Skip to content
Agent Development Kit Logo

Agent Development Kit (ADK) for TypeScript

An open-source AI agent framework integrated with Gemini and Google, built for TypeScript/Node.js

⚠️ This is a Work-in-Progress TypeScript port of the original Python ADK. Features and stability may vary.

What is ADK TypeScript?

Agent Development Kit (ADK) for TypeScript is a flexible and modular framework for developing and deploying AI agents using TypeScript. ADK can be used with popular LLMs like Google's Gemini and is designed with a focus on tight integration with the Google ecosystem. ADK TypeScript makes it easy to get started with simple agents powered by Gemini models and Google tools, while providing the control and structure needed for more complex agent architectures and orchestration within the Node.js environment.

Get started:
npm install -g adk-typescript
(This installs the CLI tool for creating and running agents)

Quickstart Tutorial Sample Agents (Python) Contribute ❤️

Learn more

  • Flexible Orchestration

    Define workflows using workflow agents (SequentialAgent, ParallelAgent, LoopAgent) for predictable pipelines, or leverage LLM-driven dynamic routing (LlmAgent with AutoFlow) for adaptive behavior.

    Learn about agents

  • Multi-Agent Architecture

    Build modular and scalable applications by composing multiple specialized agents (LlmAgent, BaseAgent) in a hierarchy using subAgents. Enable complex coordination and delegation.

    Explore multi-agent systems

  • Rich Tool Ecosystem

    Equip agents with diverse capabilities: use pre-built tools (googleSearch, codeExecutionTool), create custom functions (FunctionTool), integrate 3rd-party libraries (LangChain via LangchainTool, CrewAI via CrewaiTool), or use other agents as tools (AgentTool).

    Browse tools

  • Deployment Ready

    Containerize and deploy your TypeScript agents anywhere – run locally, scale with Vertex AI Agent Engine (check compatibility), or integrate into custom infrastructure using Cloud Run or Docker.

    Deploy agents

  • Built-in Evaluation

    Systematically assess agent performance using the evaluation module (AgentEvaluator) by evaluating both final responses and execution trajectories against test datasets (.test.json).

    Evaluate agents

Preview

This feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. Pre-GA features are available "as is" and might have limited support. For more information, see the launch stage descriptions.