Video & Animation

All animation knowledge for HyperFrames — atomic motion rules, multi-phase scene blueprints, scene transitions, broader motion-design techniques, AND the seven runtime adapters (GSAP default, plus Lottie, Three.js, Anime.js, CSS keyframes, Web Animations API, TypeGPU). Use for any motion or animation task: pick 2-4 rules and compose, or load a blueprint, or look up runtime-specific API (e.g. GSAP eases / Lottie player / Three.js mixer). Also covers auditing an existing composition's choreography (animation map) and 24 named text-animation effects. HyperFrames-native: single paused timeline, seek-safe, deterministic.

The HyperFrames composition contract — build one renderable project. Use for composition structure, the `data-*` timing attributes, `class="clip"`, tracks, sub-compositions, variables, framework-owned media playback, deterministic-render rules, and validation. Read before writing composition HTML.

Video and audio processing with FFmpeg. Use for format conversion, resizing, compression, audio extraction, and preparing assets for Remotion. Triggers include converting GIF to MP4, resizing video, extracting audio, compressing files, or any media transformation task.

Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated motion-first unit, including an animated title. Route fresh creation through hyperframes before using this skill.

Turn one topic into a knowledge explainer video that is easy to follow, good to look at and makes people want to watch to the end (on-screen text over music, the kind that took off under Douyin's #Vibe知识大赏), rendered with HyperFrames. Gives a path that works (research → copy → images → look and storyboard → render) and small tools for music editing, beat timing, sound effects and self-checks; the design is yours to make for each topic. Use it whenever someone wants a knowledge, science, history or how-it-works video — 'make a video explaining X', 'turn the history of X into a short video', 'explainer video', '做一条知识视频', '科普视频', 'Vibe知识大赏' — even if they don't name this skill. For a single short animation or just a new look, use code-video instead.

Edit videos in DaVinci Resolve Studio 21.1 with the analyses of Video Frame Expedition for DaVinci Resolve, a local application. Use when the user asks to look at a Resolve timeline and find what its clips contain, to build a new edit or modify an existing one from their rushes, to reframe clips (e.g. 16:9 to 9:16) using subject positions, to add markers or metadata from the analyses, or to pick shots for a montage. Needs the application running (its MCP server comes with this plugin) and, for what the application's own Resolve tools do not cover, the MCP server of DaVinci Resolve Studio.

Generate Mandarin and multilingual narration with Volcengine Doubao Speech 2.0. Use when creating Chinese voiceovers, when the user prefers Doubao/Volcengine/火山引擎/豆包 TTS, or when narration needs character-level timestamp metadata for subtitles.

YouTube Data API v3 complete wrapper - search videos, get video/channel/playlist details, get comments, download subtitles, etc. Supports filtering and sorting by time/views/rating and more.

Create AI avatar videos with precise control over avatars, voices, scripts, scenes, and backgrounds using HeyGen's v2 API. Use when: (1) Choosing a specific avatar and voice for a video, (2) Writing exact scripts for an avatar to speak, (3) Building multi-scene videos with different backgrounds per scene, (4) Creating transparent WebM videos for compositing, (5) Using talking photos as video presenters, (6) Integrating HeyGen avatars with Remotion, (7) Batch video generation with exact specs, (8) Brand-consistent production videos with precise control.

Edit existing videos using AI — remix style, upscale, remove background, and add audio via fal.ai's hosted video models.

Yatay YouTube konuşma videosunu (kamera kaydı, isteğe bağlı ayrı ekran kaydı) kaba kurgular. Harfiyen döküm çıkarır, duraklamaları ve art arda tekrarları sesten keser; Claude dökümü okuyup gerekçeli bir atma listesi yazar, mp4 üretilir, kesilmiş ses yeniden çözülüp her birleşim denetlenir. Şu durumlarda kullan - "videoyu kurgula", "kaba kurgu", "ham kaydı kes", "beklemeleri/duraklamaları kes", "tekrarları at", "takılmaları temizle", "jump cut", "keseceğim dediğim yerleri at", "kamera ile ekran kaydını senkronla", "ekran kaydının köşesine kamera koy", "sessiz yerleri hızlandır".

Turn a still image into a seamlessly looping ambient video for a web header or background, using MiniMax H3 in the local ComfyUI. Use when asked for a looping background/hero video, "make this image move", ambient motion from artwork, or when a generated loop visibly jumps, has no wind/atmosphere, or barely moves. Covers the settings that actually produce motion, how to close the loop, and how to measure whether it worked.

Make a 24-second square (1440², 60fps) looping product launch video where the whole story happens *inside the period of the wordmark* — iris opens onto a night photo, glass UI morphs ask → find → write → night → results → phone lock screen → proof → CTA, then closes back into the period beside the app icon. One canvas, pure seek(t), Playwright render with motion blur, beat-synced SFX, -14 LUFS. Arabic RTL template (Geist + Thmanyah when present, else IBM Plex Sans Arabic), config-driven. Use this whenever the user asks for "the night-shift video", "نفس فيديو النقطة", "فيديو إطلاق مربع", "إعلان يحدث داخل النقطة", a launch/promo video like the مستخرج one, or pastes a brief with "inside the period of the wordmark" / iOS glass panels / goo-split pills — even if they don't name this skill. Also covers getting the matched night/day photos (Higgsfield, local Codex image_gen, or ready prompts for Gemini/ChatGPT).

Sayı ve yazıyla hikâye anlatan editoryal hareketli grafik videosu üretir (Remotion) — X/Twitter gönderisi, Reels/Shorts, YouTube video arası. Hikâye JSON sahne listesi olarak yazılır (başlık, yazılan istem, sayaç, ölçer, sütunlar, karşılaştırma, liste, soru), 16:9 ve 9:16 render edilir, yumuşak whoosh + klavye + sakin müzik yatağı numpy ile sentezlenip eklenir, sonuç ölçülerek denetlenir. Şu durumlarda kullan: "X için animasyon/video", "hareketli grafik", "motion graphics", "sayılarla anlat", "sonuçları grafikle göster", "karşılaştırma videosu", "test sonucunu animasyonla anlat", "reels grafiği", "araya grafik", "Remotion".

Kodla (görsel üretim modeli OLMADAN) hikâye anlatan kısa mürekkep çizgi animasyonu üretir — X/Twitter gönderisi, Reels/Shorts, YouTube video arası ya da soğuk açılış. Görüntü PIL mürekkep motoruyla, müzik ve efektler numpy sentezi ile aynı zaman programından üretilir, sonuç ölçülerek denetlenir. Şu durumlarda kullan: "animasyon yap", "çizgi animasyon", "halimi anlat", "durumumu göster", "twitter için animasyon", "reels animasyonu", "araya animasyon", "çöp adam animasyonu", "kodla video", "one-shot animasyon".

Konuşan kafa kaydından (kamera, ses içinde; yatay 4K ya da dikey) dikey Instagram Reels / YouTube Shorts / TikTok videosu (1080x1920, 30 fps) üretir, "kart" formatında - üstte söylenen kelimede beliren sade arayüz kartları (sohbet balonu, liste, harf harf yazılan terminal, dosya ağacı, büyük sayı, gerçek ekran görüntüsü), altta yuvarlak köşeli yüz kartı, aralarında kelime kelime altyazı, kart geçişinde yumuşak whoosh, yazarken klavye sesi, sonda "yorumlara yaz" sorusu. Şu durumlarda kullan - "reels yap", "bundan reels çıkar", "shorts yap", "dikey video", "kartlı reels", "bilgi veren reels", "ipucu videosu", "üstte kart altta yüz", "yorumlara yaz diye bitsin", "tips reels".

Make and edit videos with Claude Opus 5.5, end to end, from a blank page or from the user's own footage — product and launch films, ads, talking-head and founder videos, interviews, explainers, tutorials and screen demos, event and trip montages, highlight reels, podcast clips, Reels/TikTok/Shorts, animated logos and intros, music and lyric videos, and changes to a finished video (shorter cut, vertical version, subtitles, translation, new music). Remotion is the backbone; Codex (image_gen) makes images, Seedance via OpenRouter animates them, ElevenLabs makes music and sound effects, ffmpeg and whisper handle footage and speech. It starts by asking what kind of video it is and everything that kind needs (platform, format, voice, music, rights, who must not appear, budget), then goes through direction, a script or paper edit approved piece by piece, production, revisions and delivery for each platform. Use it whenever someone wants to make, cut, edit, shorten, reformat, subtitle or improve a video, even if t...

Makes a finished marketing video for any brand or product (feature launch, product ad, social loop) in a precise motion-design style: one morphing UI shape, a real cursor, springs, a camera that zooms, a soundtrack cut on the beat, rendered to MP4 from code. Starts by collecting the person's own context (subject, UI screenshots, design system, inspiration video), so anyone can use it. Use for '/marketing-video', 'make me a marketing video', 'a video like this one', 'fais-moi une vidéo marketing', 'vidéo de lancement', 'vidéo pub produit'.

<!-- skill: animation-systems | version: 1.0.0 | updated: 2026-03-17 -->

<!-- skill: guimkt-hero-videos | version: 1.0.0 | updated: 2026-03-17 -->

<!-- skill: threejs-animation | version: 1.0.0 | updated: 2026-03-17 -->

Gemini-powered video inspection for quality analysis and debugging.

Single-script video production from project config.

Keep a bilingual document set honest — verify that a translated markdown file still matches its source by recompiling its Lua blocks on LuaJIT, comparing code fences after comment removal, matching heading structure and links, and comparing directory trees by structure instead of prose. Use after editing either side of a docs pair, or when translations start drifting from the code.