Phantra AI: UGC Tutorial
Create Phantra UGC tutorials with the person’s own steps, claims, face and product photos, one action per shot, and the result at the start and end.
UGC tutorial
A tutorial shows a way of doing something, in the order it is done, by the
person who does it. Each step is one shot, because each timed shot is
rendered as its own clip: one action, one camera setup, 2 to 10 seconds. You
write the prompt here (free), save it with write_scene, then price it.
The two elements it needs
elements: a character (the face) and a prop (the product), each
with photos. Missing one? Ask the person for the picture: upload_link with
element_kind and element_name for a file on their device (give them the
url, they pick the file, then elements again), or add_element for the
https address they gave you of an image or of the product's own page. Never
find or guess a picture for them. On write_scene the face has role
subject and goes FIRST, because the render anchors on it; the product is
product. A tool or a second product used in a step is another prop.
Whose words these are
Every line spoken, every claim ("it cleared my skin in a week"), every number and every review comes from the person, in this conversation. Ask what they want said and what they can stand behind, and use their words. Never write a testimonial, a star rating, a "customers say", a before-and-after or a statistic they did not give you, and never put a line in the mouth of a customer, a friend or an expert. A claim you cannot source to them is left out, and you say why.
In a tutorial that covers the method itself. Ask for the steps as THEY do them and write those: never add a step, a quantity, a waiting time, a temperature or a safety instruction of your own, and never promise a result ("in one week") they did not state. When a step needs a warning, ask them for it rather than writing one.
The beats (15 to 30 seconds, one step per shot)
Windows run contiguously from 0 to the total (a scene is 4 to 30 s), which
fits three or four steps. More steps is a second scene, joined afterwards
with assemble_clips (no credits).
- The result first: the finished look, dish or setup, held up or turned to the lens, with their promise in one line.
- Step one, then step two, then step three: each a single action in close-up, hands and product in frame, their instruction said while doing it ("two pumps, palms first").
- The result again, to camera, same framing as the open. End abruptly.
Keep the framing the same from step to step (same surface, same height, same light) so the cuts read as one continuous session. Phone feel, in craft terms: a phone propped at chest height or held at arm's length; the room's own light, no fill; true colour and phone sharpness. "No background music. Only diegetic sound: their voice and the product's own sounds."
A worked skeleton (replace everything; the quoted lines are the person's)
Ultra-realistic UGC-style tutorial in 9:16. The attached photo of Ines is her
exact face and clothing; the attached photo of the hair oil is the exact
product and label. Style: phone propped at chest height on a bathroom shelf,
light from a mirror strip, no fill; locked frame; true colour, phone
sharpness; the same framing in every shot.
(0-4s) Ines turns her head so the finished hair catches the light, to
camera: "this is the whole routine, three steps."
(4-10s) Close on her palm: two pumps of oil from the bottle, label to the
lens: "two pumps, warm it up."
(10-17s) Medium: she works it through the ends only, head tilted: "ends
first, never the roots."
(17-23s) Close: she twists one section round a finger and lets it fall:
"twist, and leave it."
(23-27s) Same framing as the open, to camera: "that's it." End abruptly.
No background music. Only diegetic sound: her voice, the pump, the room.
Avoid: plastic AI sheen, CGI-looking hair or skin, on-screen step numbers or
captions, slow motion, named brands or people.
Save, price and render
Show the prompt and take their edits. Then write_scene with prompt,
seconds, refs (the face first), roles, a title, and project_id when
it belongs to a project. quote the scene_id: it prices the keyframes and
the clip (one render per timed shot, each with a token) beside the
balance; more steps means more shots to render, so say the count. Say the
quote's own numbers and WAIT for a yes in this chat naming what to run. Only
then approve with what (keyframes or clip), the quote's tokens and
the same renderer choice; job (wait 15) until done; scene shows the
stills, the clips and the studio link. Tokens last an hour and work once; an
edit means a new quote and a new yes. Never approve unasked or twice;
refused insufficient_credits, stop and say credits are added in the
studio. The steps in full are price-and-render.
- Downloads
- Security audit
- Pass
- Last updated
- 2w ago
- Current version
- v1.0.0
- License
- MIT-0
