{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiedkhqtl4x2ufm4djwmzqkot2qu7ca3cljaotswggij3p6uxzlcym",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mptd2nnw5p52"
  },
  "path": "/t/how-to-curb-gpts-ambition-to-become-a-leader/1385720#post_2",
  "publishedAt": "2026-07-04T14:36:34.000Z",
  "site": "https://community.openai.com",
  "tags": [
    "/goal",
    "developers.openai.com",
    "Follow a goal | Codex use cases",
    "Run long horizon tasks with Codex | OpenAI Developers"
  ],
  "textContent": "mgm2:\n\n> it insists on grandstanding with its own “insights” and assigning tasks to the sub-agents anyway.\n\nI 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.\n\n/goal — Gives Codex a persistent target to track while a larger task runs.\n\nAdditional references:\n\ndevelopers.openai.com\n\n### Follow a goal | Codex use cases\n\nUse `/goal` when a task needs Codex to keep working across turns toward a verifiable stopping condition.\n\ndevelopers.openai.com\n\n### Run long horizon tasks with Codex | OpenAI Developers\n\nOpenAI Developer Blog\n\nThere 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.",
  "title": "How to Curb GPT's Ambition to Become a Leader"
}