yuv-viral-video — Claude Code skill for viral short-form video editing
yuv-viral-video — a Claude Code skill that turns a selfie recording or screen capture into a viral short in YUV.AI's house style. It outputs both 9:16 and 16:9 and saves every render with a _V<N> suffix so no version gets overwritten.
no license ★ 301 updated 2026-10-23
npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a claude-codeWhat yuv-viral-video does
The pipeline chains two skills. video-use handles transcription via ElevenLabs Scribe, picks word-snapped cut points, and assembles a base.mp4. hyperframes takes that base, adds liquid-glass cards built on CSS backdrop-filter, word-by-word karaoke captions, GSAP animations with back.out springy easing, and scene transitions, then renders to MP4 through Chromium. Both must run; neither alone produces the full result.
Typography comes from yuv-design-system (Phoenix Ink): Anton uppercase for English, Karantina 700 and Assistant for Hebrew. The speaker's face is never covered by overlays. Source must be a .mp4, .mov, or .mkv file — the skill is not intended for audio-only podcast edits.
Both companion skills are auto-discovered by Claude Code at session start. Set ELEVENLABS_API_KEY once in ~/Developer/video-use/.env; both video-use and hyperframes pick it up from there. Final renders go to ~/Documents/yuv-projects/videos/<slug>/renders/.
What it helps with
- Cut a selfie recording into a vertical reel with fast-paced edits
- Add word-by-word karaoke captions synced to the transcript
- Composite GSAP-animated glass cards over a background video
- Export 9:16 and 16:9 versions from a single source file
- Keep a versioned render history with
_V<N>suffixes
Example prompts for the agent
- Edit this video:
/path/to/file.mp4 - Turn
demo_screen.movinto a viral short - Make a 9:16 reel from
selfie.mp4in YUV.AI style
How to install yuv-viral-video
1Skills CLI (Vercel)
npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a claude-code2Manually: put the files in place
~/.claude/skills/yuv-viral-video/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/yuv-viral-video/
1Skills CLI (Vercel)
npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a codex2Manually: put the files in place
.agents/skills/yuv-viral-video/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a cursor2Manually: put the files in place
.cursor/skills/yuv-viral-video/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install hoodini/ai-agents-skills skills/yuv-viral-video2Manually: put the files in place
.github/skills/yuv-viral-video/SKILL.md1Skills CLI (Vercel)
npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a gemini-cli2Platform CLI
gemini skills install https://github.com/hoodini/ai-agents-skills --path skills/yuv-viral-videoManually: put the files in place
.devin/skills/yuv-viral-video/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about yuv-viral-video
How do I install yuv-viral-video in Claude Code?
Run npx skills add hoodini/ai-agents-skills --skill yuv-viral-video -a claude-code; the skill will then be available at ~/.claude/skills/yuv-viral-video/SKILL.md.
What does the yuv-viral-video skill do?
It takes a .mp4, .mov, or .mkv file, transcribes it, assembles cut segments, overlays glass cards with karaoke captions and GSAP animations in YUV.AI style, and renders both 9:16 and 16:9 outputs.
What is required before the skill can run?
An ELEVENLABS_API_KEY must be set in ~/Developer/video-use/.env. Both video-use and hyperframes read it from that single file.