https://openai.com/index/unrolling-the-codex-agent-loop/
Drawing from this article explaining Codex, I found this snippet interesting:
In Codex, the instructions field is read from the >model_instructions_file(opens in a new window) in ~/.codex/>config.toml, if specified; otherwise, the base_instructions >associated with a model(opens in a new window) are >used. Model->specific instructions live in the Codex repo and are bundled into the >CLI (e.g., gpt-5.2->codex_prompt.md(opens in a new window)).
As you can see, the order of the first three items in the prompt is determined by the server, not the client. That >said, of those three items, only the content of the system message is also controlled by the server, as the tools and >instructions are determined by the client. These are followed by the input from the JSON payload to complete the >prompt.
So essentially it's just the system instruction sits on Openai's servers and that actually changes the behavior of gpt-5.2. This whole article is actually pretty fascinating and I recommend it for a good read if you're interested in learning agentic ai (and how that might help you use Cursor more efficiently) and the usage of tools for agentic ai.
[link] [comments]