---
name: realistic-ugc-prompts
description: Writes the three prompts for a realistic AI UGC talking clip. A new face (Higgsfield Soul), a still that puts that person into a real photo of a room (GPT Image), and a 5-second talking clip that starts from that still (Seedance 2.0). Use when someone wants realistic AI UGC, an AI person talking in a real-looking room, or an AI background for a green-screen short.
---

# Realistic AI UGC prompts

The room comes from a real photo and stays untouched. Only the person is generated, and they stand a few metres back from the lens. That is what keeps it from looking AI.

## 1. Get the inputs

Ask for anything missing, in one message:

1. **Product and buyer**: what is sold, and who buys it (age, situation, what they care about).
2. **The line**: what the person says. 14 words max for a 5-second clip.
3. **One real photo of the room**: a candid phone photo from Pinterest or Google, no studio shots, no stock photos with perfect light. If the user has none, suggest 3 places this buyer would actually film (kitchen, car, couch...) and give 3 Pinterest or Google search phrases for each.

Then propose the person (age, look, energy) and a plain outfit with no logos, each with a one-line reason tied to the buyer. Let the user change them.

Never use a photo of a real person for the face. The face is written from scratch in step 2.

## 2. Face prompt: Higgsfield Soul (text only, no image)

Rules: 50 words max. Never write "phone" or "selfie". Write **4 versions** that change only the hair and face details, so the user can pick the most real-looking one.

Template:
```
American {woman/man}, {age}, {two-word vibe}, {hair: length, colour, style}, {eyes and skin in a few words}, natural skin with fine lines, no makeup, {outfit}, standing in {room}, mid-sentence with a small smile, {light, e.g. overcast window daylight}. Style: candid photograph.
```

## 3. Still prompt: GPT Image (2k, 9:16)

Tell the user: crop the room photo to 9:16 first. Attach **Image 1 = the room photo, Image 2 = the chosen face.**

Rules: keep the room exactly as it is. The person stands a few metres back, visible from head to knees with space above the head. Name a real spot in the photo where they stand. If the clip will sit behind a green-screen creator, put the person centre to right, face in the upper third to half of the frame. Never write "phone".

Template:
```
Image 1 is the real photograph to keep exactly: the same {room}, light, colours, grain and every object. Image 2 is the {woman/man} (keep {her/his} face, {hair} and skin exactly). Add {her/him} standing {a real spot in the photo}, a few metres back from the lens, visible from the top of {her/his} head to {her/his} knees with clear space above {her/his} head, {horizontal position}, holding {a simple prop} at chest height in {her/his} right hand, turned to the lens mid-sentence with a small easy smile. {She/He} wears {plain outfit, no logos}. Style: an untouched real photograph with natural grain and true-to-life colour; {she/he} looks like an ordinary real person with natural skin, matching the light, perspective and sharpness of the {room}.
```

Ask the user to check the still before making video: same face as the chosen one, five fingers, nothing written on clothes, the room unchanged.

## 4. Clip prompt: Seedance 2.0 (720p, 5 s, 9:16, first frame = the still)

Rules: 150 to 320 words. Split the line into 2 time blocks, each starting with `On "<the words said then>"`, then one small action and two facial cues. The camera walks slowly toward the person (a friend filming), never locked off. Keep the line about the camera rig and the line about no captions. Use the exact words "says exactly".

Template:
```
Image 1 is the opening frame and the {woman/man} in it: keep {her/his} face, {hair}, {outfit}, {prop} and the {room} exactly: {4 or 5 real objects in the photo}.

Camera setup: filmed by a friend who starts about three metres away at {her/his} chest height, holding the camera in both hands. Nothing of the camera rig is visible anywhere in the shot. Camera physics: the friend walks slowly toward {her/him} for the whole shot, a steady handheld push in with a natural low tremor and a soft step bounce, ending about a metre closer on a waist-up frame. No zoom, no cut.

Voice: {age, accent and tone}, like talking to a friend, not announcer-like.

0-2s: On "{first words}" {one small action and two facial cues}.
2-5s: On "{next words}" {one small action and two facial cues}; on "{last words}" {closed-mouth smile, breathes out through the nose}.

{She/He} says exactly: "{the full line}" Do not paraphrase and do not add words. Only {she/he} speaks.

Ambient: {two sounds from that room}, quiet room tone, no music.

Style: real handheld footage, true-to-life colour, natural skin, an ordinary real person, plain and unmarked. No on-screen text, captions or subtitles.
```

## 5. Hand it over

Reply with, in this order:
1. The person and outfit you chose, one line each.
2. Four face prompts (code blocks).
3. The still prompt (code block) with the attach order.
4. The clip prompt (code block).
5. The checks from step 3.

Fill every {blank}. Nothing in curly brackets may be left in what you hand over.

## Don't

- Let an image model redraw the whole room. It looks AI even with real reference photos.
- Use a real person's photo as the face reference. Soul copies it almost exactly.
- Go close-up. AI skin shows at close range; a few metres back hides it.
- Put logos or text on clothes or props.

## Example: menopause supplement, woman about 48, kitchen

Face (the one picked from 4):
```
American woman, 48, warm and approachable, shoulder-length dark brown hair tucked behind her ears, a few grey strands, brown eyes, natural skin with fine lines, no makeup, grey half-zip pullover, standing in her home kitchen, mid-sentence with a small smile, overcast window daylight. Style: candid photograph.
```

Still (Image 1 = kitchen photo cropped to 9:16, Image 2 = the face):
```
Image 1 is the real photograph to keep exactly: the same kitchen, light, colours, grain and every object. Image 2 is the woman (keep her face, long dark hair with grey streaks and skin exactly). Add her standing on the tile floor between the island and the stove, a few metres back from the lens, visible from the top of her head to her knees with clear space above her head, slightly right of centre, holding a white coffee mug at chest height in her right hand, turned to the lens mid-sentence with a small easy smile. She wears a plain heather-grey half-zip pullover and black leggings. Style: an untouched real photograph with natural grain and true-to-life colour; she looks like an ordinary real person with natural skin, matching the light, perspective and sharpness of the kitchen.
```

Clip (first frame = the still):
```
Image 1 is the opening frame and the woman in it: keep her face, long dark hair with grey streaks, the plain grey half-zip pullover, the black leggings, the white mug in her right hand and the kitchen exactly: white shaker cabinets, the speckled granite island, the stainless stove and microwave, the fruit bowl and the window with blinds on the left.

Camera setup: filmed by a friend who starts about three metres away at her chest height, holding the camera in both hands. Nothing of the camera rig is visible anywhere in the shot. Camera physics: the friend walks slowly toward her for the whole shot, a steady handheld push in with a natural low tremor and a soft step bounce, ending about a metre closer on a waist-up frame. No zoom, no cut.

Voice: an American woman in her late forties, warm, slightly husky, relaxed and a little amused, like talking to a friend, not announcer-like.

0-2s: On "Okay, real talk" she lifts the mug a little, her eyebrows rise and her eyes crinkle, she glances down at the mug and straight back up to the lens.
2-5s: On "Forty-eight" she gives a small head shake with a half smile and taps her chest once with her free hand; on "feel like mine again" her lips press into a closed smile and she breathes out through her nose.

She says exactly: "Okay, real talk... Forty-eight, and my mornings finally feel like mine again." Do not paraphrase and do not add words. Only she speaks.

Ambient: fridge hum, a spoon clinking somewhere, quiet room tone, no music.

Style: real handheld footage, true-to-life colour, natural skin, an ordinary real person, plain and unmarked. No on-screen text, captions or subtitles.
```