> ## Content Index
> Fetch the complete content index at: https://globalfeed.ai/llms.txt
> Use this file to discover other available public pages before exploring further.

# OpenAI opens the Codex harness to developers with the Agents API
- URL: https://globalfeed.ai/en/openai-opens-the-codex-harness-to-developers-with-the-agents-api/
- Published: 2026-09-11T17:08:55.000Z
- Updated: 2026-09-11T17:08:55.000Z
- Description: OpenAI launched the Agents API, bringing the harness and infrastructure behind Codex to developers with no extra fees beyond tokens and tools used.
- Author: GlobalFeed Editor
- Tags: OpenAI, Codex, ai agents, x-OpenAI, dil-en, denetim-gerekli, video

OpenAI has announced the [Agents API](https://openai.com/index/introducing-the-agents-api?ref=globalfeed.ai), which brings the agent harness and infrastructure behind Codex to developers. A production-ready agent can be created in a single API call by specifying the task, model, tools and environment. OpenAI hosts and maintains the harness itself.

Developers choose the compute environment: an OpenAI-managed sandbox, their own infrastructure, or a partner sandbox. Ecosystem providers with first-class integrations include Blaxel, Cloudflare, Daytona, DigitalOcean, E2B, Modal, Oracle, Runloop and Vercel. OpenAI also introduced its own hosted sandbox, built on the same infrastructure that powers Codex and ChatGPT, configurable with files, packages, skills and plugins.

## Context, tools and subagents

For models running for hours, the API automatically compacts earlier context as a session approaches its context limit, preserving what the agent needs to continue. Developers can build workflows spanning multiple context windows without writing their own compaction logic.

Tool search loads relevant tool definitions only as needed, reducing token usage and cost while preserving the model's cache. Programmatic tool calling lets agents run calls in parallel, chain operations, and filter or combine results in code, returning only relevant data to the context. MCP, custom functions and built-in tools such as web search are supported.

Multi-agent support splits complex tasks into independent pieces handled by subagents working in parallel. Each subagent keeps its own context while the main agent coordinates and merges results, removing the need for custom orchestration.

## Open-source harness

The Agents API runs on the open-source Codex harness, so developers can inspect the logic that coordinates model calls, tools and context, while OpenAI operates and maintains it.

The API is available in beta to all developers today. There are no additional fees; billing covers only the tokens and tools agents use. OpenAI said it will iterate based on feedback as it works toward general availability.