GoogleBuilding agents with the ADK and the new Interactions API
Read Full ArticleSummary
The article introduces the new Interactions API as a significant advancement in AI development, transitioning from stateless to stateful workflows. It outlines how developers can leverage this API alongside the Agent Development Kit (ADK) to enhance agent capabilities. The API serves as a unified interface for both raw models and managed agents, enabling improved state management, background execution, and native thought handling. Two primary integration patterns are discussed: powering ADK agents with the Interactions API and using the API as a remote agent in an existing Agent2Agent (A2A) ecosystem. Code examples illustrate how to implement these patterns effectively.
Key Learnings
- 1The Interactions API provides a unified interface for managing AI agents, allowing for improved state management and background execution.
- 2Developers can enhance their ADK agents by utilizing the Interactions API for complex state management and reasoning chains.
- 3The API enables seamless integration with existing A2A protocols, allowing developers to incorporate new capabilities without extensive refactoring.
- 4By treating the API as both an inference engine and a remote agent, developers can expand their agentic systems with minimal friction.
Who Should Read This
Senior AI Developers implementing stateful workflows in AI agents using the ADK and Interactions API
Test Your Knowledge
What are the trade-offs of using the Interactions API compared to the previous generateContent inference API endpoint?
How does the Interactions API improve state management for agents built with the ADK?
In what scenarios would background execution be beneficial for agent workflows?
What design decisions are necessary when integrating the Interactions API with existing A2A clients?
How does the InteractionsApiTransport facilitate communication between A2A clients and the Interactions API?
Topics
More articles about Agent Development Kit
Explore Agent Development Kit engineering →Real-World Agent Examples with Gemini 3
The article explores the capabilities of Gemini 3 as a core orchestrator for building complex AI agents capable of handling real-world tasks. It highlights various open-source frameworks and tools...
Introducing Agent Development Kit for TypeScript: Build AI Agents with the Power of a Code-First Approach
The Agent Development Kit (ADK) for TypeScript is an innovative framework designed to facilitate the development of intelligent, autonomous multi-agent systems. By adopting a code-first approach, ADK...
Agent Garden - Samples for learning, discovering and building
The article introduces Agent Garden, a platform designed to facilitate the creation and deployment of AI agents. It highlights the challenges developers face in designing sophisticated multi-agent...
Delight users by combining ADK Agents with Fancy Frontends using AG-UI
The article presents a comprehensive overview of integrating the Agent Development Kit (ADK) with AG-UI to create interactive AI applications. It highlights the capabilities of ADK in building...
Agent Garden - Samples for learning, discovering and building
The Agent Garden is a platform designed to simplify the development and deployment of AI agents, particularly multi-agent systems. It offers a repository of curated samples and tools that assist...
More from Google Engineering
View Google engineering blogs →Introducing Finish Changes and Outlines, now available in Gemini Code Assist extensions on IntelliJ and VS Code
The article introduces two new features in the Gemini Code Assist extensions for IntelliJ and Visual Studio Code: Finish Changes and Outlines. Finish Changes acts as an AI pair programmer, allowing...
Unleash Your Development Superpowers: Refining the Core Coding Experience
The article outlines recent feature enhancements in the Gemini Code Assist tool, designed to streamline the coding experience for developers. Key features include Agent Mode with Auto Approve for...
Introducing Wednesday Build Hour
The 'Wednesday Build Hour' is a weekly initiative designed for developers to engage in hands-on learning and skill enhancement in cloud technologies. Led by Google Cloud experts, the sessions cover a...
What's new in TensorFlow 2.21
TensorFlow 2.21 introduces significant enhancements, particularly with the LiteRT stack, which is designed for high-performance on-device inference. This new runtime offers improved GPU performance,...
You can't stream the energy: A developer's guide to Google Cloud Next '26 in Vegas
The article serves as a guide for developers attending Google Cloud Next '26 in Las Vegas, highlighting the importance of in-person collaboration and the value of hands-on learning. It outlines key...