OpenAI’s AgentKit launched on October 6, 2025, as a collection of tools for building and deploying AI agents. But there is an important lifecycle change: OpenAI says Agent Builder and Evals will leave its platform on November 30, 2026. The company recommends its Agents SDK for workflows that need to continue as code, and Workspace Agents in ChatGPT for use cases built around natural-language prompting.
What is OpenAI AgentKit?
AgentKit is OpenAI’s toolkit for building, deploying, and improving AI agents. Announced on October 6, 2025, it brought together tools for designing workflows, managing connections, embedding a chat experience, evaluating agent behavior, and fine-tuning models. It built on OpenAI’s Responses API. OpenAI’s announcement now includes a June 3, 2026 lifecycle update, so the original launch description should not be read as a guarantee that every component remains available.
![Claude AI for Beginners Bible: [5 in 1] The Ultimate Guide to Automate Your Work, Save Hours Every Week, and Use AI for Real-World Results](https://m.media-amazon.com/images/I/415+fSJacsL._SL160_.jpg)
What tools did AgentKit include?
Agent Builder: design workflows visually
Agent Builder provided a visual canvas for composing agent logic with drag-and-drop nodes. Developers could connect tools, configure guardrails, preview runs, set up evaluations, and version workflows. Its planned role was to make workflows easier to assemble and iterate on without requiring every part of the orchestration to be written and managed as custom code. OpenAI’s later update says Agent Builder is being wound down.
Connector Registry: manage data and tool connections
Connector Registry was described as a central administrative place to manage how data and tools connect across OpenAI products. The launch announcement named connectors including Dropbox, Google Drive, SharePoint, and Microsoft Teams, as well as third-party MCPs. At launch, OpenAI said the registry was beginning a limited beta rollout for some API, ChatGPT Enterprise, and Edu customers with a Global Admin Console; that is historical availability information, not a statement of current access.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware match#1 Best Overall
ChatKit: embed a chat interface
ChatKit was designed for adding a customizable, chat-based agent experience to a product or website. OpenAI said it was generally available at launch. That launch-era statement does not establish its current availability or terms.
Evals: test and improve agent performance
The launch-era Evals capabilities included datasets, trace grading, automated prompt optimization, and support for evaluating third-party models. OpenAI’s June 2026 update says it is winding down Evals along with Agent Builder.
Reinforcement fine-tuning
OpenAI’s launch post described reinforcement fine-tuning (RFT) as generally available on o4-mini and in private beta for GPT-5 at that time. It also described custom tool calls and custom graders as part of the beta. These are dated launch details, not current model or availability specifications.
Is OpenAI Agent Builder being discontinued?
Yes. In an update dated June 3, 2026, OpenAI said it is winding down Agent Builder and Evals. From November 30, 2026, they will no longer be available on the OpenAI platform. As of October 8, 2026, that end date is still in the future. Check OpenAI’s AgentKit page for any changes to the date or migration guidance.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The update names two alternatives, depending on how a workflow is meant to be built: the Agents SDK for workflows that should continue as code, and Workspace Agents in ChatGPT for use cases better suited to natural-language prompting. OpenAI’s update does not provide a full feature-by-feature migration comparison, so teams should assess their own workflow and requirements rather than assume either option is a direct replacement.
What should you use instead of Agent Builder?
| Need | OpenAI’s stated direction | What to consider |
|---|---|---|
| Continue a workflow as code | Agents SDK | Choose this direction when the workflow is intended to be maintained in code. OpenAI’s update does not specify a one-to-one migration path for Agent Builder projects. |
| Build a workflow using natural-language prompting | Workspace Agents in ChatGPT | Consider this for use cases suited to prompting within ChatGPT. The update does not promise feature parity with Agent Builder or Evals. |
Before the November 30, 2026 cutoff, identify workflows that depend on Agent Builder and evaluations that depend on Evals. Then decide whether each workflow should be re-created as code or whether its use case fits natural-language prompting in ChatGPT. Validate the replacement against the controls, integrations, and evaluation needs your team actually uses; the announced alternatives are guidance by workflow type, not a published migration checklist.
What did AgentKit look like in practice?
An OpenAI cookbook example illustrates how the components could fit together in a career-development app: the workflow extracts skills from a resume, compares them with a target job, and recommends courses. In the example, Agent Builder creates the workflow, ChatKit provides an embedded chat interface, and Evals supports prompt optimization and trace grading. The cookbook is marked archived and may refer to outdated models or APIs, so it is useful as a conceptual illustration rather than current implementation guidance.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What results did customers report?
OpenAI’s launch announcement quoted customer organizations describing their experiences. These are company-reported testimonials published by OpenAI, not independently verified or general-purpose benchmarks.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Best Value
- Ramp: OpenAI quoted Ramp as saying Agent Builder reduced iteration cycles by 70% and helped it get an agent live in two sprints rather than two quarters.
- Carlyle: OpenAI quoted Carlyle as reporting over 50% less development time for its multi-agent due-diligence framework and a 30% increase in agent accuracy.
- Canva: OpenAI quoted Canva as saying it saved over two weeks building a support agent with ChatKit and integrated it in less than an hour.
- LY Corporation: OpenAI quoted the company as saying its team built and ran its first multi-agent workflow in less than two hours.
The announcement names the organizations but does not identify individual speakers or their roles, so the statements should be attributed to the companies rather than to named executives.
Did AgentKit have a separate subscription price?
At launch, OpenAI said the tools were included with standard API model pricing. That statement does not establish a separate fixed-price AgentKit subscription, nor does it confirm current pricing or access. Check OpenAI’s current product and pricing information before estimating costs for a new or migrated workflow.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
