Results count (web mode, 1-20) | |
| `iaskMode` | string | `"thinking"` | IAsk mode (backend=iask only) | |
| `detailLevel` | string | — | IAsk detail level (backend=iask only) |
Agent Prompt
Add this to your agent's system prompt to enable web search:
When you need to search the web or find current information, use the DDG Search CLI:
npx -y @oevortex/ddg_search@latest "" -m ai
This runs an AI-powered web search via IAsk. For standard DuckDuckGo results (titles + URLs + snippets), use mode "web" instead:
npx -y @oevortex/ddg_search@latest "" -m web -n 5
For a different AI backend (Monica), add -b monica. Use this for any research, fact-checking, or current events task.Proxy Support
Respects `HTTP_PROXY`, `HTTPS_PROXY`, and `NO_PROXY` environment variables automatically.
HTTPS_PROXY=http://proxy:8080 ddg "search query"Install
npm install -g @oevortex/ddg_searchOr use `npx` without installing.
Project Structure
bin/cli.js CLI entry point
src/
index.ts MCP server
tools/
searchTool.js Unified search tool
utils/
search.js DuckDuckGo scraper
search_iask.js IAsk AI client
search_monica.js Monica AI client
user_agents.js User agent rotation
skills/
ddg-search/ Agent skill for marketplaceContributing
Pull requests welcome. Open an issue for bugs or feature requests.
License
Apache 2.0
Frequently asked questions
What is ddg_search?
ddg_search is A powerful Model Context Protocol (MCP) server for web search and URL content extraction using DuckDuckGo.
How do I install ddg_search?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is ddg_search open source?
Yes — it is hosted on GitHub at https://github.com/oevortex/ddg_search and has 16 stars.
Related MCP tools
This sample app uses the Vercel MCP Adapter that allows you to drop in an MCP server on a group of routes in any Next.js project.
The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder, MCP compatibility, and more.
An AI-powered task-management system you can drop into Cursor, Lovable, Windsurf, Roo, and others. Built for the Model Context Protocol to enhance AI capabiliti
Fay是一个帮助数字人(2.5d、3d、移动、pc、网页)或大语言模型(openai兼容、deepseek)连通业务系统的mcp框架。 JavaScript-based implementation. Trusted by 12000+ developers.
CTTF: MCP integration between Cursor and Figma, allowing Cursor Agentic AI to communicate with Figma for reading designs and modifying them programmatically.
🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients. JavaScript-based implementation.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP