google/adk-samples

★ 9,198PythonAudience · developerComplexity · 3/5Setup · moderate

Mindmap

mindmap
  root((adk-samples))
    What it does
      Demo AI agents
      Adapt-and-run templates
      Multi-agent examples
    Use cases
      Customer service
      Travel planning
      Financial advising
      Data engineering
    Languages
      Python
      TypeScript
      Go and Java
    Setup
      Install ADK
      Google Cloud credentials

mindmap root((adk-samples)) What it does Demo AI agents Adapt-and-run templates Multi-agent examples Use cases Customer service Travel planning Financial advising Data engineering Languages Python TypeScript Go and Java Setup Install ADK Google Cloud credentials

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Adapt the customer service agent example as a starting point for your own AI-powered support chatbot.

USE CASE 2

Use the travel planning agent template to build a multi-step AI assistant that coordinates searches and recommendations.

USE CASE 3

Study how Google's ADK structures multi-agent coordination by reading the supply chain or data engineering examples.

Tech stack

PythonTypeScriptGoJavaGoogle ADK

Getting it running

Difficulty · moderate Time to first run · 30min

Requires installing the Google Agent Development Kit and setting up Google Cloud credentials for each agent example.

In plain English

This repository is a collection of example AI agents built by Google using their Agent Development Kit, referred to as ADK. An AI agent in this context is a program that uses a large language model to carry out tasks, answer questions, or coordinate with other agents to complete more complex work. These examples are meant to show developers what is possible with the ADK and to give them starting points they can adapt for their own projects. The agents cover a wide range of use cases. On the Python side alone, there are examples for academic research, blog writing, customer service, financial advising, medical pre-authorization, personalized shopping, supply chain management, travel planning, data engineering, and several others. There are also agents written in TypeScript, Go, and Java, though those sections contain fewer examples. Each language's examples live in a separate subfolder with its own setup instructions. To run any of these agents, you first need to install the Agent Development Kit itself, which is a separate Google open-source library available for each supported language. The README links to the installation guide and to the documentation for each language variant. Google notes that this repository is for demonstration purposes only and is not an officially supported product. It is not intended for production use. Developers who find bugs or want to contribute new examples can do so through standard GitHub issue reporting and pull requests.

Copy-paste prompts

Prompt 1

I want to build a customer service AI agent using Google's Agent Development Kit in Python. Using the adk-samples customer service example, how do I adapt it to answer questions about my own product?

Prompt 2

Show me how to run the Python travel planning agent from google/adk-samples locally, what do I install and what commands do I run?

Prompt 3

How does the adk-samples financial advisor agent handle multi-step reasoning? Walk me through its structure so I can adapt it.

Prompt 4

What's the difference between the Python and TypeScript agent examples in adk-samples, when should I choose one over the other?

Open on GitHub → Explain another repo

← google on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.