What is AgentKit used for?

AgentKit is used to build, deploy, and test AI agents that handle multi-step tasks like customer support, research, and workflow automation. It includes visual design tools, embeddable chat interfaces, evaluation frameworks, and connector management.

Is AgentKit free to use?

There's no separate fee. It's included with OpenAI's API pricing. You pay for model token usage, tool calls, and data storage. You can design workflows in Agent Builder at no cost until you run them.

Can AgentKit work with non-OpenAI models?

Not currently. AgentKit is built exclusively for OpenAI's ecosystem. If model flexibility matters, consider alternative frameworks or a custom integration approach.

What's the difference between AgentKit and the Agents SDK?

The Agents SDK is the open-source library for coding agents in Python, TypeScript, and Go. AgentKit adds a visual builder, ChatKit for deployment, Evals for testing, and Connector Registry for governance on top of that foundation.

How does AgentKit compare to no-code platforms like n8n?

AgentKit is purpose-built for AI agent logic including branching, reasoning, guardrails, and evaluation. Tools like n8n are better for broad SaaS automation, and many teams use both together.