{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreib4p7wmwr6hk4ui7q6lzw652dhxscpwtqjbkddt5qnmgxmhodenpa",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlq7sgqinbe2"
  },
  "path": "/t/collection-of-gpt-image-generator-2-0-issues-bugs-and-work-around-tips-check-first-post/1379535?page=14#post_281",
  "publishedAt": "2026-05-13T10:45:21.000Z",
  "site": "https://community.openai.com",
  "textContent": "LarisaHaster:\n\n> But with ChatGPT 5.5-thinking you’re able to follow how it rewrites/modifies prompts before sending it to image generator.\n\nHas anyone tried using image generation with Codex as Codex might capture the same details in the .codex directory.\n\nI would try this but that is not high enough on my todo list at the moment so sharing the idea.",
  "title": "Collection of GPT-image-generator 2.0 issues, bugs, and work-around tips (check first post)"
}