ConnectDocsSkills & CLIPricingBlog
Start generating

Docs/Prompting

Prompting

Write the shot like a director.

On the loose surface the prompt is yours, so the realism work the fixed pipelines did server-side is now in your hands. This is what they injected, and how to use it.

On this page

The rubric

Verbatim from the render worker. Fold the lines that matter for your shot into the prompt, in your own words.

  1. 00Critical realism rules (must all be visible in the frame):
  2. 01- skin shows pores, oil sheen on T-zone, baby hairs at hairline, slight under-eye softness;
  3. 02- single mixed light source (soft window daylight + warm interior bulb), realistic shadows;
  4. 03- stable iPhone-like framing by default (no noticeable shake/drift), with slight off-axis angle (about 5-12 degrees) and imperfect centering;
  5. 04- 9:16 vertical, looks like raw iPhone footage, NOT a studio shot;
  6. 05- NO plastic AI sheen, NO uncanny symmetry, NO ultra-smoothed skin;
  7. 06- NO shiny/plastic face, NO glowing light on the face, NO beauty-filter glow;
  8. 07- subtle asymmetry: head tilt, blink, micro-expressions;
  9. 08- hands are always doing something, gesturing, holding a product, adjusting clothing, or otherwise occupied (never limp at the sides);
  10. 09- mouth caught mid-syllable when talking, not closed and not open-smile;
  11. 10- eyes slightly off-center to camera, not a dead stare;
  12. 11- no visible phone, selfie-stick, or outstretched selfie arm unless explicitly requested.

How to use it

  1. 01Write the shot as prose, not tags. The models read sentences better than keyword lists.
  2. 02Order: who (age, look), where (setting, light), what they do with their hands, camera (phone framing, slight off-axis), the spoken words in quotes.
  3. 03Pick 4 to 6 rubric lines that matter for THIS shot and fold them in naturally: "natural skin texture, soft window light, slight head tilt, hands busy with the bottle".
  4. 04For a series, keep the wording of the person and setting identical across calls and pass the same refs.
  5. 05With references, name them: @image1 is the person, @image2 the product, @video1 the move to copy, @audio1 the voice. Numbering is per list, from 1.
  6. 06With a first frame, describe the motion, not the frame: the frame already sets who and where. Say what changes between the first and the last frame.
  7. 07Never put edit or extend wording ("edit the video", "remove", "replace", "extend", "continue") in a prompt that carries video_refs. Describe the new clip.
  8. 08Do not say "selfie" or "phone" unless a phone should be in the frame; say "talking to camera".
  9. 09Speech: quote the words verbatim; about 2.3 words per second. 5 s is 10 to 12 words, 10 s is 20 to 25, 15 s is 30 to 35.

Worked prompts

Talking head, 5 s, seedance-2.0, text mode generate_video

A 28-year-old woman in a bright apartment kitchen, phone-camera framing slightly off-axis, natural skin texture with a little T-zone sheen, soft window daylight from the left and a warm lamp behind her. She holds a small amber serum bottle up near her cheek, tilts her head and says: "Okay. I did not expect this to actually work." Eyes just off the lens, mouth caught mid-word.

Product frame, with the product photo in refs generate_image

The same woman holding THIS bottle (from the reference) up to the lens with both hands, label facing camera, bedroom corner, soft window light, phone photo, natural skin, no beauty-filter glow.

Reference mode, a portrait and a product generate_video

@image1 sits at a kitchen counter, phone-camera framing, holds @image2 up to the lens with the label facing camera and says: "This is the one I kept coming back to." Natural skin, soft window light, a slight lean-in on the last word.

Image-to-video, from a still generate_video

The woman in the frame lowers the bottle, looks straight into the lens and says: "Told you." Handheld drift, the window light stays where it is, nothing else in the room moves.