LLM Skills
HiggsfieldTurn one winning video into dozens of variantsDiscover GenjutsuSponsored
~/catalog/video generation//SKILL
Video generationGitHub source

Video animations (Hyperframes/Remotion)

/SKILL

Animations in image sequences compatible with Hyperframes / Remotion, with automatic playback

nexu-ionexu-io
★102.4k
May 22, 2026
Apache License 2.0
// skill content

--- name: video-hyperframes zhname : "Hyperframes Video Script" enname : "Hyperframes Video" emoji: "🎞️" description: "Hyperframes / Remotion-compatible sequential-frame animation that plays automatically" category: video scenario: video aspecthint : "1920×1080 (16:9)" recommended: 5 tags: ["video", "hyperframes", "remotion", "video"] exampleid : sample-hyperframes-workflow examplename : "Hyperframes · AI-workflow Video" exampleformat : markdown exampletagline : "8-frame autoplay, includes progress bar + metadata" exampledesc : "Cinematic animation script, can be fed directly into Remotion to create an MP4" examplesourceurl : "https://github.com/heygen-com/hyperframes" examplesourcelabel : "heygen-com/hyperframes" od: mode: video surface: video scenario: video featured: 0.13 upstream: "https://github.com/nexu-io/html-anything" preview: type: html entry: index.html reload: debounce-100 designsystem : requires: false exampleprompt : "Use the 'Hyperframes Video Script' template to turn my content into a 'Hyperframes / Remotion-compatible continuous-frame animation that plays automatically.' Preserve the template's visual signature, use real content and data, and avoid lorem ipsum and placeholder images." --- 【Template: Hyperframes Video Frames】 - Output N consecutive<section class="frame"> , eachw-[1920px] h-[1080px] ; N is determined by the information density of the 【User Content】 (short scripts start at 6:10 frames; longer scripts should have more, with each frame carrying only one shot/concept). - Each frame should express one shot/concept: text + visual composition (centered composition / golden ratio / rule of thirds). - A hidden tag at the bottom of each frame (<!-- frame:N duration:3000 transition:fade --> ) is provided for subsequent Remotion/Hyperframes rendering scripts to read. - Add a JavaScript autoplay script at the top: advance to the next frame every 3 seconds; also supports click and arrow key controls; display a progress bar in the corner. - Frame 1 is “hook ” (one data point / one counterintuitive fact / one question); frames 2:N present the argument; the final frame contains the conclusion and a CTA. - Use a very large font size (text-9xl); limit each frame to a single sentence:avoid clutter. - Use a consistent, cinematic color scheme (dark background + 1 neon accent color). - The final output should include a short comment (<!-- HYPERFRAMES_META: ... --> ) containing metadata for each frame (JSON ), including duration, transition, andsceneSummary , for later conversion to Remotion.

// original public source
nexu-io/open-design
/skills/video-hyperframes/SKILL.md
Independent project, not affiliated with Anthropic. This skill remains the property of its original author.
// install this skill
Paste this command in your terminal at the root of your project:
mkdir -p .claude/commands && curl -o ".claude/commands/SKILL.md" "https://raw.githubusercontent.com/nexu-io/open-design/main/skills/video-hyperframes/SKILL.md"
Then in Claude Code, type /SKILL to activate it.
open_in_newOpen original source
// save
Save available after sign in.
loginSign in to save
// information
Creatornexu-io
Stars★ 102.4k
LicenseApache License 2.0
UpdatedMay 22, 2026
Format.md
AccessFree
HiggsfieldSponsored

One video, ten variants, zero reshoots

With Genjutsu, take an existing video and swap the character, the product or the set while keeping the motion and the edit. Perfect to scale an ad that works or adapt it to each market.

Discover Genjutsu
// similar

Skills Video generation

View allarrow_forward