- Home
- MCP servers
- DuckDuckGo
DuckDuckGo
- python
0
GitHub Stars
python
Language
7 months ago
First Indexed
3 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"varlabz-duckduckgo-mcp": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/varlabz/duckduckgo-mcp",
"duckduckgo-mcp"
]
}
}
}You can use the DuckDuckGo MCP server to access structured search across text, images, videos, and news, plus prompts and resources, all through MCP clients or the command line. This setup lets you run searches from your preferred client or terminal and receive structured results that you can incorporate into your workflows.
How to use
Set up the MCP server in your chosen client or run it directly from the terminal to perform DuckDuckGo searches. You will access a dedicated search tool that supports different content types and region settings, use prompts to analyze results, and consult the regional resources to discover supported region codes. Use the provided configuration blocks to connect your MCP client to the server, then issue search queries and receive structured results suitable for automation and further processing.
How to install
Prerequisites: you need a runtime capable of running MCP clients, typically Node-based tooling. Ensure you have a tool to load MCP servers locally or remotely, such as uvx, and access to a terminal.
- Prepare your environment for running MCP servers and clients. Install the required local tooling if you have not yet: install uvx and any dependencies your platform requires. 2. Run the MCP server ad‑hoc from the terminal to verify it starts. 3. Integrate with your MCP client by adding the provided configuration blocks to your client’s MCP configuration so the client can communicate with the DuckDuckGo MCP server.
Configuration and usage notes
# Quick sanity check to start the MCP server via uvx (ad-hoc)
uvx --from git+https://github.com/varlabz/duckduckgo-mcp duckduckgo-mcp
# Claude Desktop configuration (example)
{
"mcpServers": {
"duckduckgo": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/varlabz/duckduckgo-mcp",
"duckduckgo-mcp"
]
}
}
}
# VS Code / Workspace configuration (example)
{
"servers": {
"duckduckgo": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/varlabz/duckduckgo-mcp",
"duckduckgo-mcp"
]
}
}
}
Examples and usage tips
- Run a search from the terminal and get JSON output for scripting by using the CLI wrapper: use the ad-hoc start command above, then query results via the MCP client tooling you use. - Explore the region resources to discover supported region codes and tailor your searches to local results.
MCP capabilities and tools
The DuckDuckGo MCP server provides several capabilities you can use through MCP clients or the CLI. You can perform a DuckDuckGo search across text, images, videos, or news with configurable limits and regions. You can also access region resources and use prompts to analyze results or plan research.
Developer notes and troubleshooting
If you encounter issues starting or connecting to the MCP server, verify that your client is configured with the correct command and arguments and that the MCP runtime you use (uvx) is accessible in your environment. Ensure you are using the exact command strings shown in the configuration blocks and that any required environment setup is completed before starting the server.
Available tools
search
DuckDuckGo search across text, images, videos, or news with configurable max results, region, safesearch, and timelimit.
regions
Resources: duckduckgo://regions JSON that lists supported region codes and names.
search_assistant
Prompt tool that generates an analysis prompt to examine search results.
research_planner
Prompt tool that creates a structured plan for researching a topic with adjustable depth.