These are reusable fill-in-the-blank skeletons for Wan 2.7, Alibaba's open-weight video model (Tongyi Wanxiang). Anything in [BRACKETS] is a slot you replace — swap [SUBJECT], [SETTING], [ONE MOTION], [CAMERA MOVE], and [LIGHTING] for your scene, then delete the brackets. Keep the word order, because the Subject + Scene + Motion + Lighting + Camera + Style sequence, with audio lines underneath, is what makes Wan follow the shot. Every template ends with audio slots — SFX, Ambient, Music, or a spoken line — since native synced audio is Wan's signature. Wan ships under an Apache 2.0 license, so the clips are free to use commercially. Set aspect ratio, resolution, and duration in the app; the notes like (16:9, 1080p, 10s) are reminders, not text to paste. New to this? Start with the 35 best Wan prompts or the Wan prompt cheat sheet.
Cinematic & B-roll
Director-style skeletons for atmospheric and hero shots. One clear action, one camera move, one lighting term per fill, and let the ambient audio carry the mood.
1. Establishing hero shot
[SUBJECT] in [SETTING] at [TIME OF DAY], [ONE MOTION]. [SHOT SIZE], [CAMERA MOVE] to reveal the full scene. [LIGHTING], [LENS], shallow depth of field, cinematic film look.
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: [SUBJECT] and [SETTING] set the scene; [ONE MOTION] is the single action (mist drifting, a figure walking); [CAMERA MOVE] like slow dolly-in or crane up does the reveal. Keep it to one move.
2. Emotional close-up mood piece
Close-up of [SUBJECT] as they [ONE MOTION], [SETTING] softly out of focus behind. Slow push-in, otherwise static. [LIGHTING], [LENS], subtle film grain.
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: [ONE MOTION] is a small human beat — a blink, a breath, a glance. Keep [LIGHTING] low-key for emotion; the push-in is the only camera move.
3. Moody tracking shot
[SUBJECT] [ONE MOTION] through [SETTING] as [ENVIRONMENTAL DETAIL] surrounds them. [SHOT SIZE], [CAMERA MOVE] alongside at a constant distance. Low-key [LIGHTING], practical [LIGHT SOURCE], anamorphic-style flares, [LENS].
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: [CAMERA MOVE] should support the subject — a handheld follow or steadicam track. [ENVIRONMENTAL DETAIL] (rain, neon, fog) gives Wan continuous motion to render.
Product & ad video
Clean, controllable skeletons for commercial video. For label accuracy, start from a real photo of the item with image-to-video before filling these in — see the finished product prompts.
4. Product hero orbit
A [PRODUCT] rests on [SURFACE], [PRODUCT DETAIL] catching the light. [SETTING], [TIME OF DAY] light. [SHOT SIZE], slow orbit [clockwise/counter-clockwise] around the product. [LIGHTING], softbox key with a bright rim light, glossy reflections, [LENS] macro look.
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: [PRODUCT] and [PRODUCT DETAIL] name the hero and its texture; the orbit is your one move. Start from a product photo so the label stays on-model.
5. Product on a rotating platform
A [PRODUCT] sits on a slowly rotating platform against a seamless [COLOR] backdrop, [PRODUCT DETAIL] crisp in the light. [SHOT SIZE], static camera as the platform turns. High-key studio [LIGHTING], soft even shadows, clean commercial look, [LENS].
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: Move the platform, not the camera — that keeps [PRODUCT] razor-sharp. Set [COLOR] to your brand backdrop and [MUSIC] to the ad's energy.
6. Liquid / pour macro
[LIQUID] pours into [CONTAINER] and swirls as [BUBBLES/STEAM/CONDENSATION] forms around it. Extreme macro close-up, [CAMERA MOVE]. [LIGHTING], backlight rim light, high-speed slow-motion feel, [LENS] macro.
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: The pour is a clean physics moment Wan renders convincingly. Keep [CAMERA MOVE] to static or a slow push-in so the liquid stays the star.
Talking characters & voiceover
Skeletons built around Wan's synced speech. Keep the [SHORT SPOKEN LINE] to one sentence — a clip is only 5-15 seconds — and start from a reference photo when you need a consistent face. More in the realistic-video guide.
7. Presenter to camera
A [PRESENTER DESCRIPTION] looks directly at the camera in [SETTING], [ONE MOTION] lightly as they speak. [SHOT SIZE], very slow push-in. [LIGHTING], softbox key, clean background, [LENS].
Dialogue: [PRESENTER] says, "[SHORT SPOKEN LINE]"
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: [PRESENTER DESCRIPTION] sets the look; [SHORT SPOKEN LINE] is one natural sentence. The push-in is the only move so lip-sync stays clean.
8. Two-person conversation
[PERSON A] and [PERSON B] talk across [SETTING]. Over-the-shoulder shot on [PERSON B], static camera. [LIGHTING], shallow depth of field, [LENS].
Dialogue: [PERSON A] says, "[SHORT SPOKEN LINE]" [PERSON B] replies, "[SHORT SPOKEN LINE]"
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: Give each person one short line only. Keep [LIGHTING] and [SETTING] identical if you plan to cut to a reverse angle in a second clip.
9. Narrated voiceover cutaway
Close on [SUBJECT] as [ONE MOTION] happens in [SETTING]. [SHOT SIZE], static camera. [LIGHTING], appetizing saturation, shallow depth of field, [LENS].
Voiceover ([VOICE STYLE] narrator): "[SHORT SPOKEN LINE]"
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: The voiceover carries the step while the visuals show [ONE MOTION]. Set [VOICE STYLE] like "calm female" or "energetic male" to steer delivery.
Nature & landscapes
Wide, slow, and let the ambience breathe. One gentle move per shot renders far cleaner than a busy fly-through.
10. Landscape reveal
[NATURAL LANDSCAPE] stretches to the horizon as [WEATHER/ATMOSPHERE] moves across it at [TIME OF DAY]. Extreme wide shot, [CAMERA MOVE] to reveal the scale. [LIGHTING], volumetric light, [LENS].
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: [WEATHER/ATMOSPHERE] (drifting mist, rolling clouds) is your one motion. Keep [CAMERA MOVE] to a single steady crane or forward push so the aerial stays coherent.
11. Texture / detail cutaway
Extreme close-up of [NATURAL TEXTURE] as [ONE MOTION] happens. Static camera, macro focus. [LIGHTING], shallow depth of field, subtle film grain, [LENS] macro.
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: [NATURAL TEXTURE] plus a tiny [ONE MOTION] — a dewdrop falling, a leaf trembling — gives Wan a clean, believable macro moment.
Action & motion
High-energy skeletons. Give the model one dominant motion and a camera that supports it — stacked moves turn to mush.
12. Chase / pursuit
[SUBJECT] [runs/rides/drives] fast through [SETTING], [ONE MOTION] kicking up behind them. [SHOT SIZE], [CAMERA MOVE] alongside at speed. [LIGHTING], motion in the background, crisp foreground, [LENS].
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: Match [CAMERA MOVE] to the subject's speed — a tracking or handheld follow keeps it sharp while the background blurs. One motion, one move.
13. Peak-action moment
[SUBJECT] performs [ONE MOTION] in [SETTING], [ACTION DETAIL] flying through the air. [SHOT SIZE], [CAMERA MOVE] following the peak of the motion. [LIGHTING], high-speed slow-motion feel, [LENS].
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: [ONE MOTION] is the single hero beat — a leap, a splash, a drift. A single fast pan or handheld track reads as energetic without fighting competing motions.
Image-to-video
Wan's image-to-video is one of its best tricks: upload a start frame and describe only the motion you want added. Keep the described change small and physical, and anchor the source so it stays on-model. Wan 2.7 also accepts an optional end frame.
14. Bring a portrait to life
[Start frame: DESCRIBE YOUR IMAGE — e.g. a studio portrait of a young woman]
Animate the source: [SUBJECT] [ONE MOTION — blinks softly, a gentle smile forms] and [SECONDARY MICRO MOTION — a light breeze lifts a few strands of hair]. Keep the features, outfit, and colors identical to the source image. Static camera with a very slow push-in. Preserve the original [LIGHTING].
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: Describe only subtle motion in [ONE MOTION] so the person stays on-model instead of morphing. The "keep identical to the source" line is what anchors it.
15. Animate a product photo
[Start frame: DESCRIBE YOUR IMAGE — e.g. a clean product photo of a [PRODUCT] on a white background]
Add motion: the [PRODUCT] [ONE MOTION — rotates slowly clockwise] while the studio reflection shifts across its surface. Keep the product, colors, and logo exactly as in the source image. Static camera. Preserve the clean background and soft shadow.
SFX: [SFX].
Music: [MUSIC].
([ASPECT], 1080p, [DURATION])Swap: This turns a single catalog photo into a rotating clip with accurate branding. Lock the logo with the "exactly as in the source" instruction.
16. Animate a still scene or artwork
[Start frame: DESCRIBE YOUR IMAGE — e.g. a painted [LANDSCAPE] with clouds and water]
Bring the image to life: [ONE MOTION — clouds drift slowly] and [SECONDARY MICRO MOTION — the water shimmers] while keeping the exact [STYLE — painted brushwork / photographic] look and colors of the source. Static camera, very slow push-in.
SFX: [SFX].
Ambient: [AMBIENT].
([ASPECT], 1080p, [DURATION])Swap: Anchoring "keep the exact [STYLE]" stops Wan from photo-realizing artwork while it adds motion. Keep the described changes to gentle, ambient movement.
Vertical social 9:16
Phone-first skeletons framed for the vertical crop — subject centered, hook in the first second, and lean on native audio for voiceover and music.
17. Vertical talking hook
A [CREATOR] looks straight into the camera in [SETTING], framed vertically waist-up, [ONE MOTION] lightly. Static phone-on-tripod framing with a slow push-in. [LIGHTING], bright and clean, [LENS] phone look.
Dialogue: [CREATOR] says with energy, "[SHORT SPOKEN LINE]"
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
(9:16, 1080p, [DURATION])Swap: [SHORT SPOKEN LINE] is your hook — front-load it so it lands in the first second. A ring-light [LIGHTING] sells the native phone feel.
18. Vertical product demo / reveal
Hands [ONE MOTION — open the packaging / present the product] to reveal a [PRODUCT], framed vertically. Top-down close-up, slow push-in on the reveal. [LIGHTING], high-key, [COLOR] palette, [LENS].
Voiceover: "[SHORT SPOKEN LINE]"
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
(9:16, 1080p, [DURATION])Swap: A top-down reveal with one push-in is the highest-retention unboxing shape. Keep [SHORT SPOKEN LINE] to a single benefit or pitch.
19. Vertical satisfying loop
[SUBJECT] as [ONE MOTION — thick honey pours and spreads] catching the light, framed vertically for a seamless loop. Close-up, static camera. [LIGHTING], glossy highlights, high saturation, [LENS] macro.
SFX: [SFX].
Ambient: [AMBIENT].
Music: [MUSIC].
(9:16, 1080p, [DURATION])Swap: Pick a [ONE MOTION] that starts and ends similar so the clip loops cleanly. A static camera and one glossy action hold attention on repeat.
That is 19 skeletons across seven use cases. Keep the audio lines every time, hold to one motion and one camera move per fill, and set your ratio, resolution, and length in the app — never inside the prompt text. When a fill starts to feel natural, study the polished versions in the 35 best Wan prompts, keep the Wan prompt cheat sheet open while you write, or tighten your realism with how to prompt Wan for realistic video.
Frequently Asked Questions
How do the [brackets] work?
Each [BRACKET] is a slot you replace with your own detail. Swap [SUBJECT], [SETTING], [ONE MOTION], [CAMERA MOVE], [LIGHTING], and the audio slots for your scene, then delete the brackets. Keep the sentence order intact — the Subject + Scene + Motion + Lighting + Camera + Style sequence is what makes Wan follow the shot reliably.
Do I keep the audio lines in every template?
Yes. Native synced audio is Wan 2.7's signature feature, so every template ends with SFX, Ambient, Music, Dialogue, or Voiceover slots. Even a plain ambient line like "Ambient: quiet room tone" improves realism. Put spoken lines in quotation marks and keep them to one or two short sentences per clip.
Where do I set aspect ratio, resolution, and duration?
In the app or API before you generate — Alibaba Cloud Model Studio or a hosted Wan endpoint. The settings notes like (16:9, 1080p, 10s) are reminders, not text to paste. Wan has no inline flags like Midjourney's --ar, so writing the aspect ratio inside the prompt does nothing.
Why does each template use only one motion?
Wan follows a motion hierarchy. Give it one clear primary motion and it stays stable; stack several equal motions and the shot turns chaotic. Each template has one [ONE MOTION] slot and one [CAMERA MOVE] slot on purpose — pick a single action and a single move per shot.
How do I keep a character or product consistent?
Use image-to-video: upload a start frame of your character or product and Wan carries it into the motion. Wan 2.7 also accepts an optional end frame, so you can lock both the first and last image. This matters most for the product and talking-avatar templates, where labels and faces have to stay on-model.
How long should the dialogue in a template be?
Short. A Wan clip runs roughly 5 to 15 seconds, which is about one or two brief lines. Overwriting dialogue rushes the delivery and hurts lip-sync. Fill the [SHORT SPOKEN LINE] slot with a single natural sentence and let the visuals and ambient audio carry the rest.
Is Wan free to use commercially?
The Wan 2.7 weights are released under Apache 2.0, which permits commercial use without licensing fees, so clips you generate can go into ads and client work. If you run it through a hosted service, that platform's pricing and terms still apply, but the output itself is yours to use.
Can I combine templates into a longer video?
Yes. Generate each template as its own shot, then chain them by feeding the last frame of one clip in as the start frame of the next, or edit the clips together in any editor. Keeping the same [LIGHTING] and [SETTING] values across templates makes the cuts feel like one continuous scene.