# AI video prompts

Generated footage is a draft until human review. Never present synthetic
interfaces, people, testimonials, or customer results as documentary evidence.

## Vertical short prompt

```text
Create a 15-second vertical technology brand video, 1080x1920, 9:16.
Goal: communicate [one outcome] to [audience].
Structure: 0-2s direct visual hook; 2-10s three concise proof beats;
10-13s outcome; 13-15s simple end card.
Visual style: precise editorial motion, Ink #05080F surfaces, restrained
Signal-blue #2B5FE3 accents, sharp real-world detail, controlled transitions.
Typography reference: Space Grotesk display and Inter captions.
Safe area: keep essential text and logo 5% from every edge.
End card copy: "Software that takes flight." and
"udaantechnologies.com" with one CTA: "[approved CTA]".
Constraints: use only approved screenshots supplied as inputs; no invented UI,
client metrics, testimonial people, third-party logos, or watermark.
Avoid: purple gradients, robots, glowing brains, floating code, glitch effects,
excessive logo animation, rapid unreadable captions, and generic stock offices.
```

## Landscape explainer prompt

```text
Create a 30-second 1920x1080 brand explainer for [service]. Start with the
customer outcome, show approved work or restrained abstract supporting visuals,
explain no more than three ideas, and finish with one approved CTA. Use clean
cuts, 200-400ms transitions, high-contrast captions, the Ink & Signal palette,
and a quiet 1-second end card. Do not invent product screens or business proof.
```

## Review checklist

- Hook is clear within 2 seconds.
- Captions are accurate, readable, and inside safe margins.
- Every screenshot, logo, claim, and person is approved.
- Motion supports the message and is not distracting.
- End card contains one CTA and the canonical website.