🤖 Agents SDK—our new open-source SDK for orchestrating multi-agent workflows, improving upon Swarm. Configure agents with built-in tools, hand off tasks, add safety guardrails, and visualize execution traces for debugging and optimizing performance.

We're launching new tools to help developers build reliable and powerful AI agents. 🤖🔧 Timestamps: 01:54 Web search 02:41 File search 03:22 Computer use 04:07 Responses API 10:17 Agents SDK

Our new API primitive: the Responses API. Combining the simplicity of Chat Completions with the tool-use of Assistants, this new foundation provides more flexibility in building agents. Web search, file search, or computer use are a couple lines of code!

🔎 Web search—integrate with a few lines of code and your agent can get up-to-date answers from the web (with citations). Available in preview:

📄 File search—retrieve precise information from large document collections, with built-in query optimization and custom reranking.

🧑‍💻 Computer use—powered by the same CUA model behind Operator, you can now build agents capable of completing tasks on a computer, such as web QA testing or data entry. Available to select developers on usage tiers 3–5 as a research preview:

These are the first building blocks to help you easily build, deploy, and scale reliable, high-performing AI agents. We’ll be continuing to invest in building the platform for agents—and we can’t wait to see what you build next. Read more in our blog:

Are you searching for a simple way to deploy device certificates with #Intune? Check out SCEPman at

@grok can you explain this in NBA terms

Multi-agent workflows sound powerful. Is there an oversight mechanism in place to prevent emergent AI behaviors from going rogue?