{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreigfahxwoxnh36y6fnm5c4l4cxls76kuvbxtj7te2iev7olkoov4um",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mejdrb2cjry2"
  },
  "path": "/t/apps-sdk-multiple-widgets-in-a-single-chat-turn-fail-to-render-inside-the-native-chatgpt-mobile-app/1373881#post_1",
  "publishedAt": "2026-02-10T14:52:51.000Z",
  "site": "https://community.openai.com",
  "textContent": "I’ve been developing an app for ChatGPT with Apps SDK and sometimes the agent decides to perform multiple tool calls that each can render its widget in a single chat turn. Why this might not be the most common use case for most apps, it’s a completely justified use case for my app.\n\nThis works OK on the web (both desktop and mobile), but fails quite randomly inside the native mobile app.\n\nExpected behavior:\n\n  * Multiple widgets are correctly rendered.\n\n\n\nActual behavior:\n\n  * 1 o 2 widgets are rendered, rest are not for no reason at all (no Sentry errors, widget code is directly not being initialized)\n\n",
  "title": "Apps SDK: multiple widgets in a single chat turn fail to render inside the native ChatGPT mobile app"
}