OpenAI has introduced the Agents API in public beta, bringing the agent harness behind Codex to developers through a managed API. The service is designed for agents that need to keep working across longer sessions, use tools, coordinate subagents, and operate in environments where they can work with files and code.

Developers can choose where the agent runs: in an OpenAI-managed sandbox, on their own infrastructure, or with supported sandbox partners. OpenAI manages the harness itself, while developers provide the tools, knowledge, and workflows that make the agent useful for a specific job.

The release reflects a broader shift from one-shot AI responses toward software that can carry work forward over time. OpenAI says the Agents API can coordinate context, tool use, and subagents, and is built on the open-source Codex harness.

The API is available to developers in public beta. OpenAI says there is no separate Agents API fee beyond the tokens and tools the agent uses.