Greeter

Quick Example for adding an MCP to Gemini

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": {
    "jrmlhermitte-gemini-mcp-example": {
      "command": "python",
      "args": [
        "gemini-mcp-example/main.py"
      ]
    }
  }
}

You can run a local MCP server in Python and expose a simple tool that responds to a greeting. This setup uses the standard stdio transport, letting a client send requests and receive responses through the command’s standard input and output streams. It’s a lightweight way to prototype an MCP-enabled service and validate integration with clients like Gemini CLI.

How to use

You use an MCP client to connect to your local MCP server and discover its available tools. Once connected, you can invoke a tool by name and provide the required arguments. The server will respond with structured results or error information. This flow is ideal for quick testing, experimentation, and validating client integrations before deploying a production endpoint.

How to install

Prerequisites you need to have before running the server:

Install UV to bootstrap projects
Install Python and set up a virtual environment
Clone the example project repository

Install UV (Linux and macOS):

curl -LsSf https://astral.sh/uv/install.sh | sh

Clone the example project, change into the directory, and synchronize dependencies. Then activate the virtual environment.

git clone https://github.com/jrmlhermitte/gemini-mcp-example.git 
cd gemini-mcp-example
uv sync
source .venv/bin/activate

Start the MCP server using the local Python file included in the project. This runs the server in stdio mode, so it can communicate with a client via standard input and output streams.

python gemini-mcp-example/main.py

Once the server is running, you initialize the protocol version and interact with it using the client. The client will negotiate capabilities and then you can list and call tools exposed by the server. When you are done testing, you can terminate the process. The setup uses the stdio transport by default.

Additional notes and troubleshooting

If you run into issues starting the server, ensure you activated the virtual environment and that you are executing the Python script from the project root.

Gemini CLI integration is demonstrated by mapping the MCP server to a local extension. You can start Gemini from the project folder and use the extension to locate and interact with the MCP server.