AI Ad Prompt Guide — AI skill for Claude Code

---

name: ai-ad-prompt-guide description: | Battle-tested prompting guide for AI video and image generation in advertising. Covers universal prompting rules, hallucination prevention, the SLCT framework for structured prompts, camera movement vocabulary, UGC formulas, product shot techniques, and the pass³ quality test. Includes model-specific guidance for Sora 2, Veo 3, Kling, Flux, Nano Banana Pro, Seedance, and more.

Use when: "write a video prompt", "fix hallucination", "prompt for Sora", "generate B-roll", "UGC prompt", "product shot prompt", "which model should I use", "AI image prompt", "prompt engineering for video", "camera movement", "SLCT", "pass³ test", "AI generation prompt", "text to video prompt", "image to video prompt", "model comparison", or any AI generation prompting question.

AI Ad Prompt Guide

Universal prompting framework for AI video and image generation — works across all major models.

---

Part 1: Universal Prompting Rules (Standalone)

1.1 The SLCT Framework

Every effective AI generation prompt has four components. Use SLCT as a checklist:

**S — Subject**: What is the main focus? **L — Lighting/Look**: What's the visual mood? **C — Camera**: What angle, movement, and framing? **T — Technical**: Resolution, aspect ratio, duration, style keywords?

SLCT Examples

**Product B-roll:**

S: A glass bottle of amber serum on a marble bathroom counter
L: Soft golden morning light streaming from the left, creating gentle highlights on the glass
C: Slow push-in from medium to close-up, shallow depth of field
T: 4K, 5 seconds, photorealistic, product photography style

**UGC-style:**

S: A woman in her late 20s opening a package on her couch, looking excited
L: Natural indoor lighting, warm tones, slightly imperfect like a phone camera
C: Medium shot, handheld slight shake, selfie-style front camera angle
T: 9:16 vertical, 5 seconds, realistic, casual home setting

**Cinematic hero shot:**

S: A sleek electric car driving along a coastal highway at sunset
L: Dramatic golden hour, long shadows, warm highlights on the car body
C: Low-angle tracking shot from the front quarter, smooth dolly movement
T: 16:9, 8 seconds, cinematic, film grain, anamorphic lens flare

1.2 Hallucination Prevention

AI models hallucinate when prompts are ambiguous, contradictory, or physically impossible. These rules minimize bad outputs:

The 5 Rules of Hallucination Prevention

1. **Be spatially explicit**: "A bottle on the LEFT side of a marble counter, a plant on the RIGHT" — not "a bottle near a plant on a counter"

2. **Limit entities**: Maximum 3 main subjects per scene. More = more chance of merging/distortion

3. **Avoid negatives**: Don't say "no people in the background" — instead describe what IS there: "empty cafe with wooden chairs"

4. **Use real-world references**: "lighting like a Vogue cover shoot" anchors the model better than "beautiful professional lighting"

5. **Specify quantit