A modern, visually rich, GitHub‑ready README to help you set up and use the Python ADK for building Gemini-powered AI agents.
- ⚡ Quickstart-friendly
- 🎨 Visual diagrams & icons
- 🤖 Agent creation made simple
- 🌐 Optional Web UI
- 📦 Clean folder structure overview
- Prerequisites
- Create Virtual Environment
- Install ADK
- Create Your First Agent
- Run Your Agent
- Web UI
- Project Structure
- Troubleshooting
- Learn More
| Requirement | Version |
|---|---|
| 🐍 Python | 3.10+ |
| 🌐 Internet | Required |
| 🔑 Google API Key | From Google AI Studio |
👉 Get API Key: https://aistudio.google.com/apikey
python -m venv <env_name><env_name>/Scripts/activateSet-ExecutionPolicy RemoteSigned -Scope CurrentUser -ForceThen activate again.
pip install google-adkadk create <agent_name>Choose a model for the root agent:
1. gemini-2.5-flash
2. Other models
Choose model (1, 2): 1
Enter Google API key: <your_api_key>
🎉 Your Agent Is Ready!
adk run <agent_name>🎉 Your Agent Is Ready!
adk web --port 8000Open browser:
📦 my_agent
│
├── 🧠 agent.py → Your agent logic
├── 🔐 .env → Contains GOOGLE_API_KEY
└── 📄 __init__.py → Package marker
| Issue | Fix |
|---|---|
| 🔴 Script blocked in PowerShell | Set-ExecutionPolicy RemoteSigned -Scope CurrentUser -Force |
| 🔴 API key not found | Ensure .env contains GOOGLE_API_KEY=your_key |
| 🔴 Module not found | Activate virtual environment |
| 🔴 Cannot install google-adk | pip install --upgrade pip |
📖 Full docs & examples: 👉 https://google.github.io/adk-docs/#learn-more