Build an Agent Improvement Loop with Traces, Evals, and Codex
This notebook builds an improvement flywheel for an agent. We start with real traces, add human and model feedback, turn that feedback into
Research Layer
Focused guides, prompts, and workflows with a platform-specific visual system.
This notebook builds an improvement flywheel for an agent. We start with real traces, add human and model feedback, turn that feedback into
When an agentic system fails, the problem is often larger than a single bad response. A handoff may happen too late, a specialist agent may
This cookbook is about closed-loop agent workflows: agents that produce an output, validate it, and use the feedback to improve the next pas
OpenAI's gpt-image generation models are designed for production-quality visuals and highly controllable creative workflows. They are well-s
Codex models advance the frontier of intelligence and efficiency and our recommended agentic coding model. Follow this guide closely to ensu
The Responses API is a new way to interact with OpenAI models, designed to be simpler and more flexible than previous APIs. It makes it easy
Codex and the gpt-5.2-codex model (recommended) can be used to implement complex tasks that take significant time to research, design, and i
Goals are persistent objectives in Codex that keep a thread working toward a defined outcome across turns. A Goal gives Codex a completion c
OpenAI Developer Blog
How we preserved private network boundaries while supporting MCP streaming, authentication, and an inspectable client.
Using skills and GitHub Actions to optimize Codex workflows in the OpenAI Agents SDK repos.
Practical techniques for steering GPT-5.4 toward polished, production-ready frontend designs.
Five stories from developers building agentic products with the Responses API in its first year.
Use Codex and Figma to bring real, running interfaces into Figma, refine them, and bring changes back to Codex.
Lessons from how Perplexity Computer's voice agent was built with the Realtime API.
Turn your phone into a Codex control center for starting, steering, reviewing, and organizing engineering work.
Function calling lets you connect GPT models to external tools and APIs. Here's how to get started.