{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidtlpnjswkmlys7isoxwk2vcgapobpar3dyov7onyzl3w5syzfz7q",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mo7722yrabn2"
  },
  "path": "/t/why-does-codex-repeat-the-same-mistakes/1383615#post_9",
  "publishedAt": "2026-06-13T21:26:31.000Z",
  "site": "https://community.openai.com",
  "textContent": "I have a similar case right now. I once added an instruction to cover a situation that, in hindsight, was rather specific. When I tried to clarify the actual intention later, I realized it no longer worked as expected.\n\nLooking closely at the semantics, I can see that the instruction hints at the specific unwanted behavior. But when I confront the model with this, it does not “see” the issue unless I point it out directly.\n\nThe solution in that case is to revert to the original instructions and re-evaluate from there.\n\nThis assumes that you recently changed the instructions and still have the previous version available.",
  "title": "Why does Codex repeat the same mistakes?"
}