A Kling prompt for the Fridge POV Tiramisu AI trend, free of names

Prompt text

Vertical 9:16 image-to-video, about 8 seconds, from reference image 1.
Only one person appears: the adult in image 1, with face, skin tone, hair and build unchanged throughout.
Setting: the interior of a bright white fridge, seen from the rear of a shelf through the open door. Cool light across the top, a large foil tray of tiramisu filling the lower foreground, an unlit silent kitchen behind. Nothing legible: no lettering, stickers or logos.
Camera: one locked take at shelf height, slightly wide, without cuts or zoom. The person stays still behind the tray, head to chest in frame.
0-1s: holding a spoon above the tray, calmly looking at the lens.
1-2.5s: the spoon lifts a big scoop and it is eaten.
2.5-5s: another big scoop from the front corner, a hole showing in the dessert, eaten.
5-7s: scoop number three from that same side, equally calm.
7-8s: the spoon goes back to the tray as the person gazes at the camera, content.
A single spoon. No talking, no lip-sync, nobody else in view.
Audio: fridge hum, foil clinks from the spoon, a barely audible instrumental pad. No vocals, no voices.
Photorealistic and naturally coloured, with no subtitles, no logos and no watermark.

Tips

• Attach one adult photo as image 1.
• Lay your sound over it in TikTok later; a text prompt cannot place every scoop, so you may need to nudge the sound.
• Leave out names, sound titles and brands, since tools can filter them; the prompt describes the action instead.
• Tools change quickly, so check the length and aspect ratio yours allows, then edit the first line.