Python framework for building and orchestrating multi-agent systems with sequential, concurrent, graph-based and hierarchical workflows.
google/adk-go
A code-first Go toolkit for building, evaluating, and deploying AI agents, with multi-agent composition and tool integration.
- Stars
- 8,946
- Forks
- 992
- Open issues
- 279
- Last commit
- 5 hours ago
What is it?
ADK for Go is a modular framework for creating AI agents directly in Go code. It targets workflows ranging from simple single-agent tasks to larger systems where several specialized agents are composed together. Agent logic, tools, and orchestration are defined in Go, so they can be tested and versioned like any other part of a codebase.
The toolkit provides a set of pre-built tools, support for custom functions, and the ability to integrate existing tools. It is optimized for Gemini but is described as model-agnostic and deployment-agnostic. The project is aimed at cloud-native applications, and it includes support for containerizing agents and deploying them to environments such as Google Cloud Run.
The module is imported as google.golang.org/adk/v2 and requires Go 1.26.10 or later. Documentation, samples, and a machine-readable index for AI coding assistants are published alongside the repository.
Key features
- 01Define agents, tools, and orchestration in idiomatic Go code.
- 02Compose multiple specialized agents into modular multi-agent systems.
- 03Use pre-built tools, custom functions, or existing tool integrations.
- 04Containerize agents and deploy them to cloud-native platforms such as Cloud Run.
- 05Model-agnostic design, optimized for Gemini.
Use cases
- Building cloud-native backend services that include LLM-driven agents written in Go.
- Orchestrating several specialized agents to handle a multi-step workflow such as issue triage and report generation.
- Adding agent capabilities with custom tools to an existing Go codebase.
Quick start
- 1
Add ADK Go to your Go project
go get google.golang.org/adk/v2 - 2
Browse the samples in the examples directory to get started
Things to know
- Requires Go 1.26.10 or later.
In collections
Topics
FAQ
- What is adk-go?
- A code-first Go toolkit for building, evaluating, and deploying AI agents, with multi-agent composition and tool integration. ADK for Go is a modular framework for creating AI agents directly in Go code. It targets workflows ranging from simple single-agent tasks to larger systems where several specialized agents are composed together. Agent logic, tools, and orchestration are defined in Go, so they can be tested and versioned like any other part of a codebase.
- Is adk-go free to use? What license does it have?
- adk-go is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is adk-go still maintained?
- Actively maintained. The last commit to adk-go was on Nov 7, 2026.
- What can I use adk-go for?
- Building cloud-native backend services that include LLM-driven agents written in Go. Orchestrating several specialized agents to handle a multi-step workflow such as issue triage and report generation. Adding agent capabilities with custom tools to an existing Go codebase.
- How do I install adk-go?
- Add ADK Go to your Go project: go get google.golang.org/adk/v2 Browse the samples in the examples directory to get started
- Where can I get adk-go?
- The source code and installation instructions are on GitHub: https://github.com/google/adk-go
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source personal AI assistant that indexes your work into a local knowledge graph, with built-in email, notes, browser, meeting notes and shared Spaces.
An SDK for building AI agents and multi-agent systems in Python, .NET and Java, with model, plugin and vector database integrations.
A collection of hands-on Jupyter labs for managing AI models, MCP tools, and agents through Azure API Management's AI Gateway.
Python framework for building LLM applications with multiple cooperating agents, tools, and retrieval-augmented generation.
A Python framework for building autonomous AI agents that run on schedules or react to events, with agents able to join the Fetch.ai network.