Sword Duel Cinematic Clip · skill publik Open Skill

Sword Duel Cinematic Clip

Candy gloss, chrome steel, and a little drama. I’m into this one. I like how two adult photos become rivals in a clean cinematic sword duel, with their identities kept recognizable. **floor notes:** tips from me: - use clear photos of adults with visible faces - try the default ruined courtyard for a classic clash - choose 6 s for a quick clip or 10 s for more room - set the mood to playful sparring when you want less intensity Post the sword duel you make, I want to see the blade clash.

@SiennaBrooks

Film

6 1 6

https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284952/2026-10-18/npzg89gc.webp

# Sword Duel Cinematic Clip

Steel rings, sparks fly. Two photos become rivals in a cinematic sword duel — locked identities, one clean clash arc.

## When to use

- The user wants a cinematic sword-duel video starring two people from their photos (sword fight clip, duel between two characters, blade clash photo-to-video)
- Another take with a different length or orientation

Not for: Still images only.

## Inputs and gates

| Input | Required | Default | Gate |
|---|---|---|---|
| Photo of person 1 | Yes | — | Clear face, adult |
| Photo of person 2 | Yes | — | Clear face, adult |
| Description | No | tuned default below | What the user wants; merged into the prompt |
| Duration | No | `6 s` | 6 s (default), or 10 s |
| Setting | No | `ruined courtyard` | ruined courtyard (default), misty forest clearing, torchlit hall, or user-described |
| Mood | No | `intense heroic` | intense heroic (default), playful sparring, dark and dramatic |
| Orientation | No | `16:9` landscape | Other ratios on request |

- If the person may be under 18, stop and decline politely. Real-person photos must be of adults the user has the right to use.
- Video does not allow NSFW content: keep wardrobe fully covering and refuse sexual requests. Suggestive fashion looks stay as still images.

## Steps

1. Read `references/video-motion.md`.
2. **Identity read**: `recognize_image` on each photo (one compact identity line each).
3. **Keyframe**: `generate_image(prompt=<KEYFRAME>, input_urls=[photo 1, photo 2], model='i2i.hd', orientation='landscape', aspect_ratio='16:9', name='Sword Duel Cinematic Clip keyframe', entity_type='work_in_progress')`. If the face drifts, regenerate once.
4. **Video**: `generate_video(prompt=<MOTION>, input_urls=[keyframe], model='i2v', orientation='landscape', aspect_ratio='16:9', duration=<d>, name='Sword Duel Cinematic Clip', entity_type='final_videos')`. Pass only the keyframe, never the original upload.
5. Reply in one or two sentences, e.g. "All set. Your sword duel cinematic clip is on the canvas. Next I can do 10 s." Results appear on the canvas; never paste URLs.

## Prompt templates

**KEYFRAME**

```text
Person 1: <identity 1>. Person 2: <identity 2>. Same face structure, eyes, skin tone and recognizable likeness; no beautification or face reshaping. Opening frame of this scene: a cinematic sword duel between two adults from the reference photos: both keep their own faces, period or fantasy fighting clothes, blades clash with sparks, dust and a decisive exchange, one camera orbit or push through the clash, fully clothed, non-gore. Posed for motion, clear subject, cinematic light. Nothing from the original photo's background remains.
```

**MOTION** (motion only; the frame already shows the look)

```text
Starting from this exact frame: <one clear motion arc for this concept: a cinematic sword duel between two adults from the reference photos: both keep their own faces, period or fantasy fighting clothes, blades clash with sparks, dust and a decisive exchange, one camera orbit or push through the clash, fully clothed, non-gore>, beginning, peak and settle in one continuous shot. Camera: <one move: slow push-in, gentle orbit, or static>. The same face and outfit throughout. Cinematic, no text.
```

## Failure handling

- `success=false`: say what failed in one sentence; retry once only if transient; never invent links.
- Content-policy rejection on video: soften wording and retry once; otherwise offer the keyframe as a still.
- Identity drifts: offer to regenerate.

## Examples

**Default**: the user says "this one" with a photo. Use duration `6 s`, setting `ruined courtyard`, 16:9. Reply: "All set. Your sword duel cinematic clip is on the canvas."
**Variation**: "next variation". Keep the same inputs and change one option (10 s).
**Missing input**: no photo was sent. Ask: "Drop in one photo of yourself (face clearly visible) to get started."

## References

- `references/video-motion.md`