Video & Animation

Creates or edits visual content using only the production modules a request needs - Blender 3D modeling, materials, lighting, still renders and 3D animation; 2D motion graphics, kinetic typography and HTML/SVG/Canvas scenes rendered to MP4; product ads and app demo videos; music-synced brand showreels; footage editing (trims, silence and mistake removal from transcripts, Whisper transcription, subtitles, compositing, audio mix and loudness); and animated HTML slide decks or PowerPoint (.pptx) files with native animations. Use for anything from a single model, shot or slide fix to multi-scene productions, whether the request is in English or Korean (블렌더, 3D 애니메이션, 모션 그래픽, 제품 영상, 앱 시연, 쇼릴, 영상 편집, 무음 제거, 자막, 발표 자료, 파워포인트).

Show a pull request instead of its code — render a 25–40 second video (what changed, how it works now, before and after, the fix running, and the review key and tests together in one scene) and post it with the PR link to Slack. Use after opening a PR when the repository's delivery profile has a Demo section, when asked for a demo, a video, a recording or a visual summary of a change, or to notify someone that a PR is ready.

Edit real footage with the local video-tools MCP (ffmpeg) - trim, assemble, jump-cut pauses out of talking-head video, burn captions, lay a ducked music bed, reframe to 9:16 / 1:1 / 4:5, normalize loudness and export per platform. Use when the user hands you video files and says "cut this", "edit this", "remove the pauses / dead air", "add captions/subtitles", "add music", "make it vertical", "make a short from this", "export for TikTok/YouTube", or "clean up the audio".

Use when the user wants a narrated video made — data explainers, documentaries, short-form vertical clips — from a topic, article or dataset. Runs a staged, gated pipeline (research → proposal → script → storyboard → assets → voice → compose → render → review → publish) with sourced facts, sentence-level TTS timing, Remotion rendering, automated QC and a publish kit.

Edits filmed talking-head footage into release-ready vertical shorts (1080x1920) and 16:9 YouTube long-forms with ffmpeg and Whisper. Keeps exactly one clean read of every script line (no stumbles, restarts or double reads), never leaves a flat single shot (punch-ins, reframes, B-roll, motion graphics), captions from the finished audio, normalises loudness to -14 LUFS, makes a real-frame cover and runs a QA gate that re-transcribes the render and refuses delivery until it passes. Syncs extra cameras by sound and writes YouTube chapters. Use when someone says "edit my video", "cut this clip", "remove my retakes", "edit out the fluffs", "add captions to my video", "make a short from this footage", "sync my two cameras", "make a thumbnail from my video", or hands over raw footage and a script.

Trigger: explainer 3D, animacion tipo Stupy Doo, video explicativo cartoon, cel shading, toon animation, Blender animation, video vertical animado. Build and render cel-shaded 3D explainer videos from a storyboard.json with headless Blender.

Create professional, After Effects-style video ads / commercials / promos / Shorts for any product, service, app, event, brand or channel, rendered to real MP4 with Remotion (React) — with kinetic typography, 3D, glitch and transition effects, synced narration and sound design. Use whenever the user asks for an ad, commercial, promo video, motion graphics, product teaser, YouTube Short / TikTok / Reel, news-style explainer, kinetic typography, logo reveal, channel logo or banner — including Hebrew requests like "פרסומת", "סרטון פרסומי", "מושן גרפיקס", "אנימציה לפרסומת", "טיזר", "שורט", "סרטון לטיקטוק".

Make multi-scene animated stories, mini cartoons, and motion-design GIFs or MP4s with story beats, character references, short generated takes, local animated titles, and continuity review. Use for longer narrative edits, not just extending a short loop.

Método para hacer vídeos demo verticales de un SaaS o servicio (1080×1920, 25-35 s) con aspecto de producto profesional, en Remotion y three.js. Dispositivos 3D modelados en código con la interfaz real y nítida, rótulos legibles sin sonido, neumorfismo, transiciones conectadas, apariciones «mágicas», un móvil que gira y suelta monedas 3D, voz opcional, y efectos y música de ElevenLabs sin silencios muertos. Incluye una plantilla completa sin voz que se rellena con un archivo de ajustes, la receta de 27 efectos de interfaz y scripts para generarlos, medir volumen y silencios, nivelar la música y cortar teclas. Úsala cuando el usuario pida un vídeo demo, un anuncio vertical, un reel o un short de su producto, una explicación animada de una app o una web, o un vídeo para captar clientes, aunque no diga «Remotion» ni «3D». También sirve para revisar uno o para hacer solo su sonido (voz, efectos o música).

Build a 60-second motion-graphic product/brand video (showreel, promo, launch teaser, explainer) as a single self-contained animation, optionally exported to MP4. Use whenever the user asks for a motion graphic, product video, showreel, promo reel, animated brand film, or "1 minute video" for a product, company, or idea, even if they don't say "skill". Token-efficient by design.

Cria vídeos narrados com animação 2D, tudo em código no Remotion e sem ferramenta externa de vídeo. Serve para qualquer pessoa. Primeiro procura um brandbook de animação (um "universo" com personagens, cenários e narrativa); se não existir, conduz uma entrevista detalhada, uma pergunta por vez com opções recomendadas, e gera o brandbook em HTML mais a ficha de cada personagem. Depois usa esse brandbook como referência para escrever o roteiro, gerar a narração e montar o vídeo animado. Pode ter nenhum, um ou vários personagens. Use SEMPRE que a pessoa pedir vídeo animado, vídeo narrado com animação, avatar 2D, personagem animado, brandbook de animação, universo de personagens, "módulo 7" do curso, ou "faz um vídeo com o desenho/boneco falando", mesmo sem citar a skill.

Makes rendered 3D video clips from code with Remotion and three.js on the user's own computer, delivered as MP4 in 9:16, 16:9 or 1:1. Covers product shots and turntables, 3D logo reveals, extruded title cards, ad B-roll, looping brand clips and interiors. Use this skill whenever someone asks for a 3D video, 3D animation, animated logo, logo intro, rendered product video, or a 3D reel, short or TikTok, including Italian requests such as "video 3D", "animazione 3D", "logo animato" or "spot prodotto", even if they never mention Remotion or three.js and only describe the shot they want. Not for editing existing footage, flat 2D motion graphics, or interactive 3D websites.

Turn any feature, page or dashboard in the current project into a polished AutoZoom-style demo video with Remotion. The product UI is rebuilt (or imported) from the project's own code, then "filmed" on a wallpaper desktop with AutoZoom's smooth auto-zoom camera, a real-feeling cursor and clicks, spotlights, typing, keycaps, word-by-word captions, toasts and a REC chip, and rendered to MP4. Installs Remotion automatically. Use when someone asks for a demo video, feature walkthrough, product/launch video, onboarding animation, landing-page hero video, "show how X works", or anything in "AutoZoom style".

Turn a video script into per-scene voiceover for Remotion in one command - one MP3 per scene in a saved or cloned voice, measured durations converted to durationInFrames, word-level captions in Remotion's Caption format, and an SRT. Use when building a narrated video with Remotion (or any React/code video tool), when the user asks to "add voiceover", "narrate these scenes", "sync scenes to the narration" or "add captions that follow the voice". Voice and timing by Miragea, no GPU or ElevenLabs a

Caption any video or audio file in one command. Produces clean, readable SRT and WebVTT subtitles (whole words, 42-char lines, proper timing) and can burn them into an MP4, including big centred 1-3 word "pop" captions for TikTok, Reels and YouTube Shorts. Use when the user asks to add subtitles or captions to a video, make an SRT/VTT file, transcribe a talk or podcast into captions, or burn captions into a short-form video. English-first; speech recognition by Miragea.

Create editable, animated PowerPoint presentations from the conversation and supplied materials. Cover the user's content, infer the brief and theme, study and decompose local templates, and recombine their assets and motion patterns into original slides. Use for 精美PPT、动画PPT、素材整合、模板拆解 or presentations that need both visual design and purposeful motion.

Edit raw trip / car vlog footage (phone HLG talking-head clips recorded while driving, dash-cam road footage, photos, OBD/Torque screen recordings) into a finished HDR (HLG 10-bit) YouTube film with a teaser, jump cuts on speech, road inserts under music, captions, route map, licence-plate blurring, levelled voice, chapters, thumbnail, subtitles in several languages and dubbed audio tracks in the user's cloned voice. Use this whenever the user wants to edit, cut, assemble or finish a video from

Video Essayist specialist, tagged @video-essayist or consulted for script-led commentary and analysis: narration locked first, every line matched to a visual that proves it, short annotated clips, quotes with sources, thesis by 60 seconds, chapters. Use when the user tags @video-essayist or asks for a video essay, film analysis or commentary edit.

Wedding & Event Filmmaker specialist, tagged @wedding-filmmaker or consulted for wedding and event highlight films, teasers and vertical teasers: the couple's own words as the spine, cut to licensed music, slow motion only on peaks, true skin and an unclipped dress. Use when the user tags @wedding-filmmaker or asks for a wedding or event highlight video.

Style, time and place captions and subtitles so they read on a phone with the sound off. Sets line length, reading speed, cue timing, size, contrast and platform safe zones, and checks SRT or VTT files against them. Use when the user wants captions or subtitles added, restyled, fixed or made accessible, says captions are too fast, too small or hidden under app buttons, or is preparing TikTok, Reels, Shorts or YouTube video, even if they only ask to "put what I say on screen".

Designed graphics with HTML elements, animating the objects inside them on the timeline clock. Use for animated stat cards, infographics, diagrams, UI mockups, logo stings, or moving an object inside a graphic.

Motion graphics people can edit afterwards: compositions of layers with keyframes, After Effects style. Use for lower thirds, animated titles, kinetic type, stat reveals, logo stings and callouts that the person may want to retime or restyle by hand, and when a composition is already on the timeline.

Use when the user drops or points at a local talking-head video and wants the kinetic supers + rapid camera reposition prompt for video-to-video AI (one real take turned into split-second camera snaps that lock off, with bold motion-graphics text synced to the speech). Triggers on /kinetic-multicam, "kinetic supers", "kinetic typography video", a dropped .mp4 plus a motion-graphics-text + camera request. PT examples for reliability, "faz o prompt de kinetic supers desse vídeo", "gera supers ciné

把"一个主题"或"一份主题文档"做成可交付的动画短片（MP4 + SRT + 可编辑工程）。端到端覆盖选题调研、分镜、Remotion 逐帧绘制、风格插件、语音合成、混流与量化验收。当用户说"做一支讲解动画 / 把这篇文档做成视频 / 给这个主题配个片头 / 用某种风格出片 / 给片子配音"时使用。