This is a simple MCP server that provides a few different examples of how tools can be defined.

Install virtualenv or conda for python 3. Run:

pip install -r requirements.txt

By default, server.py runs using the stdio transport.

python server.py

Alternatively, you can uncomment the line mcp.run(transport="sse") in server.py and run the above command again to use SSE transport to connect over Http.

If you are running the server with http/sse, use test_client_http.py

If you are not running the server, you can just run test_client_local.py to run the server in stdio transport mode.

The easiest way to test this with Claude Desktop is to use the stdio transport. Find the file claude_desktop_config.json and add a local MCP server. For example:

"mcpServers": {
    "aifarms": {
      "command": "/path/to/virtualenv/mcp-example/venv/bin/python3",
      "args": [
        "/path/to/source/mcp-example/server.py"
      ],
      "env": {
        "PYTHONPATH": "/path/to/source/mcp-example"
      }
    }
  }

Restart Claude Desktop and you should be able to run the following prompt:

What crops are less than $6?

In this example, we use Ollama to demonstrate a simple LLM with tool calling capabilities that talks to our example MCP server. To get started, you will need to download and install Ollama from here:

https://ollama.com/download

You also will need to make sure you have llama3.1 model pulled so Ollama can use the model in our example.

ollama pull llama3.1

Once this is done, you can run the following in a terminal to bring up the assistant:

python test_llm_prompt.py

This will start a connection with Ollama and a prompt to enter questions. Some example questions include:

1. What are the available crops?
2. What are the available crops less than $6?
3. What is the weather in Urbana?

When you are finished, type quit in the terminal to exit.