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.

referenceslottie.md

≈762 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Lottie — branded motion graphics (live, lottie-web)

Use for designed brand motion: logo stings, stat reveals, icon pops, the neural pulse. For plain kinetic text, prefer GSAP.

The trap (read this)

diffusionstudio/lottie (text-to-lottie) previews in Skottie (Skia). HyperFrames renders in lottie-web. JSON that's perfect in Skottie can render blank in lottie-web (different AE feature support, group-wrapping rules). Always screenshot-verify the Lottie in the actual HyperFrames render — not just the Skottie preview.

Authoring a Lottie that renders in lottie-web

  • Plain Bodymovin JSON (v, fr, ip, op, w, h, assets:[], layers:[]).
  • Shape layers (ty:4) with a group (ty:"gr") → el/sh + fl/st + a trailing tr transform.
  • Animate via the layer transform ks (scale s, opacity o) with keyframes carrying i/o bezier handles — lottie-web needs them.
  • Colors are normalized RGBA arrays: pink #FF1464 → [1, 0.0784, 0.3922, 1], cyan #00E5FF → [0, 0.898, 1, 1].
  • Template: ../assets/neural-pulse.json (cyan core + expanding pink/cyan rings, loops, renders in lottie-web — verified).

Wiring into HyperFrames (the contract)

<div id="pulse" style="width:200px;height:200px"></div>
<script src="https://cdnjs.cloudflare.com/ajax/libs/bodymovin/5.12.2/lottie.min.js"></script>
<script>
  window.__hfLottie = window.__hfLottie || [];
  const a = lottie.loadAnimation({
    container: document.getElementById("pulse"),
    renderer: "svg", loop: true, autoplay: false,   // autoplay MUST be false
    path: "assets/neural-pulse.json",                // local asset, not a remote URL
  });
  window.__hfLottie.push(a);                          // the adapter seeks every entry
</script>

HyperFrames' lottie adapter seeks each __hfLottie entry to composition time via goToAndStop(t·1000, false). Don't call .play(). Keep the container size stable. For dotLottie (.lottie), use DotLottie + setCurrentRawFrameValue; it must be pushed manually.

⚠️ Seek-clamp gotcha (sections later than the lottie's duration)

The adapter seeks to COMPOSITION time — if a section starts at 55s but the lottie's total duration is 35s, lottie-web clamps to the LAST frame (often an invisible cycle-start pose). Fix: wrap goToAndStop with a deterministic modulo (see the L() helper in teaser-explainer.md — use it in every composition; harmless when not needed).

Content-synced beats + transparency (the standard)

Lotties must illustrate the SPOKEN content at the exact moment it's said — director-timed beats, fully transparent (no chip/box; drop-shadow only), persistent+continuous design so any window reads. Six production-grade content animations + their generator are bundled: ../assets/gen_content_lotties.py. Verify transparency by rendering the test grid over an odd-colored background. Full method: teaser-explainer.md.

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