What is Adk Golang

adk-golang is an open-source Go toolkit designed for building and deploying AI agents with flexibility and control, allowing developers to define agent behavior and orchestration directly in code.

Use cases

Use cases for adk-golang include building modular AI applications, creating adaptive workflows with LLM-driven routing, developing real-time interactive experiences, and deploying scalable AI solutions across different platforms.

How to use

To use adk-golang, developers can start by defining agents, tools, and orchestration logic in their Go code. They can leverage the CLI and visual web UI for local development, testing, and debugging before deploying their agents to various environments.

Key features

Key features include code-first development, multi-agent architecture, a rich tool ecosystem, flexible orchestration, integrated developer experience, built-in evaluation, deployment readiness, native streaming support, and state management.

Where to use

adk-golang can be used in various fields such as cloud computing, AI development, and any application requiring sophisticated AI agents that can be integrated with Google Cloud services.

Agent Development Kit (ADK)

An open-source, code-first Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

Important Links: Docs & Samples

The Agent Development Kit (ADK) is designed for developers seeking fine-grained control and 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.

✨ Key Features

• Code-First Development: Define agents, tools, and orchestration logic for maximum control, testability, and versioning.
• Multi-Agent Architecture: Build modular and scalable applications by composing multiple specialized agents in flexible hierarchies.
• Rich Tool Ecosystem: Equip agents with diverse capabilities using pre-built tools, custom Go functions, API specifications, or integrating existing tools.
• Flexible Orchestration: Define workflows using built-in agents for predictable pipelines, or leverage LLM-driven dynamic routing for adaptive behavior.
• Integrated Developer Experience: Develop, test, and debug locally with a CLI and visual web UI.
• Built-in Evaluation: Measure agent performance by evaluating response quality and step-by-step execution trajectory.
• Deployment Ready: Containerize and deploy your agents anywhere – scale with Vertex AI Agent Engine, Cloud Run, or Docker.
• Native Streaming Support: Build real-time, interactive experiences with native support for bidirectional streaming (text and audio).
• State, Memory & Artifacts: Manage short-term conversational context, configure long-term memory, and handle file uploads/downloads.
• Extensibility: Customize agent behavior deeply with callbacks and easily integrate third-party tools and services.

🚀 Installation

You can install the ADK CLI using Go:

go install github.com/nvcnvn/adk-golang/cmd/adk@latest

Or download a pre-built binary from the releases page.

🔑 Setup API Key

Follow this guide to get and setup your key.

🏁 Getting Started

Create your first agent (my_agent/agent.go):

// my_agent/agent.go
package main

import (
	"github.com/nvcnvn/adk-golang/pkg/agents"
	"github.com/nvcnvn/adk-golang/pkg/tools"
)

func main() {
	// Define your agent here
	rootAgent := agents.NewAgent(
		agents.WithName("search_assistant"),
		agents.WithModel("gemini-2.0-flash-exp"), // Or your preferred Gemini model
		agents.WithInstruction("You are a helpful assistant. Answer user questions using Google Search when needed."),
		agents.WithDescription("An assistant that can search the web."),
		agents.WithTools(tools.GoogleSearch),
	)

	// Export the agent for the CLI to use
	agents.Export(rootAgent)
}

Run it via the CLI:

adk run my_agent

Or launch the Web UI:

adk web

For a full step-by-step guide, check out the quickstart or sample agents.

📚 Resources

Explore the full documentation for detailed guides on building, evaluating, and deploying agents:

• Get Started
• Browse Sample Agents
• Evaluate Agents
• Deploy Agents
• API Reference

🤝 Contributing

We welcome contributions from the community! Whether it’s bug reports, feature requests, documentation improvements, or code contributions, please see our Contributing Guidelines to get started.

📄 License

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

Happy Agent Building!