How to Curb GPT's Ambition to Become a Leader

mgm2:

it insists on grandstanding with its own “insights” and assigning tasks to the sub-agents anyway.

I am still learning how to use this and have only used it once so far. However, the task ran for seven hours without drifting and completed without hallucinations. The final completion reason was a hard failure, caused by not allowing the AI to download and install code.

/goal — Gives Codex a persistent target to track while a larger task runs.

Additional references:

developers.openai.com

Follow a goal | Codex use cases

Use /goal when a task needs Codex to keep working across turns toward a verifiable stopping condition.

developers.openai.com

Run long horizon tasks with Codex | OpenAI Developers

OpenAI Developer Blog

There is still a lot more to learn about /goal and long-horizon tasks. My suggestion would be to use it on a real-world project; learning by doing will likely be faster and more useful.