Instagram · 19 Sept 2026
View the original on instagram.comTopicOpenAI Agents API launch
OpenAI Agents API launch
By @romippatel · Instagram
Source: https://www.instagram.com/reel/DdO7agIoHDq/?stkn=MWpieHU2NDlxcTk2NA==
AI generated summary of the linked source. Posted anonymously. Transcripts are published only for public links, never for uploads, documents or pasted text.
Share
The AI link is plain markdown. Paste it into any assistant and it can read the whole public entry.
Summary
OpenAI released an Agents API that packages the "harness" (context management, tool orchestration, sub-agent coordination, sandboxing) most teams have been building by hand into a single managed service. It's the same infrastructure that runs Codex, now exposed for anyone to call.
Core ideas
- The harness is the hard part: The model itself is easy to use; the real engineering work is managing context, tools, sub-agents, and safe code execution around it.
- One API replaces custom infrastructure: A single call gives you a task, model, tools, and environment, and OpenAI runs the loop instead of you writing orchestration code.
- Automatic compaction solves long sessions: When context limits approach, the API compresses earlier context on its own, letting agents run for hours across multiple context windows.
- Tool efficiency cuts context waste: Tool search only loads definitions when needed, and programmatic tool calling lets agents run calls in parallel and filter results in code instead of dumping junk into context.
- Sub-agents come built in: The main agent can delegate independent tasks to parallel workers, each with its own context, with no orchestration work required from the developer.
- Sandbox choice stays flexible: You can run on OpenAI's own infrastructure or partners like Cloudflare, Modal, and Vercel.
- The harness code is open source: It's the same Codex harness, so you can inspect exactly what's running under the hood.
- Pricing stays simple: No extra fees for the service itself, you just pay for tokens and tools as usual.
- Model providers are climbing the stack: First they sold tokens, then tools, now they're selling the whole agent runtime.
Quotes
“when you build an agent the model is the easy part the hard part is everything around it”
“first they sold you tokens then tools now the whole agent runtime”
“the agents API is the harness that powers Codex, exposed as a managed service”
Resources
Links to the tools, products, repos or reading this entry mentions. Found by a web search of what the entry names, so you can go and use them.
Anyone can run this once. After that the links are saved on this entry and shown to everyone.
Remix this knowledge
Turn this entry into something you can post. Pick a format and get a fresh, original rewrite.
Remixes are AI generated, original wording, and free to reuse.
Talk it through
Open a round table on this entry and argue with other people about what it means.
Start a round tableDrop your own
Paste any video, podcast or audio link and get the transcript, the core ideas and the quotes back in one pass.
Try Drop