Installation¶

This guide will help you install and set up the Gemini Agent Framework.

Requirements¶

Python 3.8 or higher
pip (Python package installer)
A Google Generative AI API key

Installation Methods¶

Using pip¶

The simplest way to install the framework is using pip:

pip install gemini-agent-framework

From Source¶

If you want to install from source:

Clone the repository:

git clone https://github.com/m7mdony/gemini-agent-framework.git
cd gemini-agent-framework

Install the package:
```
pip install -e .
```

Development Installation¶

For development, install with development dependencies:

pip install -e ".[dev]"

This will install additional tools for development: - pytest for testing - black for code formatting - isort for import sorting - mypy for type checking - flake8 for linting

API Key Setup¶

Get your API key from Google AI Studio
Create a .env file in your project root:
```
GEMINI_API_KEY=your-api-key-here
```

Load the API key in your code:

from dotenv import load_dotenv
load_dotenv()

Verification¶

To verify the installation, run:

from gemini_agent import Agent

# Create an agent instance
agent = Agent(api_key="your-api-key")

# Test the agent
response = agent.prompt("Hello, world!")
print(response)

Troubleshooting¶

Common Issues¶

API Key Not Found
Make sure your .env file exists and contains the correct API key
Check that you're loading the environment variables correctly
Import Error
Verify that the package is installed correctly
Check your Python version (must be 3.8 or higher)
Dependency Issues
Try reinstalling the package
Check for conflicting packages

Getting Help¶

If you encounter any issues: - Check the GitHub Issues - Create a new issue with details about your problem