Weather Ai Agent

🚀 Google Gemini API Integration with MCP Tools Project Description

This project demonstrates how to integrate the Google Gemini API with custom tools managed by the Multi-Cloud Platform (MCP) framework. It leverages the Gemini API to handle natural language queries and uses MCP tools to perform specific operations based on the query intent.

🚀 Quick Start

Prerequisites

Before running this project, ensure the following conditions are met:

• Python 3.7 or higher.
• The Gemini API in your Google Cloud project is enabled, and you have obtained an API key.
• The MCP environment is set up, and the necessary tools are installed.
• Create a .env file in the project root directory and add the following environment variables:

GEMINI_API_KEY=<Your Gemini API Key>
GEMINI_MODEL=<Your Gemini Model Name>
MCP_RUNNER=<Path to the MCP Runner>
MCP_SCRIPT=<Path to the MCP Script>

📦 Installation

1. Clone the repository:

git clone <Repository URL>
cd <Repository Directory>

2. Create a virtual environment (recommended):

python3 -m venv venv
source venv/bin/activate  # On macOS/Linux

3. Install the required dependencies using uv:

uv pip install dotenv google-generativeai mcp
uv add "mcp[cli]" httpx
uv pip install python-dotenv google-generativeai mcp

4. Create a .env file in the project root directory and add your environment variables.

GEMINI_API_KEY=Your API Key
GEMINI_MODEL=gemini-pro
MCP_RUNNER=Path to the MCP Runner
MCP_SCRIPT=Path to the MCP Script

💻 Usage Examples

Basic Usage

To run the application, execute the following command:

python main.py

🔧 Technical Details

How It Works

1. The application loads and verifies the existence of environment variables.
2. It establishes a connection with the MCP client.
3. It retrieves the available tools in the MCP session.
4. It sends a prompt to the Gemini API along with tool definitions.
5. It processes any tool calls from the model.
6. It returns the final response, including the results of tool calls.

Customization Guide

To customize the prompt or behavior:

1. Modify the prompt variable to set the desired text.
2. Adjust the get_contents() function to change the prompt format.
3. Extend the process_response() function to handle different types of responses.