OpenAI Agents API
OpenAI has introduced the Agents API, a new toolset designed to help developers build and manage autonomous AI agents. The platform handles session orchestration and sandbox environments, allowing agents to execute code, access MCP servers, and perform web searches.
Why it matters
This release lowers the barrier for developers to deploy complex, multi-agent systems that can interact with real-world tools and environments.
The Agents API gives your application access to the Codex harness through an OpenAI-managed API.
OpenAI manages sessions, orchestration, context compaction, and recovery while your application provides tools and chooses its execution environment.
Agents can operate in a sandbox where they can execute code, edit files, connect to MCP servers, and produce artifacts.
Model usage is billed at the selected model’s API rates . OpenAI tools use their standard rates , and OpenAI-hosted sandboxes use standard container rates .
The Agents API is built around four main concepts:
Start with an OpenAI-hosted sandbox in the quickstart :
With an OpenAI-hosted session, your application sends input and receives events, while OpenAI runs the agent and provisions and manages its sandbox. See environment options for setup and limitations.
What the managed harness provides The managed Codex harness supports:
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in