{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidrnt64lc6iwvk7ba5dibs57yaz4idjaxquagvo3k45dblauzgmcy",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mni3mcv2woc2"
  },
  "path": "/t/feature-request-opt-in-shared-context-between-chatgpt-and-codex-including-memory-and-reference-chat-history/1382529#post_5",
  "publishedAt": "2026-06-04T16:44:45.000Z",
  "site": "https://community.openai.com",
  "tags": [
    "@qguo551.",
    "@Hey_Stu",
    "@T.Maillet"
  ],
  "textContent": "Thanks for laying this out so clearly @qguo551. I can also see where @Hey_Stu and @T.Maillet are coming from.\n\nThe request makes a lot of sense. Many people use ChatGPT for planning and architecture, then switch to Codex for implementation. Re-copying goals, constraints, and project decisions between the two can get repetitive.\n\nWhat stands out here is the focus on user control:\n\n  * Opt-in sharing\n  * Selective access (memories, projects, specific chats)\n  * Transparency about what context is being used\n  * Easy revocation and exclusions\n\n\n\nThat also helps address some of the concerns around token usage and conflicting context. A summarized, project-level context layer may be more useful than sharing full conversation history.\n\nFor long-term projects, a shared context option between ChatGPT Projects and Codex workspaces would be a valuable addition.\n\n-Mark G.",
  "title": "Feature request: Opt-in shared context between ChatGPT and Codex, including Memory and Reference Chat History"
}