Video & Animation

Make vertical 3D cartoon explainer shorts (30-60 s, 9:16) on any topic. Big-headed toon characters live a story told to "you"; painted faces swap through blinks; the camera never stops; every spoken number lands as a physical object; editorial captions (bold sans capitals + italic serif words in brand red, fonts included); a topic question at the top; a designed sound on every action. Use for history, science, tech, "how it works" and "what happens when" explainers.

Make motion-graphics videos from a script, from short reels to long videos up to 4K: config, script, storyboard with playable preview, checks, render, audio, publishing texts and thumbnails. Use when Petr wants a video, reel, short, explainer, script or storyboard.

לוח סגנונות לסרט אנימציה חדש. הסקיל מציע 3–4 סגנונות מתוך שבעה שנבנו ונבדקו בסרטים אמיתיים (ציור בחול, לוח מצויר, צעצועי פלסטלינה, צבעי מים כמו בספר, לבד ונייר גזור, מילים חיות, טיפוגרפיה תיעודית), ומראה לוח תמונות אחד עם דוגמה לכל סגנון וסרטון קטעים קצר. אם רוצים, הוא מראה גם את הסיפור עצמו מצויר בכל סגנון: דרך Codex כשיש מנוי ChatGPT, או עם דף פרומפטים שהמשתמש מדביק בעצמו ב-ChatGPT, ב-Gemini או ב-Copilot. הוא מציע גם וידאו AI במודלים החדשים (Seedance 2.5, Kling 4.0, Veo 3.1, Gemini Omni) ומסביר כמה זה עולה לקליפ, לשוטים מרכזיים ולסרט שלם. הפעל בתחילת כל סרט חדש כשהסגנון עוד לא נבחר, וכשאומרים "איזה סגנונות יש", "תציע סגנונות", "לוח סגנונות", "בורד", "מודבורד", "style board", "באיזה סגנון לעשות", "מה האפשרויות לסרט", "תראה לי איך זה ייראה", "וידאו AI", "כמה עולה וידאו", "Kling", "Seedance", "Veo". הפעל גם כשמביאים צילומי ספר, סיפור, שיר או רעיון לסרט ולא אומרים סגנון.

Auto captions for videos, made on the user's own computer by their AI agent: transcribes speech with a time for every word (free local Whisper by default, or the user's own Groq or Deepgram key, or existing subtitles), proofreads names and numbers, burns creator-style captions into the video (8 presets: word highlight, bold, pill, reveal, two-line, clean subtitles, minimal, one word) in any shape (9:16, 4:5, 1:1, 16:9, crop or blurred background), writes SRT, VTT and a transcript, translates the captions into other languages with the same timing, shows a review sheet with each app's buttons drawn on, and posts or schedules the captioned videos to TikTok, Reels, Shorts, LinkedIn, Facebook, X, Threads and Bluesky through WoopSocial after a yes. Use when someone says "auto captions", "add captions", "add subtitles to my video", "caption generator", "subtitle generator", "burn in subtitles", "TikTok captions", "make an SRT", "translate my subtitles" or "caption this folder of videos". No watermark, no per-minute

把动作描述、分镜或口播文字做成无字幕、无配乐、无配音的 GIF / 视频素材。使用 GPT Image 原生 GIF，按 69 种风格资料保持人物与画风；用户可选择保存单个长 GIF 或静音 MP4，保留原始短 GIF 和来源清单。

Drive Blender from Claude Code through an already-connected Blender MCP to build, clean up, validate and export game-ready 3D assets (props, modular kits, blockouts, batch cleanup, GLB/FBX export for Godot, Unity, Unreal, Roblox or three.js). Enforces an inspect → brief → build → look → fix → check → export → verify-in-engine loop with measured done-criteria. Use when the user asks Claude to make or fix a 3D model, prop, kit or scene in Blender, or to export assets for a game engine.

Use when asked to make a short motion-graphics ad or promo film (9:16 Reels/TikTok/Shorts, 4:5 feed). Builds a seek-safe HTML/GSAP film with every move locked to voice-over words and music beats, gates it by script, and learns each run. ElevenLabs optional.

手順書（Markdown やテキスト）から、イラストの手順動画（MP4。Swift で描くアニメーション＋ナレーション＋字幕）を作る。左上にタイトルと「STEP 01」工程バッジ、下部に字幕バーと進捗バー。型は 5 種類：部屋に立つ人物が道具を持って作業する「部屋」（拡大カードで手元）、スマホ・ブラウザの画面をタップ／クリックで操作する「画面操作」、真上から作業台に手が伸びる「手元俯瞰」、線画の等角図と矢印で見せる「組み立て説明書」、1080×1920 の「縦型」。料理・家事・掃除・アプリや Web サービスの操作・家具の組み立て・機器の設置・手入れなどの「やり方」「作り方」「手順」「マニュアル」「使い方」「How to」「チュートリアル」の動画や、手順書・説明書・メモを動画にしたい依頼、既にこの skill で作った動画の手直し・多言語版（日本語・英語・中国語・ベトナム語など）の追加に使う。ナレーションは既定では macOS の音声で、書き出した動画は個人的・非商用の利用に限る（Apple の使用許諾）。公開・業務に使うなら Google・Azure・Amazon・OpenAI のクラウドの TTS（環境変数にキーを入れるだけ）か、コマンドでつないだ TTS に切り替える。

Turn a voice note or audio clip (no camera footage needed) into a ~1-minute vertical explainer for TikTok / Reels / Shorts that PROVES what the speaker says - full-screen screenshots of real websites zoomed onto the exact line (with highlighter), company logos stamped in with a camera click, a step-by-step architecture diagram a lay person can follow, plain-English cards for every jargon word, a "ka-ching" on every successful money/transaction step, a hook first and a follow CTA last, word-by-word subtitles anchored to the bottom. Use whenever someone shares an audio recording / voice memo / podcast snippet / mp3 / m4a and asks for a TikTok, short, reel, explainer, "1 minute video", "video with evidence / receipts / sources / screenshots", "show the websites", "add logos", "make it easy for a lay person", or wants to explain how a product, fintech, payment flow, app or company works behind the scenes. Also for re-cutting or restyling an explainer made this way (zooms, sounds, logos, subtitles). Not for editin

录屏/讲解视频全自动剪辑（不依赖剪映，ffmpeg 直出成片）：faster-whisper 转写带词级时间戳 → AI 纠错字幕 → 静音/卡壳/重复/口误自动剪除 → 讲错处画面叠更正条 → 一次渲染出片。当用户说"帮我剪这条视频"、"剪掉停顿/卡顿/重复"、"自动剪辑讲解视频"、"视频去气口出成片"、"给视频加更正字幕"，或给出一个录屏/讲解视频文件要求剪辑处理时，务必使用本 skill。要"回写剪映草稿/在剪映里继续编辑"的走 jianying-subtitle-fixer；本 skill 的产出是直接可发布的成片。

Turn a brand brief or visual reference into an original motion video. Use for promos, product videos, showreels, kinetic typography, title sequences, storyboards, reference analysis, or prompt handoffs. Research Prompt Motion when available, adapt the craft to any brand and existing production stack, and support multilingual and right-to-left copy.

Build videos as React code with Remotion - compositions, sequences, springs, interpolation, audio sync, data-driven props, studio preview and rendering to MP4. Use for explainers, product demos, release notes, data charts, branded templates, or any video that should be reproducible from code.

Turn a voiceover or narration audio file into a finished, fully coded 2D animated video (MP4) with no video-generation models and no stock footage - every frame is drawn by code (Canvas 2D), cut to the exact word timings of the narration, and rendered with headless Chromium + ffmpeg. Use this whenever the user hands over an audio file (voiceover.mp3, lektor, narration, podcast clip) and wants an animation, animated short, explainer, motion graphics, "Kurzgesagt-style" video, coded animation or b-roll made from scratch for it; when they say things like "animate this voiceover", "make a video for this audio", "zrób animację do tego lektora", "zanimuj ten voiceover"; and also when they have only a script and want a coded animated video (make or request the audio first). Prefer this over one-off improvised HTML/CSS animation whenever the result has to be a real video file synced to speech.

Create, edit, translate and burn in captions and subtitles - transcribe speech with Whisper, write SRT/VTT/ASS files, style them, and hard-burn word-highlight captions for Shorts, Reels and TikTok. Use for any captioning, subtitle, transcript or accessibility request.

Use when the user asks for a 3D promo, teaser, launch, "coming soon", "what's new" or feature video for their iOS app, with a real 3D iPhone and a mascot character that hops around and presses the app's UI, or a revision or localized cut of such a video.

Use when animating a 2D game character from a single painting on a skeletal rig (Spine or similar), such as cutting the painting into parts, authoring an idle, walk, run or combat move as code on the rig, making a reference video for timing, painting hidden parts or swappable gear, gating captured clips with scripted checks, or planning a team of AI models for that work.

Guided, step-by-step workflow for making a short motion-graphics / SaaS / app / product ad with Claude. Interviews the user for everything it needs (product, length, assets, ending, level, sound), enforces storyboard-first approval, then hands the build to the saas-animation skill and finishes with sound design, voiceover and music. Use when someone says "make an ad / promo / motion graphic / app video / product video", or invokes /motion-ad-director.

Fully automated talking-head video editor. Turns raw camera footage into a finished video - cuts silences, punch-in zooms on the face, colour grade, voice mastering, animated stat/step/comparison/quote cards at the right moments, optional b-roll - with one command. Works for vertical reels/shorts and landscape YouTube. Use when the user drops footage and says edit my video, cut the silences, make this a reel, finish this video, or types /auto-video-edit.

Turn an app or product into a short promo/ad video built from HTML animations recorded to MP4 clips, with real screenshots, video frames and data from the product, plus a shot-by-shot script, voice-over lines, cut order with trims and speeds, music (ElevenLabs Music, a generated licence-free bed, or the user's own track), an optional ElevenLabs voice-over placed on the cuts, and a rough-cut preview. Use this whenever someone wants a social ad, promo video, product demo video, explainer clip, launch video, feature teaser or "video for our website/Instagram/YouTube/LinkedIn" about their software, even if they only say "make a video showing how the app works" or "I want to record some clips for an ad". Also use it to redo or add a single shot, re-time clips, or write the voice-over and editing instructions for an existing set of animation clips.

Make a self-introduction film for a creator on X (30 s by default, as long as what it says), starring the creator as a smooth 3D vinyl toy that pops out of their profile page, rendered in Blender Cycles. Use when someone asks for a self-intro or intro video, 自我介绍视频, 自我介紹短片, 个人介绍片 or a creator intro with a 3D toy. Austin's fork: also for Threads profiles in Traditional Chinese (自我介紹影片、公仔介紹片、Threads 自介影片、做一支跟 ChrisHKAI 一樣的影片).

Create videos in which a camera approaches a framed painting in a room, the depicted scene comes alive, and optionally the camera pulls back. Use for оживающие картины and subsequent videos in the Paintings series; not for generic films or standalone image edits.

Art direction and motion-design craft for animation projects — creative briefs, storyboards, style frames, colour and type systems, composition, timing and easing, transition vocabulary, sound direction, accessibility, and right-to-left (Persian/Arabic) typography. Use before building scenes, when a video looks amateur or generic, when choosing palette and fonts, when a request says "make it feel premium/cinematic/modern", or when text, rhythm, or hierarchy is not landing. Tool-agnostic; pairs with blender-animation, remotion-motion-graphics, and manim-animation.

Produce mathematical, scientific, and technical animations with Manim Community — equations, proofs, graphs, coordinate systems, geometry, algorithms, data structures, and precise diagram transformations. Use when correctness of mathematical relationships is central, for educational explainers, research or conference visuals, and for animated derivations where every element must be exactly placed. Covers installation with uv, scene design, LaTeX/MathTex, camera and 3D scenes, rendering quality presets, and compositing with other tools.

Produce a finished, narrated Loom-style screen-recording video (MP4, 1080p, the user's own cloned voice via Fish Audio, captions, no webcam) that demonstrates a website or web app. Use when the user runs /loom:video or asks to record, demo, or make a walkthrough/tutorial video of an application.