Best Kling AI Prompts for UGC Ads: Hooks, Demos and Scripts
By the Prompt Shelf team · Updated November 18, 2026 · 9 min read
If you make short product ads for a Shopify or DTC brand, the best kling ai prompts are the ones that keep the product looking real while still stopping the scroll. This guide is for UGC creators, freelance ad makers and small e-commerce teams who use Kling to turn a script into hook shots, product demos and B-roll without booking a studio for every test. You will get a reusable prompt structure, four copy-paste prompts, a before-and-after example and a claims check to run before you generate anything.
Quick answer: The best Kling AI prompts state the subject, action, setting, camera, lighting, style, duration and what to avoid, in that order. For UGC ads, pair a scroll-stopping hook with a product demo and a creator-style explainer, generate two to four versions of each, keep the clips where the product looks accurate, and add your price and logo in editing.
What the best kling ai prompts have in common
A Kling AI prompt is a short block of text that tells Kling's video model what to generate: the subject, the action, the setting, the camera, the lighting, the style, the clip length and what to avoid.
Kling is good at motion, texture and believable handheld camera work, which is exactly what a UGC ad needs. It is less good at remembering that your bottle is matte black and not glossy, or that your logo sits on the left and not the right. A structured prompt is how you hold those details in place.
For a 15 or 30 second ad, you do not write one prompt. You write one prompt per shot: a hook, a demo, a creator explainer and a piece of B-roll, then you cut them together. Each prompt follows the same skeleton:
Subject, action, setting, camera, lighting, style, duration, and a short "avoid" list.
The "avoid" list matters as much as the rest. It is where you tell the model not to invent text, not to add extra fingers and not to change the material.
Step by step
- Pick one product and one buyer problem. Write the product name, the price, the buyer and the problem in one sentence. If you cannot state the problem in the buyer's words, the ad will not land.
- Write the angle brief in ChatGPT, Claude or Gemini. Give the model the product, the price, the buyer, the problem and the proof you can actually show. Ask for five distinct angles, then pick two to test first.
- Run the claims check on the script. Read every line and ask: can I prove this? Cut anything you cannot support with a spec, a certification, a guarantee or a real review count.
- Turn the script into shot prompts. Break the script into a hook, a demo, a creator explainer and one B-roll shot. Write each one with the skeleton above.
- Generate two to four versions of each clip in Kling. Use text-to-video for hooks and mood shots, and image-to-video with a real product photo for any close-up where the details matter.
- Edit the clips together in CapCut, Premiere or your usual editor. Add your real price, offer code and logo as text, and label the ad as required by the platform.
- Test one variable at a time. Change the hook, not the demo. Keep the winner, then test the next hook against it.
AI UGC Ad Video Prompts for Sora, Veo and Kling: Product Demo and Hook Scripts
Shot-by-shot prompts and 15-30 second scripts for UGC-style product ads you can generate with AI video tools.
$24 · See what's inside →
Copy-paste prompts
When to use: the first shot of the ad, when you need a scroll-stopping hook that shows the product in the first second.
Vertical 9:16 video, 4 seconds, Kling AI.
Subject: [PRODUCT, WITH COLOR, MATERIAL AND SIZE].
Action: the product drops into frame from above, lands on [SURFACE], settles with a small bounce, then the camera pushes in slowly.
Setting: [SETTING, E.G. BRIGHT MODERN KITCHEN WITH A BLURRED BACKGROUND].
Camera: macro close-up, 50mm look, locked off for the drop, then a slow push-in, shallow depth of field.
Lighting: [LIGHTING, E.G. SOFT WINDOW LIGHT FROM THE LEFT].
Style: crisp, realistic product commercial, natural colors, phone-quality-plus.
Sound: a light landing sound, no music.
Keep the top third of the frame clear for on-screen text.
Avoid: warped product shape, extra products, invented text or logos, hands, people.
If any bracketed detail is missing, ask me before you write the shot.
When to use: a problem-first hook, when the buyer feels the pain before they see the product.
Vertical 9:16 video, 4 seconds, Kling AI, handheld phone-camera feel.
Subject: [PERSON, AGE RANGE AND CLOTHING] dealing with [THE PROBLEM IN ONE CLEAR MOMENT].
Action: [EXACT ACTION, E.G. TUGS AT A TANGLED KNOT OF CABLES, SIGHS AND DROPS THEM].
Setting: [EVERYDAY LOCATION].
Camera: medium close-up, slight natural handheld shake, eye level, 26mm smartphone look.
Lighting: [ORDINARY REALISTIC LIGHTING].
Style: authentic UGC, unpolished, realistic skin texture, no cinematic grading.
Spoken line, under 8 words, for tools with audio: "[SHORT LINE]".
Avoid: cartoon expressions, the product appearing in this shot, readable brand names, extra fingers.
If any bracketed detail is missing, ask me before you write the shot.
When to use: the middle of the ad, when you need to show the product doing the one thing that matters.
Vertical 9:16 video, 6 seconds, Kling AI.
Subject: hands using [PRODUCT, WITH COLOR AND MATERIAL].
Action: one clear step only: [THE STEP, E.G. PRESS THE TOP BUTTON TWICE AND THE LID POPS OPEN]. Movements are slow and deliberate.
Setting: [SURFACE AND BACKGROUND].
Camera: close-up from 45 degrees above, static or a very slow push-in, 50mm look, product always sharp.
Lighting: [E.G. SOFT, EVEN, BRIGHT DAYLIGHT, NO HARSH SHADOWS].
Style: realistic tutorial UGC, natural colors.
Avoid: complex finger movements, extra fingers, parts that do not exist, steps that skip ahead, on-screen text.
If any bracketed detail is missing, ask me before you write the shot.
Tip: if Kling struggles with the step, film your real hands for this shot and use AI for the hook and B-roll.
When to use: a creator-style explainer, when you need a person on camera but no booked talent.
Vertical 9:16 video, 7 seconds, Kling AI.
Subject: AI-generated presenter, [AGE RANGE, LOOK, CLOTHING], sitting [ON A COUCH / LEANING ON A COUNTER], holding [PRODUCT].
Action: talks to camera like explaining to a friend, small natural hand gestures, glances at the product once.
Spoken line, under 20 words, written as the brand or a creator and not as a customer: "[LINE]".
Setting: [ROOM DETAILS].
Camera: medium close-up, phone on a tripod at eye level, static, 26mm look.
Lighting: [E.G. SOFT DAYLIGHT FROM A WINDOW TO THE SIDE].
Style: realistic creator video, natural pauses, relaxed.
Avoid: lip-sync drift, stiff posture, a reading-from-a-script look, the product changing shape, claims of personal results.
Label the ad according to your platform's AI-content rules.
If any bracketed detail is missing, ask me before you write the shot.
Worked example
Before: a prompt that leaves Kling to guess.
Make a cool video of a water bottle for an ad. Show someone using it. Make it look expensive.
That prompt gives the model no subject detail, no camera, no lighting, no duration and no avoid list. It will invent a bottle, a person and a setting, and you will get a clip you cannot use in a paid ad.
After: the same idea with the skeleton filled in.
Vertical 9:16 video, 5 seconds, Kling AI.
Subject: a matte black insulated water bottle, 500 ml, sitting on a pale oak counter.
Action: a hand twists the lid off, pours water into a clear glass, sets the bottle down centered in frame.
Setting: bright modern kitchen, blurred background.
Camera: close-up from 45 degrees above, slow push-in, 50mm look.
Lighting: soft morning window light from the left.
Style: crisp, realistic product commercial, natural colors.
Keep the top third clear for on-screen text.
Avoid: extra bottles, invented logos, text on the bottle, warped shape, extra fingers.
The second prompt keeps the product, the surface and the camera fixed, so the output is a clip you can actually cut into an ad.
Which shot does what
| Shot type | What it does | Best for | Typical length |
|---|---|---|---|
| Hook shot | Stops the scroll in the first 2 seconds | The opening of the ad | 3-5 seconds |
| Problem-in-action | Shows the pain before the product | Ads that lead with a frustration | 3-5 seconds |
| Hands demo | Shows the product doing one thing | The middle of the ad | 5-8 seconds |
| Feature macro | Shows texture and build quality | Premium or considered purchases | 3-5 seconds |
| Creator explainer | Puts a face and a voice on the brand | Trust and a call to action | 6-8 seconds |
| Lifestyle B-roll | Sets the mood around the product | Transitions and the end card | 4-6 seconds |
Common mistakes
- Writing a one-line prompt and hoping. Fix: use the full skeleton every time. Subject, action, setting, camera, lighting, style, duration, avoid.
- Letting the product change shape or color. Fix: run the close-up shots as image-to-video from a real product photo, and generate several versions.
- Making a claim you cannot prove. Fix: run the claims check before you generate. If a line needs a statistic, a certification or a review count you do not have, cut it.
- Pretending an AI presenter is a real customer. Fix: write the spoken line as the brand or a creator, and label the ad under your platform's AI rules.
- Ignoring the first 3 seconds. Fix: write the hook as its own shot with its own prompt, and test at least three hooks against each other.
- Adding text inside the prompt. Fix: tell Kling to leave the frame clear, then add your price, code and logo as text in the editor.
Frequently asked questions
Which Kling mode should I use for UGC ads?
Use text-to-video for hooks, mood shots and anything without fine product detail. Use image-to-video with a real photo of your product for close-ups, demos and any shot where the label, stitching or finish must match reality. Generate two to four versions of each clip and keep the accurate ones.
Do these prompts work in Sora, Veo and Runway too?
Yes. The skeleton of subject, action, setting, camera, lighting, style, duration and avoid is tool-agnostic, so you can paste it into Sora, Veo, Runway or Kling. Clip length, audio and lip-sync differ by tool, so add spoken lines later as voiceover in the editor when a tool has no built-in speech.
How do I keep Kling from changing my product?
Describe the color, material and size in the subject line, and add an "avoid" line that names the specific things the model tends to invent, such as extra products, invented logos or warped shapes. For close shots, start from a real product photo with image-to-video. Then generate several versions and keep only the clips where the product matches.
Can I use an AI presenter in a paid ad?
You can, but write the spoken line as the brand or a creator, not as a satisfied customer, and follow your platform's ad and AI-content labeling rules. Keep any product claim inside the line provable. This is practical guidance, not legal advice.
How many versions should I generate per shot?
Two to four is the practical range. One version tells you nothing, and more than four costs time you could spend testing a different hook. Re-run the same prompt instead of rewriting it, so the comparison stays fair, then keep the clip where the product looks most accurate.
What should I do when Kling warps the product or adds text?
Do not try to fix it with a longer prompt. Simplify the shot: fewer objects, one action, a plainer background, and a stronger "avoid" line. If the detail still matters, film that shot with your real product and use Kling for the hook and B-roll instead.
Next step
If you want the full set of fill-in prompts, AI UGC Ad Video Prompts for Sora, Veo and Kling: Product Demo and Hook Scripts has 30 prompts across planning, hooks, demos, creator talkers, B-roll and testing, each with a filled-in example. Start with the angle brief, run the claims check, then generate your first hook. You can also browse every pack at /packs.