Announcing the Agent Development Kit for Go: Build Powerful AI Agents with Your Favorite Languages

[ad_1]

Today, we are thrilled to add Go to the Agent Development Kit (ADK) family of supported languages! We want to meet you as the builders, where you are so you can build powerful and sophisticated AI agents with the flexibility and control of ADK, using the language that best fits your stack and your use case.

ADK is an open-source, code-first toolkit designed for developers who need fine-grained control over their AI agents.

ADK moves the complexity of LLM orchestration, agent behavior, and tool-use directly into your code. This gives you:

Robust Debugging: Define logic with the same rigor you apply to all your services.
Reliable Versioning: Track changes and deploy with confidence.
Deployment freedom: Take your applications anywhere – from your laptop to the cloud.

What is ADK?

Agent Development Kit (ADK) is designed for developers seeking flexibility when building advanced AI agents that are tightly integrated with services in Google Cloud. It allows you to define agent behavior, orchestration, and tool use directly in code, enabling robust debugging, versioning, and deployment anywhere – from your laptop to the cloud.

Introducing ADK for Go

For Go developers, ADK for Go offers an idiomatic and performant way to build agents. Leverage the power of Go’s concurrency and strong typing to create robust and scalable agentic applications.

As a bonus, ADK Go also has out of the box support for over 30+ databases through MCP Toolbox for Databases – making data integration seamless and simple.

Key Features Across ADK Go

ADK Go shares the same core design principles and features as Python and Java ADK, providing a consistent development experience regardless of your language choice:

Rich Tool Ecosystem: Supercharge your agents with pre-built tools, custom functions, OpenAPI specs, and tight, seamless integration across the Google ecosystem.
Code-First Development: Define agent logic, tools, and orchestration directly in your favorite language for ultimate flexibility, testability, and versioning.
Modular Multi-Agent Systems: Design scalable applications by composing multiple specialized agents into flexible hierarchies.
Development UI: Accelerate your workflow with a built-in development UI that lets you test, evaluate, debug, and showcase your agent(s).

A2A Support for the Go ADK

We are also excited to announce that ADK Go now includes support for the Agent2Agent (A2A) protocol. This enables developers to build powerful multi-agent systems where agents can collaborate to solve complex problems. With A2A, a primary agent can seamlessly orchestrate and delegate tasks to specialized sub-agents – whether they are local services or remote deployments – ensuring secure and opaque interactions without needing to expose internal memory or proprietary logic.

As part of the support of A2A protocol in ADK Go we have also contributed A2A Go SDK to the A2A Project repo. You can dive deeper into the A2A protocol and explore our other SDKs on the A2A Protocol website.