
Build type-safe, production-ready AI agents in TypeScript with ADK-TS, supporting multiple LLMs, built-in tools, session management, and agent orchestration.
ADK-TS is an open-source framework for building production-ready AI agents in TypeScript. It provides a strongly typed foundation for composing, orchestrating, and deploying agents that interact with multiple large language models (LLMs) and external tools. The framework is designed to help teams move from prototypes to robust, maintainable AI systems with predictable behavior and clear boundaries.
At its core, ADK-TS offers type-safe abstractions for defining agents, tools, sessions, and workflows, ensuring compile-time validation of inputs, outputs, and agent interactions. It supports multi-LLM configurations, allowing developers to route tasks to different models based on cost, capability, or latency requirements. Built-in tools and adapters simplify common tasks such as retrieval, function calling, and structured output handling, while session management enables conversational context, state persistence, and user-specific workflows. Agent orchestration features allow you to coordinate multiple agents, define control flows, and implement complex, multi-step reasoning or task execution pipelines.
Please sign in to comment
💬 No comments yet
Be the first to share your thoughts!
Explore 1000+ top alternatives to ADK-TS
MindStudio is a platform for visually designing, configuring, and deploying custom AI applications and workflows using large language models and integrated data sources.

Kavout is an AI-driven investment research platform that analyzes and ranks thousands of stocks, ETFs, and cryptocurrencies, offering natural language queries, institutional activity tracking, and actionable trading signals.

MiniMax Agent provides an AI-powered workspace that assists with coding, data analysis, content creation, podcast generation, and guided meditation to support everyday creative and productive tasks.

Neuraldeep is an AI platform that converts speech and written ideas into 3D designs, supports LLM fine-tuning, and enables bio-upcycled 3D printing applications.