Gemini

Provides an MCP endpoint to interact with Gemini via the CLI, including sending messages, generating content, and listing models.

javascript

5

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 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": {
    "blesscat-gemini-cli-mcp": {
      "command": "node",
      "args": [
        "/path/to/gemini-mcp/index.js"
      ]
    }
  }
}

Gemini MCP Server enables you to run a local MCP endpoint that integrates with the Gemini CLI. It lets clients send messages, generate content, and list models through a simple, self-contained Node.js server that can be wired into MCP clients like Claude Desktop or other tooling.

How to use

Use the Gemini MCP Server with an MCP client to perform common Gemini operations from your tooling. Start the server locally and point your MCP client at the provided stdio entry to communicate through Node.js. The server exposes commands to send messages, generate content, and list Gemini models from your existing Gemini CLI setup.

How to install

Prerequisites include Node.js 18 or newer and the Gemini CLI tool installed and configured on your system.

Install dependencies for the Gemini MCP Server.

npm install

Configuration and usage notes

The MCP server assumes the Gemini CLI is available in your system PATH. Ensure you have the Gemini CLI tool installed, configured with your API credentials, and working by running gemini --help.

Integration with Claude Desktop can be configured with the following MCP server entry.

{
  "mcpServers": {
    "gemini": {
      "command": "node",
      "args": ["/path/to/gemini-mcp/index.js"]
    }
  }
}

Development

To develop and run the server with live file watching, use the development script.

npm run dev

Available tools

gemini_chat

Send a message to Gemini and get a response. Parameters: message (required), model (optional, default: gemini-pro)

gemini_generate

Generate content with Gemini using a specific prompt. Parameters: prompt (required), model (optional), temperature (optional, 0.0-1.0)

gemini_list_models

List available Gemini models. No parameters required