A Model Context Protocol (MCP) server that provides web search capabilities using DuckDuckGo. This server enables LLMs to search the web and retrieve detailed content from websites through structured data extraction.

• Fast and reliable. Uses DuckDuckGo's web interface with robust HTML parsing
• LLM-friendly. Returns structured data optimized for AI consumption
• Content extraction. Intelligently extracts and summarizes webpage content
• Related searches. Generates contextual search suggestions

• Python 3.10 or newer
• VS Code, Cursor, Windsurf, Claude Desktop, Goose or any other MCP client

First, install the DuckDuckGo MCP server with your client.

Standard config works in most of the tools:

{
  "mcpServers": {
    "duckduckgo-search": {
      "command": "mcp-duckduckgo"
    }
  }
}

Use the Claude Code CLI to add the DuckDuckGo MCP server:

claude mcp add duckduckgo-search mcp-duckduckgo

For global configuration (available in all projects):

claude mcp add duckduckgo-search --scope user mcp-duckduckgo

Follow the MCP install guide, use the standard config above.

Go to Cursor Settings -> MCP .

Install in Cursor

Go to Cursor Settings -> MCP -> Add new MCP Server. Name to your liking, use command type with the command mcp-duckduckgo.

Install in VS Code

Follow the MCP install guide, use the standard config above.

You can also install the DuckDuckGo MCP server using the VS Code CLI:

code --add-mcp '{"name":"duckduckgo-search","command":"mcp-duckduckgo"}'

After installation, the DuckDuckGo MCP server will be available for use with your GitHub Copilot agent in VS Code.

Follow Windsurf MCP documentation. Use the standard config above.

Go to Advanced settings -> Extensions -> Add custom extension. Name to your liking, use type STDIO, and set the command to mcp-duckduckgo. Click "Add Extension".

Go to Program in the right sidebar -> Install -> Edit mcp.json. Use the standard config above.

DuckDuckGo MCP server supports following arguments:

mcp-duckduckgo --help

Available options:

--port PORT        Port number for the MCP server (default: 3000)
--version          Show program's version number and exit
--help             Show help message and exit

• MCP_PORT: Set the port number for the server (default: 3000)

Example usage:

# Set port via environment variable
export MCP_PORT=8080
mcp-duckduckgo

# Or set it inline
MCP_PORT=8080 mcp-duckduckgo

• Title: Web Search
• Description: Search the web using DuckDuckGo
• Parameters:
  • query (string): Search query (max 400 characters)
  • max_results (number, optional): Maximum number of results to return (1-20, default 10)
• Read-only: false

• Title: Get Page Content
• Description: Retrieve and extract content from a web page
• Parameters:
  • url (string): URL to fetch content from
• Read-only: false

• Title: Suggest Related Searches
• Description: Generate contextual search suggestions based on a query
• Parameters:
  • query (string): Original search query
  • max_suggestions (number, optional): Maximum suggestions to return (1-10, default 5)
• Read-only: true

If you need to install from source or development:

uv is a fast Python package manager:

# Install uv if you haven't already
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install from GitHub
uv tool install git+https://github.com/gianlucamazza/mcp-duckduckgo.git

# Clone and install
git clone https://github.com/gianlucamazza/mcp-duckduckgo.git
cd mcp-duckduckgo
pip install -e .

git clone https://github.com/gianlucamazza/mcp-duckduckgo.git
cd mcp-duckduckgo

# Install in development mode
pip install -e .

# Run tests
pip install -e ".[test]"
pytest

MIT

GitHub Repository