# Cutscene prompt framework

From Throne of Aura — The Open Playbook. Replace every bracket.

## Image style line (Nano Banana Pro)

```
[what you want to make] in a 2009 video game style, PS3 Xbox 360 era in-engine render, detailed textures, realistic proportions, waxy skin shading.
```

## Scene prompt template (Seedance)

```
PART [n] — [duration], [aspect ratio]
Create a [era] videogame cutscene, with period-specific rendering and character animation. [One sentence: the scene, and whether it stays in one location or uses a continuous take.]

REFERENCE ROLES
[Game-style scene image] — PRIMARY APPEARANCE: Match this image's rendering across the entire frame: characters, architecture, pavement, background people, lighting and reflections, throughout every cut.
[Character image] — IDENTITY: Preserve face, hair silhouette, proportions, wardrobe and accessories.
[Environment image] — Guide architecture, layout and colours; translate into the game-rendered style.
[Prop image] — Preserve silhouette, colours, construction, material and markings.
[Voice sample] — Voice identity only: timbre, accent, delivery. Speak only the exact dialogue.

RENDERING STYLE
[Silhouettes, clothing folds, hair construction, surface detail, skin shading, light, shadows, reflections.]
[One exclusion sentence for drift you've seen.]

WORLD AND STAGING
[Architecture, palette, atmosphere.]
[Where each character stands, what they touch, screen side, eyelines.]
[Background life with simple purposeful actions — or state that the scene is empty.]

CHARACTER ANIMATION AND PERFORMANCE
[Delivery and pacing.]
[Body and hands: pose transitions, gestures, settles.]
[Face and gaze: mouth, blinks, eyes holding targets.]
[Background movement.]

CAMERA LANGUAGE
[What the camera does for this scene and how it behaves: locked, pushes, drifts, handheld intensity.]

TIMED SHOTS AND EXACT DIALOGUE
[start–end] — [Framing], [one dominant move or hold].
[Action.]
[Speaker]: "[Exact line.]"
[Beat / silence and what the picture shows.]

SOUND AND ESSENTIAL CONTINUITY
[Only what isn't established above: ambience, prop states, reveal restrictions.]
```

## Example: Guangzhou plaza (Higgsfield)

```
PART 1 — 24.59 seconds, 4:3
Create a stylized, dialogue-led videogame cutscene. Scene takes place in a Guangzhou plaza. Use <<<image_1>>> for styling reference.

REFERENCE ROLES
Main character image: Match his face, hair, proportions, wardrobe, and accessories consistently.
BLENDER VIDEO — CAMERA AND BLOCKING ONLY: Follow its framing, cuts, camera paths, and character positions. Apply the scene image’s finished game appearance to that structure.
Animate characters with deliberate pose transitions, restrained facial rigs, simple hand shapes, and repeating background walk cycles.
GAME-STYLE SCENE IMAGE — PRIMARY APPEARANCE REFERENCE: Match <<<image_1>>> rendering across the entire frame: character, architecture, foliage, pavement, background people, lighting, and reflections. Preserve its simplified geometry and visibly textured game surfaces throughout every camera cut.
The main character stands sideways to the railing, torso perpendicular to it, one forearm resting naturally on the rail. Keep him predominantly on the right third, looking toward the active camera. Preserve his position and screen direction across cuts.

PERFORMANCE AND CAMERA LANGUAGE
Dry, conversational confidence with quick, clearly articulated delivery. Small expressions and purposeful gestures attached to specific words; hands settle between gestures. Close shots create intimacy, wides reveal context, and locked close-ups land punchlines. Use gentle handheld sway in the wide shot; keep pushes and lateral drifts controlled.

TIMED SHOTS AND EXACT DIALOGUE
All timestamps are seconds and hundredths.

0.00–4.19 — Medium close-up, slow push-in.
Direct eye contact, curious deadpan:
“Why is it called tomfoolery and not just foolery? Who is Tom and what did he do?”

4.19–9.10 — Hard cut wide, full body.
Reveal his relaxed rail lean and the plaza. Almost imperceptible handheld sway; one small explaining gesture:
“Imagine being so ridiculous, your ‘foolery’ gets credited for generations.”

9.10–10.16 — Hard cut to a locked face close-up.
A brief beat, then absolute deadpan:
“I can.”

10.16–15.18 — Medium close-up, slight three-quarter angle, slow lateral drift.
“I came down from space to China to search for my brother in a Chinese time of HIS life. Then I got hungry.”
A small eyebrow movement emphasizes “HIS.” He recounts the story while remaining at the railing.

15.18–19.11 — Medium shot, subtle push-in.
“So I stopped and got some soup dumplings from a spot in Liwan.”
Finish around 19.00; hold briefly.

19.11–24.59 — Cut closer, restrained lateral drift.
Brief appreciation, then matter-of-fact delivery:
“They’re decadent. I order half the menu immediately. Ask him how much. He says eight dollar special.”
End after “special,” before his price reaction.

BACKGROUND AND CONTINUITY
Sparse pedestrians move in both directions at the edges of the wide frame. Animate natural walking and perspective changes. Smaller figures cross the lower plaza and ascend the escalator. Keep the main character’s face and gestures clear.
Use clean hard cuts at the specified times.
```