All skills
hoodini avatar

/yuv-video-director

@f412198

Yuval's all-in-one AI video pipeline. Turns an idea/script into a finished, on-brand MP4 by orchestrating HyperFrames (HTML→deterministic video render), Lottie (branded motion graphics), ManimCE (math / neural-network / concept animations), and a transcribe→approve caption flow — all wrapped in the YUV.AI Neon Phoenix brand via a frame.md. Use whenever Yuval wants to make, edit, or explain something as a video: promo, explainer, launch, social reel, "make a video about X", "explain X as a video", "neural network animation", "turn this into a video", captioned tutorial, 16:9 or 9:16. Triggers: video, explainer, promo, reel, manim, lottie, hyperframes, animation, "make a video", "explain ... as a video", מצגת וידאו, סרטון, הסבר וידאו. Routes each beat to the right engine, wraps in brand, self-verifies, and renders.

Use this Skill: https://skilld.dev/gh/hoodini/ai-agents-skills/yuv-video-director

This session only. Nothing lands on disk.

referencescomposition-pattern.md

≈659 tokens on demand. Your agent reads this file only when SKILL.md points to it.

The multi-engine composition pattern

The canonical index.html that combines all engines (this is exactly what shipped as the "How AI Learns" video — it lints 0/0, validates WCAG AA, renders clean).

Layer stack (z-index, NOT track-index)

z0   #net      — neural-net phoenix field (full duration, canvas, seekable)
z5   #mvid     — Manim render as <video class="clip" muted playsinline>  (body beat)
z10  .scene    — HyperFrames intro/outro clips (transparent bg, .scrim for contrast)
z50  #flash    — neon pink→cyan flash transition on each scene boundary
z60  #fade     — black, fades in only on the final 0.6s

The Manim video is opaque, so it covers the field during its window — run the field full-duration behind everything; it only shows in the intro/outro gaps.

Skeleton

<div id="root" data-composition-id="main" data-start="0" data-duration="D" data-width="1920" data-height="1080">
  <canvas id="net" width="1920" height="1080"></canvas>
  <div id="intro" class="scene clip" data-start="0"   data-duration="4.5"  data-track-index="1">…+ #pulse Lottie…</div>
  <video id="mvid" class="clip" data-start="4.5" data-duration="27.8" data-track-index="0"
         src="assets/scene.mp4" muted playsinline crossorigin="anonymous"></video>
  <div id="outro" class="scene clip" data-start="32.3" data-duration="5.2" data-track-index="2">…</div>
  <div id="flash"></div><div id="fade"></div>
</div>

Timeline (single GSAP timeline, registered on window.__timelines["main"])

  • Field: tl.to(proxy, {t:D, ease:"none", onUpdate:()=>drawNet(proxy.t)}, 0) — see ../assets/neural-net-field.js.
  • Lottie: load via lottie-web, push to window.__hfLottie (adapter seeks it). See lottie.md.
  • Each scene: gsap.from() entrances at the scene's start (offset ~0.2s, vary eases). No exit tweens — the flash handles the cut. Only the final scene may fade out (#fade).
  • Transitions: per boundary b, tl.to("#flash",{opacity:0.92,duration:0.22,overwrite:"auto"}, b-0.24) then back to 0 at b+0.02.

Fonts

@font-face { font-family:"Anton"; src:url("fonts/Anton-Regular.woff2") } (latin subset, bundled). Inter + JetBrains Mono auto-resolve — just name them. No Google Fonts <link>.

Rules carried from the hyperframes skill

Standalone root = data-composition-id div directly in <body> (no <template>). Timed elements need class="clip" + data-start/data-duration/data-track-index. Deterministic only. Always invoke the hyperframes skill when authoring; run the gates.md loop before done.

Source: SKILL.md on GitHub

1 warning3mo3 checks · Risk SAFE
  • Gen Agent Trust Hub3mo

    The yuv-video-director skill is a comprehensive and well-architected video production pipeline for AI agents. It orchestrates HyperFrames, Lottie, and ManimCE to create professional-grade video content with synchronized captions and branded motion graphics. The skill follows standard practices for asset management, using helper scripts to fetch legitimate resources like fonts and libraries from trusted sources. No malicious patterns, obfuscation, or data exfiltration risks were identified during the analysis.

  • Socket3mo

    No alerts

  • Snyk3mo

    Risk: MEDIUM · 1 issue

Signed by skilld at f412198. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Steadyupdated 4 months ago

README badge

README badge for hoodini/ai-agents-skills/yuv-video-director