{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreieusszwhfjf5ryisjytwbwadw6w3mryunzarf7tu7ovihnozpxiwu",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mo6ydd3p2dp2"
  },
  "path": "/t/codex-gpt-5-5-reliability-has-become-a-serious-trust-issue-for-daily-project-work/1383615#post_2",
  "publishedAt": "2026-06-13T19:22:43.000Z",
  "site": "https://community.openai.com",
  "textContent": "sounds like a skill issue.\n\nFor instance, if a model does a mistake you normally want to start over and steer it away from repeating the mistake rather than tellig it to fix the mistake. Because that’s how LLMs work.",
  "title": "Codex GPT 5.5 Reliability Has Become a Serious Trust Issue for Daily Project Work"
}