SocialToPrompt Create your own prompt

Liquid Glass Figures Reach Across Worlds in Moss

A vertical macro video showing translucent glass humanoid figures reaching toward each other across moss, capturing a dreamy, melancholic mood with ambient music.

AI video prompt

### Style
Visual Texture: High-contrast macro photography blending velvety moss with glossy, fluid glass forms. Lighting Quality: Soft, directional light creating specular highlights on the liquid bodies against a dark, creamy bokeh. Color Palette: Deep greens, earthy browns, and silvery transparent white. Atmosphere: Dreamy, ethereal, melancholic, and weightless.

### Cinematography
Camera: Static, locked framing mimicking a vertical screen capture. Lens: Macro focus emphasizing texture details. Lighting: Soft naturalistic key light casting gentle highlights. Mood: Serene, introspective, and distant.

### Scene Breakdown

**Scene 1 (00:00s - 00:04s):**
**Actions:** A vertical screen recording begins, showing a social media post by 'Pst Bhushan Datta'. The top status bar reads '1:30'. A black pill-shaped screen-recorder timer advances: 00:00, 00:01, 00:02, 00:03, 00:04. The central video displays a macro scene of soft brown-and-green moss. Two translucent, liquid-glass humanoid figures appear. The LEFT figure is suspended in mid-air, twisting and reaching downward. The RIGHT figure balances on a tilted grass blade, reaching upward. They do not touch. The UI shows: heart '3.5k', comment '62', 'Profile' button, caption 'You and I are in different worlds🌍💫', and reaction chips 'Nice 👏 Love it ❤️', 'View all comments'.
**Dialogue:** None.
**Background Sound:** Gentle, melancholic ambient/instrumental music with slow tempo synth/piano tones.

**Scene 2 (00:04s - 00:09s):**
**Actions:** The screen recording continues. The timer advances: 00:04, 00:05, 00:06, 00:07, 00:08, 00:09. Inside the macro scene, the LEFT figure leaps higher, performing a mid-air flip/contortion with limbs splayed. The RIGHT figure shifts balance on the grass blade. Both stretch arms toward the invisible boundary between them. The UI elements remain identical except for the timer. The camera remains static; lighting and color palette stay consistent.
**Dialogue:** None.
**Background Sound:** Continuous soft, slow-tempo ambient/instrumental music; wistful mood.

**Scene 3 (00:09s - 00:13s):**
**Actions:** The final segment of the screen recording. The timer advances: 00:09, 00:10, 00:11, 00:12. The LEFT figure drifts downward through the air; the RIGHT figure remains perched on the grass blade. Their outstretched hands hover close together but remain separated. The UI elements (status bar '1:30', back arrow, like/comment counts, poster name, caption, reaction chips) are unchanged apart from the timer. The clip ends as the figures maintain their separation.
**Dialogue:** None.
**Background Sound:** The gentle ambient/instrumental track persists to the end with melancholic, dreamy mood.

Turn any video into a prompt like this

Paste a TikTok, Instagram, YouTube or X link and get a detailed, model-ready prompt for Veo, Seedance, Kling, Wan and more.

Extract a prompt free