Mixed Reality Hot Chocolate Pour

{
  "action": "image_generation",
  "action_input": {
    "prompt": "A mixed-reality scene. In the foreground, a real human hand pours [hot chocolate] from a transparent glass server into a clear glass cup. The cup appears to emerge three-dimensionally from inside a smartphone screen that stands upright on a silver stand on a wooden cafe table. Inside the phone screen, [the man from the reference photo with his exact facial features, curly dark hair, and beard] is wearing [a warm beige winter coat and scarf], looking at the cup with a gentle smile and holding it as if receiving the liquid. The [hot chocolate] stream flows seamlessly from the physical world directly into the digital display. Soft rising steam, warm cafe lighting, shallow depth of field, cinematic realism."
  }
}