Step-by-Step ADK

This repository contains sample code that will allow you to learn the features of Google ADK and how to use it.

Learning Objectives

Through the course of the steps in this repository, you will learn:

1. 01_basic_agents: How to create a basic agent using ADK.
2. How ADK is different from the GenAI SDK (Upcoming)
3. How to define and call tools using ADK (Covered progressively, starting with exercises in 01_basic_agents)
4. How to use advanced ADK features such as session, state, and memory (Upcoming)
5. How to deploy an ADK app to a cloud runtime (Upcoming)
6. Agent workflows (Upcoming)
7. Multi-agent interactions (Upcoming)

Getting started

This repository is structured into sequential steps, each in its own directory (e.g., 01_basic_agents, 02_another_step, etc.).

To get started with this repository:

1. Clone the repository in your IDE.

2. Set up a Python virtual environment (Note: if you don't have uv installed, install it using pip install uv - I highly recommend this tool):

   uv venv
   source .venv/bin/activate

   or

   python3 -m venv .venv
   source .venv/bin/activate

3. Install dependencies using uv:

   uv sync

   or

   pip install -r requirements.txt

4. Go to the step you'd like to execute

   cd 01_basic_agents

   Each step directory contains its own README.md with specific instructions and explanations for that agent.

5. Run the ADK server

   adk web

   This will start a local server (usually at http://localhost:8000). Open this URL in your web browser to interact with the agent for that step.

Repository Structure

• /01_basic_agents: Learn the fundamentals of creating and running a simple ADK agent. The README.md in this directory includes exercises for adding tools.
• (More steps will be added here as they are developed)

Follow the README.md file within each step's directory for detailed guidance.

Creating your own agent

Any ADK agent needs to be create using the specified folder structure below.

parent_folder/
    multi_tool_agent/
        __init__.py
        agent.py
        .env

Then run the below command within a folder where you would like to create the agent to automatically bootstrap the required files.

mkdir my_agent && cd my_agent && touch __init__.py agent.py && echo "from . import agent" >> __init__.py && touch .env && echo "GOOGLE_GENAI_USE_VERTEXAI=FALSE" >> .env && echo "GOOGLE_API_KEY=<paste your key here>" >> .env

The API Key for Gemini can be obtained from Google AI Studio

I recommend that you export the API key to the environment. Either use the below command or include it in your bash or zsh env files.

  export GOOGLE_API_KEY=<paste your key here>

Contributing

Contributions are welcome! Please feel free to submit pull requests or open issues.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.