Gemini 2.5 Pro

🚀 Gemini MCP Server

A server implementing the Model Context Protocol (MCP) that enables the Claude desktop version to interact with Google's experimental Gemini 2.5 Pro AI model.

🚀 Quick Start

This server is designed to facilitate the interaction between the Claude desktop version and Google's Gemini 2.5 Pro model by implementing the MCP protocol.

✨ Features

• Full support for the MCP protocol.
• Access to Google's experimental Gemini 2.5 Pro model.
• Secure handling of API keys.
• Integration with Google Search (optional).
• Token usage reporting.
• Implemented in TypeScript.

💻 Usage Examples

Basic Usage

The MCP server provides the following tools:

1. generateWithGemini

This function uses Google's Gemini 2.5 Pro experimental model to generate content.

// Example of using generateWithGemini
const result = await generateWithGemini({
    prompt: "Your prompt here",
    temperature: 0.7,
    maxTokens: 200,
    safeMode: true,
    useSearch: false
});

• Parameters:
  • prompt (string, required): The prompt sent to Gemini.
  • temperature (number, optional): Temperature setting (ranging from 0.0 to 1.0).
  • maxTokens (number, optional): The maximum number of output tokens.
  • safeMode (boolean, optional): Enable safe mode for sensitive topics.
  • useSearch (boolean, optional): Enable the Google Search grounding tool.

2. getModelInfo

This function retrieves information about the currently used Gemini model.

// Example of using getModelInfo
const modelInfo = await getModelInfo();

📚 Documentation

Frequently Asked Questions

1. Connection Issues

• Check if you have a valid API key.

🔧 Technical Details

Security

• API keys are only handled through environment variables.
• No sensitive data is stored or logged.