OpenAI recently launched its new feature, the Agent Kit, a complete set of tools for developers and enterprises to build, deploy, and optimize agents. Building agents required juggling between different tools, custom connectors, and manual eval pipelines, prompt tuning, and weeks of frontend work. But with AGent Kit, developers will now be able to design better workflows visually and embed if they want Agentic AI’s faster workflow.
Agent Builder – a visual canvas for creating and versioning multi-agent workflows.
Connector Registry- a central hub for admins to manage how data and tools are connected across OpenAI products.
Chat Kit – A toolkit for embedding customizable chat-based agent experiences in your Product.
These will not be the only add-ons as the suite will also be updated with new features like datasheets, trace grading, automated prompt optimization, and third model support to measure and improve agent performance.
Design Workflows with Agent Builder
Agent Builder will provide the developers with clearer visibility into how they work. Agent builder provides a visual canvas for composing logic with drag and drop nodes, connecting tools, and configuring custom guard rails. The new software supports preview runs, inline eval configurations, and full versioning – ideal for fat iteration.
Beyond that, OpenAI has also launched a connector registry for enterprises to govern and maintain data across multiple workspaces and organisations. The connector registry consolidates data sources into a single admin panel across Chat GPT and the API. The registry includes all pre-built connectors like Dropbox, Google Drive, SharePoint, and Microsoft Teams, as well as third-party MCP’s
Developers will also be able to enable Guardrails in Agent Builder – an open-source, Modular safety layer that will help protect agents against unintended or malicious behavior. Guardrails can mask or flag PII, detect jailbreaks, and apply other safeguards, making it easier to build and deploy reliable, safe agents. Guardrails can be deployed standalone or via the guardrails library for Python and JavaScript.
Embed agentic chat experiences with the Chat Kit
Deploying Chat Ui’s can be surprisingly complex, handling streaming responses, managing threads, could all be very complex. With new features of the chat Kit, users will be able to embed chat-based agents that feel native to your product. These can be embedded into applications or websites and customized to match your theme or brand.
Pricing and availability
Chat Kit and the new Evals capability are generally available to all developers. Agent Builder is available in beta, and the connector registry is beginning its beta rollout to some api, chat, Enterprise, and Edu customers with a global admin. The global admin console is a prerequisite to enable the connector registry; all of these are included with standard Apin model pricing.
