OpenAI has launched AgentKit, a toolkit designed to help developers and enterprises build, deploy, and optimize AI agents.

AgentKit features several core tools, including Agent Builder for workflow creation, Connector Registry for data integration, and ChatKit for customizable chat-based experiences.

The update also introduces evaluation tools such as datasets, trace grading, automated prompt optimization, and third-party model support.

Agent Builder is currently in beta, while Connector Registry is rolling out to API, ChatGPT Enterprise, and Edu users with a Global Admin Console. ChatKit and the new evaluation features are already available.

OpenAI said all AgentKit tools are covered under standard API pricing and that it plans to expand agent deployment options within ChatGPT in future updates.

🔗 Source: OpenAI