Why this prompt works
This prompt succeeds by specifying a concrete subject (young Korean woman) with detailed actions (gentle skincare application) that create a narrative of self-care. The precise camera instructions (extreme close-up, wider shot) and lighting cues (soft daylight, cinematic lighting) guide the model to create consistent, professional-looking footage. Without these specific directions, the AI might produce distorted features, unnatural movements, or inconsistent lighting that would break the luxurious atmosphere.
The visual storytelling here creates a narrative of mindful self-care and tranquility through the progression from intimate skincare application to peaceful relaxation in natural light. The pairing of gentle hand movements with soft facial expressions creates a harmonious flow that reads well as a wellness moment, with the consistent subject appearance reinforcing the ritualistic nature of the skincare routine.
Technically, the prompt maps directly to model capabilities through specific descriptors like 'photorealistic,' 'realistic skin texture,' and 'smooth movements' that guide rendering quality. The 15-second duration suggests a complete micro-narrative, though the recommended 5s setting would need careful editing to maintain the intended pacing. The model-agnostic phrasing works well across platforms, with 'clean luxury atmosphere' effectively communicating the desired aesthetic.
This scene serves ideal social media clips or skincare ads focusing on product efficacy and lifestyle association. Pair with minimalist product close-ups or text overlays highlighting skincare benefits. For longer content, use as b-roll between testimonials or demonstrations. The natural lighting makes it versatile for day-parting in content calendars.
Prompt anatomy
- Subject
- young Korean woman with natural face
- Action
- gently touching cheek, applying skincare, massaging cheeks
- Setting
- bright white room with marble table and window
- Camera
- extreme close-up to wider shot with gentle motion
- Lighting & mood
- soft cinematic lighting, clean luxury atmosphere
Make it yours
- 1For different skin tones, modify to 'young East Asian woman' or specify ethnicity while keeping skincare focus. Avoid 'Asian' alone as it's overly broad.
- 2To extend duration, add more product application steps or slow down massage movements. For shorter clips, focus on just the bottle-to-face transition.
- 3Watch for AI inconsistencies in facial features between shots - add 'consistent facial features' as a directional cue if needed.
- 4For Instagram Reels, add text overlay in the final window shot with product benefits or a call-to-action.
Skincare Routine Cinematic
Create a 15-second photorealistic skincare video featuring a young Korean woman in a bright, elegant white room. Start with an extreme close-up of her natural face as she gently touches her cheek with her fingertip, showing realistic skin texture and soft daylight. Then show a clear skincare bottle placed on a white marble table as her hand slowly reaches toward it. Continue with her applying skincare gently to her face using both hands, keeping her facial features and appearance consistent. Show her closing her eyes peacefully while softly massaging both cheeks with her fingertips. Finish with a wider shot of her wearing a simple white dress standing beside a large window, while the sheer white curtains move naturally in the sunlight. Keep the movements smooth and realistic, with natural hands, realistic skin, soft cinematic lighting, clean luxury atmosphere, gentle camera motion, and no distortions or artificial-looking details.
How to use this video prompt
- Click "Use this prompt" — it loads into the text-to-video generator automatically.
- Pick a model, aspect ratio and duration (or keep the recommended settings above).
- Hit generate — free users get daily free generations, no watermark.
Tip: swap the subject or setting and keep the camera, lighting and style parts — that's the structure that makes these prompts work.
Need a matching cover image for your video? Reverse-engineer any image into a prompt →