ugc-video-prompts
Generate ultra-realistic UGC/influencer-style video clips inside Palmier Pro — talking selfies, product demos, candid lifestyle motion — using the proven still-then-animate pipeline instead of cold text-to-video. Drives Palmier Pro's generate_video, get_timeline, and get_media tools directly, with model selection grounded in this project's actual capabilities (dialogue/audio support, reference limits, resolution). Use whenever the user wants UGC-style video, influencer clips, talking-to-camera ad creative, "make a video of [a person/product]," or animating a still into motion — even if they don't say "UGC" explicitly. Pairs with the ugc-photo-prompts skill, which should run first to produce the anchor still.
palmier-io · Static checks on file
Installation
Read the source and dependencies before running a command. Copying does not install or execute anything.
npx skills add palmier-io/palmier-skills --skill ugc-video-promptsAbout this skill
name: ugc-video-prompts description: Generate ultra-realistic UGC/influencer-style video clips inside Palmier Pro — talking selfies, product demos, candid lifestyle motion — using the proven still-then-animate pipeline instead of cold text-to-video. Drives Palmier Pro's generate_video, get_timeline, and get_media tools directly, with model selection grounded in this project's actual capabilities (dialogue/audio support, reference limits, resolution). Use whenever the user wants UGC-style video, influencer clips, talking-to-camera ad creative, "make a video of [a person/product]," or animating a still into motion — even if they don't say "UGC" explicitly. Pairs with the ugc-photo-prompts skill, which should run first to produce the anchor still.
UGC-Style Video Prompts
Core principle
Video models are decent at motion but worse than image models at solving "doesn't look AI" from zero — skin, lighting, and identity drift across many frames in ways a single still doesn't. Don't text-to-video a UGC clip from scratch. Generate the anchor still first (with the ugc-photo-prompts skill, or reuse one already in the media library), then animate it. The still already solved realism; the video model only has to solve movement and speech, which is what it's actually good at.
The pipeline
- Anchor still. Already exists in the media library, or generate one with
ugc-photo-prompts. This locks subject identity, wardrobe, product, lighting, and skin texture before any motion is added. - Motion prompt. Describe only what changes from that frame — camera behavior, body motion, speech. Don't redescribe the subject/setting; the still already carries that.
- Model pick. Based on whether the beat needs synced dialogue, multiple combined references, or just clean simple motion — see the table below.
- One clip = one beat. A full UGC ad is a sequence of short jump-cut beats (hook, demo, reaction, CTA), not one continuous shot. Generate each beat as its own clip off the same anchor (or a small consistent set of anchors), then cut them together on the timeline.
- Lock the voice across beats. Same problem as appearance drift, different mechanism — describe the voice explicitly in every beat's prompt (or pin it with a
referenceAudioMediaRefsclip on Seedance) so the speaker doesn't sound like a different person from one cut to the next.
Motion-prompt formula
The still already did slots 1–4 and 6–8 from the photo formula. This is just what's added for motion:
- Seed reference. State plainly that this animates the provided still — most models infer this from
startFrameMediaRef/referenceImageMediaRefsalone, but naming it keeps the prompt focused on change, not redescription. - Camera behavior. Default to handheld micro-shake or a static phone propped on a surface — not a smooth gimbal move or a dolly/crane. "Smooth cinematic camera move" is as strong an AI tell in video as studio lighting is in stills.
- Subject motion. Small, continuous, natural movement — a weight shift, a blink, hair moving, a hand gesture mid-sentence. Avoid big choreographed actions unless the beat specifically calls for one (e.g., the demo beat).
- Speech, if any. Put the exact line in quotes, written the way someone actually talks — contractions, a small pause, not ad copy. "okay wait this actually smells amazing" reads as real; "Introducing our revolutionary new formula" doesn't, regardless of delivery. If this speaker appears in more than one beat, describe the voice itself — and always include all four: pitch, tone, accent, and pace, not just one or two. "A warm, slightly raspy voice" alone is incomplete; "a warm, friendly tone, medium-pitched, neutral American accent, relaxed conversational pace" is the actual bar. Reuse that exact wording verbatim in every beat's prompt — each
generate_videocall has no memory of any other, so the same anchor face will get a different voice per clip unless all four are pinned down explicitly every time. Seedance can do better than words: pass the rendered audio from beat one as areferenceAudioMediaRefsinput on later beats to lock the actual voice rather than approximate it in text. - Duration matched to the beat. Pick the shortest duration the model offers that fits one beat (often 4–6s) rather than padding a single clip to cover multiple actions — multi-action clips are where motion starts looking choreographed.
- Negative motion cues. Add: no smooth gimbal stabilization, no cinematic camera move, no professional production polish, no overly choreographed motion, no studio lighting. Carry over the standing negative list from
ugc-photo-promptsfor the visual frame itself.
Model selection
Grounded in this project's actual model catalog and observed behavior, including real failures — not just the catalog spec sheet.
Default to Grok Imagine Video or Kling V3 for most beats. Between those two: Kling for anything needing a longer or more nuanced line, Grok when a quick draft is enough.
| Model | Best for | Notes |
|---|---|---|
| Grok Imagine Video | Default pick for most beats | Renders synced dialogue audio with accurate lip-sync in this project's tests. Caps at 720p — fine for most UGC use, not a 4k hero asset. |
| Kling V3 / Kling O3 | Default pick, especially longer/dialogue-heavy beats | Renders synced dialogue audio automatically when the prompt has a quoted line — confirmed repeatedly in this project. O3 allows more references (7 vs 3) if anchoring more than one element. Up to 4k, 3–15s, 16:9/9:16/1:1. |
| Seedance 2 / Seedance 2 Mini | Beats combining multiple elements — person + product + a reference audio style | **Never default to 1080p/4k on Seedance 2 — it's expensive. If a beat doesn't specifically need that resolution, use Seedance 2 Mini (720p cap) instead, and if full Seedance 2 at 1080p/4k seems warranted, tell the user it costs significantly more and let |