Video & Animation

Spec-driven explainer films as code: one line, a script, or a recording becomes a 30s–5min cinematic knowledge video with verified data, synthesized BGM, and (optional) Volcano TTS narration with frame-accurate subtitles. 一句话/文稿/录音生成知识解说短片。

Review and fix a motion-heavy web page (scroll stories, WebGL heroes, film scrub, shader sections) with a look-and-fix loop. Use when asked to review, critique, QA, polish or "make it ship ready" an animated landing page, template or hero, or before publishing one. Takes section screenshots at desktop and phone sizes, scores the page against a 10-criterion rubric, runs blocking QA checks (console errors, overflow, reversible scroll, pause control, reduced motion, text size, tap targets, white flash, fonts, alt text, focus, page weight) and fixes what fails.

Écrit et prépare des vidéos YouTube de survie / isolement racontées à la 2ᵉ personne ("you"), jour par jour, style documentaire sombre, avec faits réels vérifiés, crises techniques, mystère jamais expliqué et twist final (format "Bunker Survival 1 Year", 10–12 min). Produit le script complet calibré à la minute, le plan visuel (shot list, prompts IA, animation, étalonnage, audio), le titre, la description, la catégorie, la miniature par code, et génère les images automatiquement dans Google Flow (Chrome piloté par Playwright, scripts inclus) en mode diaporama. Use whenever the user wants a script, scenario, storytelling or faceless video about survival, bunker, shelter, isolation, apocalypse, nuclear winter, polar station, submarine, space station, lifeboat, lockdown, blackout, "365 days in", "one year alone", "what if you were trapped/alone", or wants to copy/adapt the style of a viral second-person survival video, even if they don't say "skill" or "template". Triggers FR/EN - script vidéo survie, histoire à

Editor de video profesional para clientes (reels, talking heads, montajes de eventos, videos de aplicación) en español. Organiza el material por cliente y proyecto, analiza referencias para escribir la guía de estilo de cada cliente, selecciona las mejores tomas con alineación MFA, y arma el video en HyperFrames con un motor propio (subtítulos por palabra, zooms, b-roll, tarjetas, sonidos) cuyo look define cada cliente. Úsala cuando el usuario pida editar videos de un cliente, "nuevo cliente", "nuevo proyecto", "analiza estas referencias", pase crudos o guiones para editar, o pida reels/shorts/recaps.

Écrit et prépare des vidéos YouTube explicatives de 20 minutes au format « Why Evolution Made… » (style Apefully) : un paradoxe sur le corps ou le comportement humain, ouvert par une scène du quotidien à la 2ᵉ personne, résolu par des chapitres « scientifique + année + chiffre + analogie », retourné en « ce n'était pas un défaut mais un marché », fermé par un retour à la scène et une question aux commentaires. Produit le script complet (3 300–3 600 mots), la liste d'images une par phrase pour un diaporama 2D sans animation (prompts avec préfixe de style, familles de plans, graphiques par code), le titre, la miniature « WHY … ? », les chapitres, la description, les tags et la catégorie, et génère les images automatiquement dans Google Flow (Chrome piloté par Playwright, scripts inclus) puis la miniature par code. Use whenever the user wants a faceless educational video, a science or evolution explainer, a "why do humans…", "why did evolution…", "the real reason…" video, a 2D stickman or whiteboard-style explai

Professional video production workflow for an AI editor. Runs an intake interview (story, references, sound, source website links, file paths), then plans, cuts, captions, animates (HyperFrames or Remotion), mixes, measures loudness and runs QC before delivery, with approval gates. Use whenever the user wants to make, edit, plan, caption, mix, render, fix or check a video, reel, Short, TikTok, YouTube video, promo, ad, explainer, product demo or motion graphic, including Bangla or Banglish requests like "video banabo", "reel edit koro", "notun video start korbo", even if they never name this skill. Also use when the user gives feedback on a draft, so the fix becomes a rule.

Make a 16:9 hand-drawn, Indian-maximalist lyric music video for any song — stop-motion character animation, a persistent ornate frame, word-synced kinetic Devanagari + Roman lyrics, and an eye moment on every refrain. Lyrics are extracted from the user's own audio automatically; the character (from the user's reference images or an original design) is built with the Atlabs MCP using Nano Banana 2.1, GPT Image 2.5 Sunburst or Seedream 5, kept consistent across expressions, outfits and colour-palette looks. Use when someone wants a "lyrical video", "lyric music video", "animated lyrics", "hand-drawn music video", "desi / Bollywood / Hindi song animation", "kinetic typography for a song", "illustrated reel for this song", or shares a song plus a character or reference video in this style. REQUIRES the Atlabs MCP — the skill stops and walks the user through connecting it if it is missing.

Use whenever working with FilmCraft on this machine — the Rust video editor and Premiere Pro alternative — including launching the app with control port ("FilmCraftを起動", "FilmCraftを立ち上げて", "launch FilmCraft"), creating, editing, trimming, colouring or exporting sequences and projects (.fcproj, MP4/MOV/MXF, PNG/TIFF/GIF, WAV/AIFF/AAC, XML/FCPXML/OTIO/EDL/AAF/OMF, SRT/VTT/SCC), driving the running desktop window, or using its MCP server, filmcraft-cli or control channel. Triggers on "FilmCraft", "filmcraft-cli", "filmcraft mcp", "FilmCraft MCP", "FilmCraftを起動", "FilmCraft立ち上げて", "FilmCraft開いて", "edit this video", "build a sequence", "cut this clip", "add a transition/effect/caption", "export to MP4/ProRes/DNxHR", "export XML for Premiere", "take a FilmCraft screenshot", "drive the FilmCraft window", and requests to automate video editing or FilmCraft itself. Invoke with @filmcraft.

Turn a video link into words - the transcript as spoken, a full translation, a summary in any language, or SRT/VTT subtitles - from YouTube (videos, Shorts, playlists, channels, topic search), Bilibili, Douyin, TikTok, Instagram and Facebook Reels and Vimeo, with speech-to-text when a video has no captions and the price shown before anything runs. Use when the user says "get the transcript of this video", "transcribe this YouTube / TikTok / Reel", "what does this Bilibili video say in English", "translate this video into Spanish", "summarise this YouTube video", "download the subtitles as SRT", "transcripts of every video in this playlist / channel", "transcript with timestamps", "this video has no captions, transcribe the speech", "视频转文字" or "把B站视频翻译成英文". Routes to one Actor, lexonia/lexonia-video-scrape. Does not download video files, read private or members-only videos, or search Bilibili or Douyin by keyword.

Create polished videos with code: narration, motion captions, screen capture, zoom, redaction, 3D and reusable libraries. Use for promos, explainers, demos and editing.

Edit supplied camera/phone footage into a finished video — talks and presentations, vlogs, interviews and podcasts, tutorials, product demos, event recaps — with transcript-driven selects, a stabilized virtual camera that tracks and reframes the speaker (wides, punch-ins, pushes, pans) over wide 4K shots, HDR→SDR, deflicker, per-clip color matching, dialogue cleanup and loudness, narration, native Swift motion-graphics titles, and optional music beds, sound effects and picture effects. Use when the user supplies video files to cut, reframe, grade, fix or title; not for generating footage.

Storyboard a short-form video (Reels, TikTok, Shorts) from its script: split the script into shots word for word, give every shot a 9:16 frame, the on-screen text, an edit note and the sound, and group the filming into sittings. Writes a storyboard folder with a board you can open in the browser, one PNG per frame and a contact sheet. Use when the user says "storyboard this", "make a storyboard", "turn this script into shots", "plan the shots", "shot list for this script", or has a finished script and is getting ready to film or edit. The finished storyboard doubles as the prompt for building the video's graphics one shot at a time. Also use right after the skill is installed, and when the user says "set up the storyboard skill", "learn my style" or "here are my videos": that runs the guided setup, which studies their own videos so every storyboard matches their content. Do NOT use to write or rework the script itself.

Internal part of Ice Motion Director, not a command: start with /ice. Motion craft for every Ice Motion Director film and edit: motion grammar and movement jobs, measured timing and easing curves, closed-form spring formulas, hold and stillness policies, the cut and transition catalog with exact numbers, seam continuity and the carry principle, camera moves as punctuation, composition, hero size and the visual richness floor, 2D and 3D motion, shape morphing, the cursor as actor, one-take films, anti-slideshow and motion density rules, each scoped to the film or edit types where it holds. Called by Ice Motion Director workflows; use only when called by name.

Use when audio already placed in a HyperFrames composition needs to be mixed: fade-in/fade-out, crossfade, track gain or volume, volume automation, ducking, a music bed that fights a voiceover (voiceover carve), effects on a track (EQ, compressor, limiter, gate, saturation, delay, reverb, chorus, phaser, bitcrush), automation envelopes drawn on a track's volume or any effect parameter, or one submix bus carrying a chain, a fader and an automation clock for several tracks at once (`<hf-audio-group>`). Don't use for sourcing or generating audio — finding BGM, SFX, or making a voiceover is `/media-use`. Don't use for clip timing or track layout, which is `/hyperframes-core`.

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet `anchor` rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

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.

终极口播视频 skill：中文口播稿 + 成品配音 → CPU 字级时间戳 → SHOTBOOK 层矩阵分镜 → Remotion 电影感成片（横屏默认/竖屏）。当用户要"做口播视频"、"解说/科普视频"、"把文案变成视频"、"给配音配画面动效"时使用。默认使用成品配音，可选 Fish Audio 从稿子合成配音与时间戳；数字人生成技术不在本 skill 内（人物素材是输入）。含统一视觉语言（Apple 范式）、108 张动效配方卡、镜头三面分层工作单、七层镜头反PPT系统（极缓推拉相机/让位，运动做减法）、六式运动承接转场（每式一卡）、长镜头世界画布、anime.js+three.js 桥、自动静止检测 + 独立 subagent 评估循环。

Make a 6-second 1080p kinetic-type title video (MP4) from three short lines, a chip, a footer and three colours. Use for launch openers, release announcements, social posts and reel intros when someone asks for an animated title, title drop, kinetic typography or a short type-only video.

Edit a raw vertical (9:16) take into a finished short in the SCRAPBOOK STYLE, as validated by the user (2026-10-05), reproduced exactly on any take: every pause cut out (jump cuts, the frame alternating wide and close on each one); captions low, three words at a time, the word being said in yellow; the hook in the reference's TWO FACES (white Archivo words, THE important word in Fraunces in yellow) with a line on PAPER; the paper only for the titles and two or three very important phrases: the reference's collage of small ruled notebook scraps, flat, written in a marker hand (Gochi Hand), landing and unfolding, breathing, folding away, the red marker drawing on it as it is said (a word struck out, an underline); die-cut emoji stickers with white outlines round the head for what the speaker lists; real reaction GIFs from GIPHY as polaroids, big (0.80 of the width) in the middle of the screen, always at the same place, dropped on and flicked away; sections on the same scraps swinging in with the camera punchin…

Edit a raw vertical (9:16) take of someone talking to the camera into a finished short in the Pipecorn EXPLAINER LANGUAGE, with the motion of an After Effects edit: the speaker in a frame with round top corners at the BOTTOM of a dark stage lit like a set (charcoal lit from the middle, a vignette, a light on each scene carried by the camera, a red glow in a corner, dust floating, a grid of dots), their head (and hands) rising out of the frame (person matte); ABOVE them, the stage EXPLAINS what they say as they say it, in SCENES joined by CAMERA MOVES (whip pans with a light streak, rises, push-throughs), everything with real MOTION BLUR, springs and expo curves, never still (a slow drift and push, parallax on the dots and the dust), every figure casting a soft SHADOW on the ground and GLOWING in its colour (glossy rims, a sheen across the cards, a flash of light behind what pops in): a two-line title per section (Inter ExtraBold with one Instrument Serif italic accent, a red box or a red underline), then a n…

使用 HyperFrames 框架制作 16:9 横版「Vibe 视频大赏」科普短视频。当用户要求生成知识科普、观点讲解、金句科普类短视频，或要求生成/修改 HyperFrames 视频工程（HTML+GSAP 时间线）、读写项目 index.html、执行 npx hyperframes lint / preview / render 时使用。适配豆包工作「本地电脑」模式。第三方技能，非 HeyGen 官方出品；上游 HyperFrames 由 HeyGen 开源（Apache-2.0，仓库 https://github.com/heygen-com/hyperframes）。

把一首歌（音频 + 带时间的歌词 srt/lrc）做成有故事的像素小人卡拍 MV（ClaudeMV 系列做法）。流程是先写故事线，再出带画面的分镜表给用户看、反复修改，用户确认后才渲染。支持多角色（我、她、新男友……）、剧情动作（松手、转身、哭、心碎、摘墨镜……）、7 种歌词互动、7 种转场、电脑桌面窗口视角、27 个像素场景和天气、卡拍特效，用 HyperFrames 渲染 MP4。用户说“做个像素 MV”“把这首歌做成 Claude MV”“像素小人 MV”“卡拍 MV”“上传了歌和歌词”时使用。只做口播或解说视频时用 /xumu。

Preserve the real identity and original pixels of portrait photos while applying tasteful cinematic slow-shutter motion-smear, background removal/replacement, and optional conservative geometric correction. Use for 撕拉感、慢门拖影、人物消散、纯色背景、保留原脸 photography edits.

Analyzes genetic variant effects on gene expression (RNA-seq), chromatin accessibility (DNASE), histone marks (ChIP), and transcription factors using the AlphaGenome API. Use when the user asks about non-coding variant effects, pathogenicity, clinical significance, disease associations, functional effects, gene expression changes, splicing disruption, or regulatory effects in promoters and enhancers. Also use for resolving biological terms to tissue/cell-type ontologies (UBERON/CL) or analyzing variants in chr:pos:ref>alt format.