Agent Skills

yyl-remotion-video

Build 16:9 spoken-video compositions as standalone Remotion projects with React, TypeScript, frame-based animation, estimated durations, and direct mp4 rendering. Use when the user wants video files, Remotion, mp4 export, frame-accurate motion, post-production voiceover, or theme-driven video templates; do not use for click-driven web presentations.

Install

npx skills add https://github.com/ttfake92-lab/skills --skill yyl-remotion-video
SKILL.md

YYL Remotion Video

Turn an article, script, notes, or clear outline into a Remotion video project. This skill is video-first: no click-driven web player, no web presentation/, no TTS synthesis inside this workflow.

Core Output

my-video/
├── article.md
├── script.md
├── outline.md
├── notes.md
└── remotion/
    ├── src/Root.tsx
    ├── src/Composition.tsx
    ├── src/chapters/<NN>-<id>/
    │   ├── <Chapter>.tsx
    │   ├── narrations.ts
    │   └── timing.ts
    ├── public/assets/
    ├── preview-stills/
    └── out/

Bundled skill resources:

themes/       # color / typography systems
templates/    # copyable Remotion TSX starting points
references/   # craft, theme, and template guidance

Execution Strength

Decide before writing files:

Strength Use When Behavior
Quick <= 450 Chinese chars, <= 60s, theme test, draft demo 1 chapter, few beats, direct implementation
Standard 1-4 min, normal publishable explainer script + outline, first chapter anchor, then continue
Strict > 4 min, brand/commercial/high-risk content fuller planning, more still checks, slower chapter review

Report in one line:

I will use <Quick/Standard/Strict> because <length / risk / target>.

Workflow

  1. Input Gate

    • Proceed only if the user gives source text, a script, notes, or a clear outline.
    • If they only give a topic, ask for material. Do not invent a full video.
  2. Content Files

    • Save source material to article.md when provided.
    • Save final spoken text to script.md.
    • Save development plan to outline.md.
    • outline.md plans chapters, beats, screen content, information pool, assets.
    • outline.md does not prescribe exact animation implementation.
  3. Theme Choice

    • Use only themes bundled in this skill: themes/command-film, themes/studio-white, themes/research-desk.
    • Do not reuse or import themes from web-video-presentation.
  4. Template Choice

    • If the user asks for a known layout or animation mechanism, read references/TEMPLATES.md.
    • Use templates/luxury-perspective-gallery for luxury dark portfolio, 3D perspective horizontal carousel, or premium agency gallery requests.
  5. Remotion Project

    • Reuse existing remotion/ if present.
    • Create only if missing:
      npx create-video@latest --yes --blank --no-tailwind remotion
      
    • If a selected template requires Tailwind, create the project with Tailwind support or add Tailwind before copying template files.
    • Install dependencies only if remotion/node_modules is missing or Remotion commands fail because dependencies are absent.
  6. Timing

    • Do not synthesize audio.
    • Estimate durations from narration and visual complexity.
    • Default fps = 30, width = 1920, height = 1080.
    • Store text in narrations.ts; store estimated durationFrames in timing.ts.
  7. Implementation

    • Read references/REMOTION-CRAFT.md before each chapter.
    • Animate with useCurrentFrame(), interpolate(), Easing, and Sequence.
    • Do not use CSS transition / CSS animation / GSAP for primary animation.
    • Tailwind may style static layout, but Tailwind animation utilities are not allowed.
  8. Verification

    • Run:
      cd remotion
      npm run lint
      
    • Render at least one still per chapter, or 2-4 stills for a short full video:
      npx remotion still <composition-id> --frame=<frame> --scale=0.5 --output=preview-stills/<name>.png
      
    • Render mp4 when the stills look correct:
      npx remotion render <composition-id> out/<name>.mp4
      
  9. Notes

    • Update notes.md with fps, resolution, total frames, estimated duration, checked stills, rendered mp4 path, and that audio/subtitles are post-production.

Theme Selection

Theme Best For Feel
command-film agents, terminal, automation, developer tools dark cinematic command room
studio-white SaaS demos, launches, courses, product tutorials clean white product studio
research-desk essays, reports, papers, business analysis calm analytical desk

For full theme notes, read references/THEMES.md.

Template Selection

Template Best For Feel
luxury-perspective-gallery premium portfolio, product showcase, 3D horizontal carousel dark cyber-minimal agency slider
charity-doodle-poster nonprofit posters, hand-drawn animal heroes, line-boil editorial web effects retro ink-brush stop-motion poster

For full template notes, read references/TEMPLATES.md.

Remotion Rules

  • Use Composition in Root.tsx.
  • Use Sequence to place beats on the timeline.
  • Use useCurrentFrame() + interpolate() for motion.
  • Place static assets in remotion/public/assets/; reference with staticFile().
  • Still checks are required before final render.
  • The final mp4 is silent unless the user later supplies audio in post-production.

Related skills

video-editgenmedia-labs715KEdit existing video on RunComfy — this skill is a smart router that matches the user's intent to the right edit model in the RunComfy catalog. Picks Wan 2.7 Edit-Video (general restyle / background swap / packaging swap, identity + motion preservation), Kling 2.6 Pro Motion Control (transfer precise motion from a reference video to a target character), or Lucy Edit Restyle (lightweight identity-stable restyle / outfit swap). Bundles each model's documented prompting patterns so the skill gets shai-video-generationgenmedia-labs714KGenerate AI videos on RunComfy via the `runcomfy` CLI — a smart router across the full video-model catalog: HappyHorse 1.0 (Arena #1, native in-pass audio), Wan-AI Wan 2-7 (open weights, audio-driven lip-sync), ByteDance Seedance v2 / 1-5 / 1-0 (multi-modal cinematic), Kling 3.0 / 2-6, Google Veo 3-1, MiniMax Hailuo 2-3, ByteDance Dreamina 3-0. Covers text-to-video (t2v), image-to-video (i2v), and Veo's video-extend endpoint. The skill picks the right model for the user's intent (Arena-#1 qualitai-musicgenmedia-labs714KGenerate AI music on RunComfy via the `runcomfy` CLI — a smart router across the music-model catalog. Routes to ElevenLabs AI Music Generation (premium 44.1 kHz stereo vocal tracks, 5 s–5 min, $0.0083/s) and ACE Step / ACE Step 1.5 (StepFun-AI open-weights, tag-driven composition, multilingual lyrics, $0.0002–0.0003/s, ~27× cheaper), plus ACE Step audio-inpaint (regenerate a time range inside an existing track) and ACE Step audio-outpaint (extend a track before or after). Picks the right model fimage-to-videogenmedia-labs713KAnimate any still image on RunComfy — this skill is a smart router that matches the user's intent to the right i2v model in the RunComfy catalog. Picks HappyHorse 1.0 I2V (Arena #1, native audio, identity preservation) for general animations, Wan 2.7 with `audio_url` for custom-voiceover lip-sync, or Seedance 2.0 Pro for multi-modal animation from image + reference video + reference audio. Bundles each model's documented prompting patterns so the caller gets sharper output without burning iterat

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers