DuckDuckGo Search with MCP Agent

What is DuckDuckGo Search with MCP Agent?

DuckDuckGo Search with MCP Agent is a project that demonstrates how to use the DuckDuckGo MCP Server with a LangChain Groq LLM agent to perform intelligent search tasks via MCP (Micro Component Protocol). It is intended for developers building search-driven applications with agentic reasoning.

How to use DuckDuckGo Search with MCP Agent?

Clone the repository, install dependencies with pip install -r requirements.txt, set your GROQ_API_KEY in a .env file, install the MCP server via uvx -y duckduckgo-mcp-server, and run python main.py. The script starts an MCP client, connects to the DuckDuckGo MCP server, and uses the Groq LLM to perform a smart search.

Key features of DuckDuckGo Search with MCP Agent

• DuckDuckGo MCP Server integration
• Groq LLM (deepseek-r1-distill-llama-70b) for reasoning
• Async Python execution
• Simple and modular code structure

Use cases of DuckDuckGo Search with MCP Agent

• Performing intelligent web searches with an LLM agent
• Demonstrating MCP client-server integration with LangChain
• Building modular search pipelines that reason over results

FAQ from DuckDuckGo Search with MCP Agent

What dependencies are required?

The project requires libraries such as langchain_groq and python-dotenv, listed in requirements.txt. You also need uvx installed to run the DuckDuckGo MCP server.

What LLM model does the agent use?

The agent uses Groq's deepseek-r1-distill-llama-70b model for reasoning.

How do I get started quickly?

Set up your .env file with a valid GROQ_API_KEY, install all Python dependencies, and run python main.py. The script includes an example that searches for "the best restaurant in San Francisco."

What transport protocol does the server use?

The server uses MCP (Micro Component Protocol) over stdio, launched via the uvx command.

Is there a license for this project?

Yes, this project is licensed under the MIT License.