Mon influenceuse IA : la chaîne complète et le prompt master

Le prompt master qui minute un clip UGC de 30 secondes plan par plan, plus la chaîne MakeUGC + Claude + GPT Image 2 + Seedance 2.5.

La chaîne, dans l'ordre

Brancher MakeUGC sur Claude

Dans MakeUGC, va dans les réglages, onglet API Integration : tu y trouves Connect to Claude, ta clé (bouton Regenerate pour en créer une nouvelle) et la case à cocher pour autoriser la synthèse vocale IA. La clé se passe dans l'en-tête X- de l'API MakeUGC Platform.

Les deux images de référence

• <<image_1>> : le selfie de ton personnage. C'est la MASTER REFERENCE.
• <<image_2>> : le produit en flat-lay.

Les prompts tapés dans Claude

build an avatar for UGC videos using GPT image 2

video using the avatar and the product using Seedance 2.5

Le prompt master, recopié tel quel

<<image_1>> (tongue-out selfie in yellow hoodie) MASTER REFERENCE. Character identity, face, skin, hair, makeup, jewelry, room, wall, lighting, color temperature, voice, delivery, overall style and vibe. Everything defaults to this image.
<<image_2>> (flat-lay hoodie) PRODUCT. Butter-yellow oversized sherpa-lined hoodie. Exact color, exact cream sherpa hood lining, kangaroo pocket, dropped shoulders, ribbed cuffs and hem.

=== CAPTURE + AUDIO SPEC === iPhone 17 Pro Max, 24mm main lens, vertical 9:16, handheld-grade sharpening, mild shadow noise, HDR highlight rolloff. No grade, no LUT, no cinematic polish. Light exactly as <<image_1>>: flat soft daylight from off-frame front-left, warm-white wall, matte black horizontal wall fixture behind her at ear height, small round ceiling vent top of frame. No fill, no bounce, no rim. Phone is propped on a dresser for Shots 1, 2, 4, 5 - never visible in frame. Shot 3 is the same phone in her hand, rear camera. Audio: single phone mic at room distance. Thin midrange, auto-gain pumping as she moves nearer and further, real reflections off hard walls, clothing rustle, loud close fabric handling in Shot 2, muffled pass in Shot 4. No denoise, no compression, no music, no room tone bed. Voice matched to <<image_1>> - same speaker, breathy conversational American, mid-pace. Include breath intakes, vocal fry on line ends, one mouth click, one micro-stumble. No captions, no text overlay, no watermark. The only text in frame is the woven neck label.

PART 1 - SHOT BREAKDOWN (15s, five shots)

SHOT 1 (0.0-4.4s) - Walk-back and settle
EFFECT: locked-off propped-phone frame + near-lens motion smear + auto-exposure re-balance + garment inertia
Opens with her right hand still in the lower-right of frame, fingers retreating from the record button. Torso locks, only the finger moves. Her face is out of frame or half-cropped at the top edge. Voice arrives closer and louder to the mic here, auto-gain ducking briefly. Speed 100%. Hard cut out.

SHOT 3 (7.6-10.2s) - Rear camera, floor lay
EFFECT: camera handover to rear lens + handheld tremor + physical push-in + rolling-shutter keystone + label reveal
Hard cut. Perspective is now the phone in her hand, rear camera. Cooler, slightly higher contrast than the front-camera shots. The hoodie is laid flat on pale wood flooring, sleeves spread wide and symmetrical, hood arranged so the cream sherpa lining faces up, body straightened, no bunching. Starts wide over the whole garment, then she physically lowers the phone toward the neck - the floor plane keystones and skews as she leans over it. Real hand tremor, one mid-move tilt correction. Her free hand enters from the top of frame and flips the woven neck label. She thumbs it flat and holds for a beat. Her voice drops in level and goes off-axis. Speed 100%. Hard cut.

SHOT 4 (10.2-12.6s) - Putting it on
EFFECT: return to locked-off propped frame + muffled-audio pass + hair release
Hard cut back to the propped phone. Identical frame, identical room, identical light as Shot 1. She gathers the hoodie, pushes both arms up through the sleeves and hauls it down over the top in one continuous motion. For ~0.6s her head is inside the fabric and the audio muffles and dulls. Head pops through. Hair is dragged loose by the collar, she shakes it out and pulls the trapped strands free. One action, no stacking. Speed 100%.

SHOT 5 (12.6-15.0s) - The 360 - SIGNATURE VISUAL EFFECT
EFFECT: full-body rotation + cloth inertia + rolling-shutter lean + auto-exposure roll + overshoot-and-correct
She steps back once so the full garment fits the frame, raises both arms to shoulder height, and rotates a full 360 degrees counterclockwise over roughly 2 seconds - natural speed, not slow, not performed. The oversized body of the hoodie lags her rotation and swings outward; sleeves trail behind the arms; the hood flops against her back on the half-turn. Vertical lines in the room lean slightly during the fastest part of the turn (rolling shutter). Exposure rolls up about a quarter stop as the pale garment passes through the brighter side of the room, then back down. She overshoots by about 20 degrees, corrects, lands facing camera, drops her arms, weight already shifting to the other hip. Ends mid-energy. No wrap-up pose, no wave, no smile-and-hold. She is still mid-word and still moving when the clip cuts.

=== DIALOGUE (~33 words, gaps intentional) ===
Shot 1 (from 1.3s): "Okay. I just got this and I'm obsessed."

Comment l'adapter à ton produit

• Remplace la description de <<image_2>> par ton produit, couleur et matière exactes.
• Garde la structure CAPTURE + AUDIO SPEC telle quelle : c'est elle qui donne le rendu téléphone.
• Garde le minutage des plans, et réécris seulement les dialogues.

Une question ? Réponds-moi directement en message privé.