Video & Animation

Direct and produce a short film made entirely in code, in one of the styles of the Lemo-Opuscar library (e.g. Impasto Oil Painting 油画厚涂, Watercolor Brush 水彩笔刷, Ukiyo-e 浮世绘, Satellite Briefing 卫星简报, Exploded Teardown 产品拆解). Use when the user asks for a video, film, short, promo, explainer or animation (视频、短片、动画、宣传片) in a named style, including explainers built from real material such as satellite imagery, laser scans, CAD files or archive scans (数据解说、科普、产品拆解), and new films cut from real archive footage and documents (剪辑与纪实, e.g. 用档案影像剪一支新闻片); when they want a film about their own topic and haven't chosen a style (help them choose); when they want to change a style into their own, make their own style from a blank page, or use a style they saved (改风格、从空白开始、我的风格); when they start the skill without saying what they want; or when they ask which film styles there are. Not for plain editing, trimming or converting the user's existing video files.

Set up ReelStyles on this computer for the user's AI agent. Checks and installs Node.js 22+, ffmpeg, git and Chrome for HyperFrames, wires the ReelStyles skill into the agents the user has (Claude Code, Codex, Cursor, Roo Code, Cline, Gemini CLI, GitHub Copilot, Windsurf), then renders a 5 s test film. Use when the user says "set up ReelStyles", "install ReelStyles", or before the first ReelStyles video if setup has not passed.

Turn a video URL or video file into context an agent can read, as one bundle folder with the post text, captions or a whisper transcript, and deduplicated frames. Use when the user shares a video link (x.com, YouTube, any site yt-dlp supports) or a video file and asks what it shows, to summarise or explain it, or to learn from it. Filing what was learned as a study note is the learn skill.

Make a 9:16 comedy-sketch reel (30-45 s, English) with two 3D characters - the Nerd (spiky hair, glasses, hoodie) and Claude, the small orange low-poly robot on his desk - rendered in Blender from one storyboard.json. The Nerd has a real dev/AI pain, Claude answers deadpan and slips in one verified tip, then a twist and a loop back to frame 1. Runs research with sources, script, free `say` draft, layout stills, ElevenLabs takes, whisper check, final EEVEE render, cover and QA. Use for "make a Nerd + Claude sketch", "duo reel", "new episode", "sketch reel about X", "video con l'omino arancione".

Start or continue a multi-session learning workspace in the current folder - a mission, vetted resources, learning records, numbered interactive lessons, reference sheets, a glossary, flashcard review, and a queue for working through a course (Udemy, YouTube playlist) or a book chapter by chapter.

Review a finished Pipeline piece (a short vertical video with its caption and cover) before it goes to a human for approval. Reads the /ig skill script results, applies four critic lenses against the campaign's own brief and voice, and returns a SHIP or FIX verdict as JSON. Use whenever a piece built by Pipeline needs a fresh-eyes review, or when asked to "review", "QA" or "check" a Pipeline piece.

Cut AI tells from interface and chart copy: eyebrow labels, 01/02/03 numbering, redundant subtitles, captions that explain what is already visible, leaked design notes, decorative badges, status dots, chart clutter and legends a direct label could replace. Applies /pstack:unslop to every string it keeps.

Makes narrated lesson films for YouTube that teach one technical mechanism in about one to two minutes, with every frame drawn by code and a computed score, for STEM viewers with short attention spans. Builds on the code-film skill for the look, the taste and the defect checks, and adds a spoken voice that sets the clock (Kokoro, word timestamps), a script written as one conversation, a pause to guess, a recap and an ending that fades, a scene system in 16:9 with parallel capture, -14 LUFS mastering, exact captions, and checks that the voice said every word and the picture moves on each word it is tied to. Use when the user wants an explainer or lesson video, a YouTube video that teaches something, a narrated version of a code-film, or a Short cut from one. Not for a film with no voice, such as a music video, an ad or a visual poem, which belongs to code-film, or an animated diagram inside a page or doc.

Make a narrated, captioned faceless short video (AI images and AI voice) on the user's own AutoShort or StoryShort account. Works with either MCP server, detects which one is connected, agrees the script with the user, shows the credit price the server returns and renders only after an explicit yes, then polls the render and returns the video link. Use when the user wants to make, draft, render or publish a short video, YouTube Short, TikTok or Reel.

Make product screenshots and demo videos on Windows and macOS - capture a web app / Expo web app / desktop window / iOS Simulator / iPhone file, then style it (background, iPhone/browser/window frame, lift, loupe, arrows, steps, blur, zoom, cursor, cuts) and render PNG or MP4 into a versioned library. Use when the user asks for screenshots, mockups, help-center images, changelog pictures, social posts (X/Instagram), demo/B-roll videos or screen recordings of their app.

Turn raw footage into a platform-ready deliverable. Use when the user wants to "make this a reel", "cut this for YouTube", "turn this clip into a Short / a TikTok", "edit this video for Instagram", "package this for X", "record my tutorial and edit it", or hands a phone clip / OBS capture / screen recording and names a destination platform + intent. The director runs PRE (intake, probe, route, source-prep) and POST (platform packaging, variations, editor handoff), and hands THE EDIT (cuts, grade, captions, overlays) to the `video-use` skill via a baton-pass. If the user only wants raw editing with no platform/deliverable framing, that is `video-use`'s job directly — this skill is the director that wraps it for a target platform.

Plans and writes paid advertising copy for Google Ads, Meta, Reddit, TikTok, YouTube, and other paid media. Always confirms copy style and tone before drafting, recognizes each platform's default deliverable counts and character limits, follows capitalization conventions, and runs a final compliance and anti-generic-copy check.

Edits a talking-head / screen-recording video (Instagram Reel, Shorts, TikTok, Camtasia export, tutorial) into a motion-graphics cut in a style the user picks from 7 presets (modern gradient, bold pop, neon tech, clean minimal, hand-drawn doodle, retro VHS, broadcast news). It adds text and graphics timed to the spoken words, sound effects that land on real on-screen events, creative transitions, camera punch-ins, a CTA ending, and optional styled subtitles, voice cleanup, pause trimming, a cover image and an @handle watermark. Persian/RTL is the first-class case, with bundled Persian display fonts. Use this whenever the user hands over a video file and asks to "edit", "motion graphic", "add effects/elements/sound effects", "make it engaging", or says things like «ادیت موشن گرافیک», «افکت صوتی بذار», «المان اضافه کن», «ترنزیشن», «زیرنویس خلاقانه», «سبک ادیت» — even if they don't name this skill.

白菜价数字人口播——DeepSeek 写稿 + MiniMax H3 出镜 + Remotion 剪辑，一条 15 秒竖版视频约 1.36 元。换一张参考图 + 一段稿子，自动出带花字、逐字字幕、二维码的成片。适合抖音/视频号，也可批量产出。

Remotionで動画を制作する。Biim形式の解説動画（3Dキャラ・Aivis音声・字幕・BGM/SE付き、1人解説/2人の掛け合い、横16:9/縦9:16）と、完成音源に合わせたモーショングラフィックスのMVの2種類を扱う。「〜の解説動画を作って」「この曲でMVを作って」「Biim風の動画」「kokuren / natto-zの声で解説」などの依頼で使う。設定を集めて制作ジョブのフォルダを作り、台本・シーン・音声・音響・QAを経てfinal.mp4まで仕上げる。

从选题资料到可发布文章包的中文自媒体生产 Skill。用于证据整理、SEO/GEO 关键词与引用、平台版本、去 AI 味润色、字数规则、头像/图标资产、Remotion 封面与插图、ImageGen 概念画面、封面上传复核和发布证据准备；默认停在最终发布前，不自动点击发布。

Find short-video formats worth testing before making a video. Read the user's own post performance from their AutoShort or StoryShort account, browse the available niches and templates, study reference Shorts the user names, and turn the findings into format cards and a small test plan. Use when the user asks what to make next, which format or hook style to try, or how to learn from videos that did well.

Referanstan Konsepte — kullanıcının beğendiği bir videodan ("bunun gibi bir şey yap", yapıştırılan bir Reel / TikTok / Shorts / YouTube linki ya da yerel klip) üretim başlatma protokolü. Videoyu indir, ölç ve plan plan ayrıştır (transkript, kesimler, tempo, ekrandaki yazılar, 5 açılı plan dökümü, gerçek hareket mi hareketlendirilmiş fotoğraf mı), neyin kalıp neyin değişeceğini yaz, dürüst bir araç yoluyla 2–3 farklı konsept öner, tam kurgudan önce 10–15 sn'lik örnek yap. "Şu reels'e bak" öğrenme notları için de aynı protokol. Kullanıcı referans video gönderdiğinde, "bunun gibisini yap" dediğinde ya da bir reels / gönderi incelememi istediğinde yükle. Tetikleyiciler - "bunun gibi yap", "şuna bak", "bu reels'i incele", "make something like this", "study this reel".

Fix audio that sounds "AI". Use when a video, voiceover, podcast, short drama, screen recording or reel sounds harsh, muddy, thin, too quiet or too loud, clips, or changes level between segments - and before delivering to YouTube, TikTok, Spotify, Apple, podcast or broadcast. Measures integrated loudness (LUFS), true peak, sibilance, low-mid mud, segment drift and phase against the 7 AI Audio Tells, applies a corrective chain, then re-measures and reports whether the delivery spec was actually met.

Edits the user's own video clips into a finished vertical reel for TikTok, Reels, and Shorts: the best moment of each clip, cuts on the beat, captions, labels, and an end card. Use whenever someone has videos, clips, or footage and wants a montage, reel, recap, tour, or highlights, even before they've sent the files. Only photos: that's montage-video.

Plan, edit, grade, subtitle, and quality-check short-form vertical video (vlog, campus/travel, interview, brand story). Use when the user asks to turn raw clips into a 45–90s vertical vlog, add burned-in Chinese subtitles, build title/fact/end cards, grade flat/Log footage, deliver a finished MP4, or revise an already-approved cut without regressing it. Triggers include: 剪 vlog, 竖屏视频, 加字幕, 出成片, 换素材, 调色, 修改视频, 视频复盘, 已批准版本, 保留原版. Do NOT use for long-form film/TV finishing, pure audio editing, 3D animation, or when the user only wants a text plan.

Builds one finished piece of interface that moves well from a plain description, in a look dealt fresh for each request. It makes a control such as a switch or a slider, an animated chart, a widget or card, a screen or page inside someone's app, or a short looping animation that demos an app. It writes the piece into the person's project in its own framework, or as one self-contained HTML file when there is no project, and adds motion to a piece the person already has. Use it whenever someone wants such a piece to look and feel good, even in a beginner's loose words like "make it pop" or "feels nice to click", and even when they never mention motion. Whole websites and landing pages belong to page-design skills, and settle builds only the moving piece inside one; video editing and title sequences belong to video tools, and reviews of existing animation code to animation-review skills.

Make an animated video in a finished visual style (cut-paper, 80s anime, cartoon, collage, diagrams, 3D and more: 3 styles), animated in code and rendered with HyperFrames. Use when the user asks for a video, a film, a Short, an explainer or a ReelStyles style.

Use when the user runs /reel-split, or asks for the split-screen explainer Reel — an animated widget panel on the top ~45 % of the frame, the speaker below it the whole time, no subtitles, no SFX — typical for "I found this tool / here are 5 tips" videos, on a talking-head take in raw/.