Shorts Hook: Cold Open
Vertical. A person looks at camera and reveals a surprising object on a workbench, then a quick push-in on the object. Bright, native, photoreal. No generated captions.
These youtube video prompts cover Shorts hooks, channel intros, and faceless B-roll. Use them as youtube video prompt examples, then generate the clip in Media.io.
YouTube video prompts should make a plate you can edit. Choose Text to Video, paste the process shot, sting, or faceless desk beat, then generate a clip that can sit under a voiceover.
Open Media.io and choose Text to Video so your youtube video scene starts from a written prompt. Stay on this path for the Seedance 2.5 text-to-video workflow.
Paste one of these youtube video prompts into the box. Write the B-roll, intro sting, or faceless process you actually need under a voiceover, not a full episode. Use Inspire Me, duration, or resolution controls if you need a tighter test.
Generate a short clip first, review motion and artifacts, then download the take you want to cut. Results vary, so rewrite one note at a time instead of stacking extra effects.
Vertical. A person looks at camera and reveals a surprising object on a workbench, then a quick push-in on the object. Bright, native, photoreal. No generated captions.
Top-down landscape of hands assembling a small gadget. Calm, educational, photoreal, no face, no readable brand names. Good B-roll under a voiceover.
A fast landscape move through a stylized workspace that lands on a simple logo-safe negative space in the center. No invented logo, no tiny text, just a clean end hold.
Vertical drone-to-street descent over a coastal town at sunrise, then a handheld walk. Cinematic but short. No readable signs, no extra landmarks morphing.
Preserve the object. Camera: slow orbit and push-in on the moving part. Educational, photoreal, landscape or vertical. No extra parts appearing.
A cozy room at night, lamp, rain on the window, slow pan. Faceless, calm, loop-friendly. No people, no text.
Keep the person identity and gym outfit. Motion: one clean rep, then a hold. Camera: vertical three-quarter. Natural effort, no extra people, no morphing anatomy.
Landscape tracking shot of a city crosswalk in late afternoon, realistic, slightly documentary. No invented headlines, no logos you can read.
Over-the-shoulder of a person painting a miniature. Shallow depth of field, quiet, educational. Hands and work are the subject, face optional and stable.
Vertical slow move through a misty forest toward a creek. No people, photoreal, peaceful. Good for faceless commentary channels.
Preserve creator identity. Motion: a friendly wave, then a still hold with space on one side for later end-screen graphics. Camera: medium, landscape. No generated subscribe text.
Keep the product exact. Camera: landscape hero orbit, then a detail push. Review-style, photoreal, no fake UI, no extra products.
A cold open on a street: a look to camera, then a turn into an alley. Camera: handheld-but-stable vertical. Photoreal, no extra crowds.
Overhead craft assembly of a small object, hands only, a pause to show the result. Camera: locked top-down. Faceless YouTube, photoreal, no extra tools spawning.
A clean intro sting: a lamp, a desk, a slow push, a hold. Camera: medium. No fake logos, no extra text.
A train window landscape sliding, a reflection of a traveler. Camera: locked. Photoreal travel B-roll, no extra landmarks.
A macro of a small mechanism turning, then a pull to a wider desk. Camera: macro to medium. Photoreal, no extra gadgets.
A cozy room behind a talking space, lamp, plant, slow dust. Camera: locked wide of the set, no host required. Photoreal, no extra people.
A home yoga hold in a living room, breath, a slow transition. Camera: full-body. Photoreal, no extra equipment appearing.
A city intersection, traffic, a slow pan to a building. Camera: locked tripod feel. Photoreal, no extra signage you can read.
Over-the-shoulder of a laptop and notebook, a hand pointing once. Camera: OTS. Faceless, photoreal, no readable screen text.
A creek, stones, light through trees, no host. Camera: low tracking along water. Photoreal nature, no extra hikers.
A creator waves, then a still hold for an end screen. Camera: medium locked. Native studio, no generated subscribe buttons.
A review product on a desk, slow orbit, label locked. Camera: three-quarter. Photoreal, no extra SKUs.
Kitchen B-roll for a voiceover: a kettle, a pour, a window. Camera: medium. Photoreal, no extra cooks.
Strong youtube content prompts serve the edit: hook, B-roll, intro, or outro - not all four at once.
A kettle pour, a desk assembly, or a clean end hold is usable B-roll. Generate the plate, and leave subscribe buttons out of the frame.
Create process, travel, and atmosphere shots to sit under a voiceover.
Describe the real product or studio, then add an orbit or process beat for YouTube.
Generate 9:16 and 16:9 versions from related prompts.
End intros and outros on empty space you can brand in your editor.
Job-specific briefs: a Shorts cold open, faceless desk process, intro sting, or review hero.
Describe hands, places, macros, and camera path. Explicitly say no face if you do not want one.
Usually no. Generate a clean shot, then add branding in your editor so it stays sharp.
Generate clips, then edit. One prompt rarely produces a complete YouTube video with good pacing.
Yes. Write vertical hooks and generate in Media.io, then add captions and audio on YouTube.
Describe your real studio in the prompt, lock layout and gear in words, then generate a short clip in Media.io Text to Video. Output still varies.