Video & Animation

Structure Remotion markup for interactivity

Deep guide for game ANIMATION assets: motion cycles, action keyframes, effect sequences, and animation sprite sheets — built around a video-first pipeline. In this app-builder sandbox, execute via the video2dsprite / generate2dsprite skills (magenta + scripts), not ad-hoc ffmpeg. Use whenever generating anything that moves: walk/run cycles, attacks, idles, FX, flags, fire, animation sheets. Complements game-asset-core.

Manim CE animations: 3Blue1Brown math/algo videos.

Làm video tiếng Việt từ một yêu cầu, AI làm đạo diễn, không khung dựng sẵn: ý tưởng lớn, kịch bản, giọng VieNeu, phân cảnh theo lời, nhạc nền, dựng từng cảnh bằng agent Sonnet chạy song song (HyperFrames + GSAP), phụ đề, xuất MP4. Dùng khi người dùng gõ "/videomotion", "làm video", "video giải thích", "clip ngắn về X", "biến bài này thành video".

Write a platform-tuned caption plus a researched, non-spammy hashtag set for a post. Use when finishing a social post, asked for 'caption and hashtags', or optimizing reach for Instagram/TikTok/X/LinkedIn.

>- Use this skill when recorded footage must become a finished cut by talking to an agent ("vibe editing"): talking-head takes, interviews, podcasts, tutorials or screen recordings where the agent picks takes, removes pauses, false starts and fillers, shortens or reorders speech, or cuts a long recording into clips. It inventories sources (variable frame rate, rotation, HDR), works from word timestamps, writes an edit decision list with source and output times for approval, renders frame-accurately with FFmpeg, moves captions and inserts onto the new timeline, and plans sound (fades at joins, ducking, loudness). Also for choosing the tool for an edit and deciding which media may be uploaded. Not for HyperFrames or Remotion code, motion graphics or explainers without footage (hyperframes, general-video, remotion-best-practices), captions or overlays on footage that stays unchanged (embedded-captions, talking-head-recut), or checking a finished render (vibe-editing-review).

>- Use this skill when a rendered video must be checked before it is published or handed over: does it match the brief (size, frame rate, duration), the approved edit and the script; are there black or frozen frames, dropped sound, long silences, clicks, loudness or true-peak problems; do the captions follow the final timeline, keep to line and reading-speed limits and spell names right; is private data visible. It scans the file read-only with FFmpeg, guides one full viewing with sound, and writes a report with output times, evidence and fixes, plus what it could not check. Also for diagnosing a broken render, writing the handoff for another agent, and turning an accepted video into a template for a series. Not for making the cut (vibe-editing-cut), or validating and rendering HyperFrames or Remotion projects (hyperframes-cli, remotion-render).

Generate high-CTR thumbnail concepts for a video — layout, focal subject, text overlay (≤4 words), color/contrast, and an image-gen prompt. Use when designing a YouTube/short thumbnail, asked for 'thumbnail ideas', or planning cover art for a video.

Builds and maintains an interactive travel diary website (dark satellite map with 3D terrain, animated day-by-day "replay" of routes, photos and videos pinned where they were taken, points of interest with Wikimedia images, mobile-first) from a trip's raw material - diary text, photos/videos, GPX tracks and optionally a Google Timeline export. Use this whenever someone wants a trip website, travel blog with a map, road-trip / camper / bike-tour / hiking diary, "photos on a map with the route", a Komoot-style flyover of their days, or wants to add a new day of material (new photos, GPX, timeline, diary text) to such a project - even if they just say "there are new pictures and tracks". Three modes - init (scaffold a new project with a local setup web app for drag-and-drop input), day (process new material into the site), deploy (single HTML file, web space via SSH, or Synology nginx).

Read live Zoromi (zoromi.studio) prices and plans for short videos, vertical short dramas and illustrated books, and, with the user's own ZOROMI_API_KEY, their Gems balance, daily cap, books, video projects, job status, recent charges and a quote for filming video clips; and, with a key the user allowed to spend, make one picture or one video clip, paid in US dollars from their API balance, after the user says yes to its quoted price, or start a short drama from a script that the user then finishes on the website. Use when the user mentions Zoromi or zoromi.studio, asks what a short drama, video clip or illustrated book would cost there, asks about their own Zoromi account, books, projects, jobs or Gems, or asks for a picture, a video clip or a short drama made there. Prices, plans and what Zoromi does need no key; which other tools are open, the server's tools/list says, so call it first and ask for a key only when it lists tools that need one. Nothing is charged until the user approves the exact price.

>- Make a game's Google Play store listing: the promo FEATURE GRAPHIC (1024x500) and phone/tablet SCREENSHOTS captured from the real game running in headless Chromium, plus the listing TEXT (short / full description). Use when the user asks to "make store images / store assets / Play Store screenshots / feature graphic / store description / ストア画像 / スクショ(スクリーンショット)一式 / フィーチャーグラフィック / 簡単な説明 / 詳しい説明 / Play Console の説明文", or to capture many gameplay screenshots to choose from. Also builds a PICKER page (an Artifact with checkboxes, per-section comments and copy buttons) so the user ticks the store set -- screenshots, feature graphic, listing text, and app icon candidates too ("選定用の artifact", "チェックして選びたい", "アイコンも一緒に選びたい"). Works for Phaser/PixiJS/Three.js/plain-canvas games served locally. Sibling of `game-icon-from-screenshot` (that one makes the square app icon; its candidates can go in this picker). Not for hand-drawn/from-scratch promo art.

Premium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional logo concepting, refined composition, sparse typography, strong symbolic meaning, premium mockups, art-directed imagery, and flexible grid layouts.

Run a Scribe signature marketing campaign: build the banner content, target it, schedule/pause/resume it, and measure it. Use when the user wants to promote an event, launch, or offer through the banner under everyone's email signature.

Build and publish AI web apps on RunApps (runapps.ai) through its MCP tools, and write the app code with the RunApps JavaScript SDK (@runappsai/sdk). Covers getting authorized and connected (device-code flow, agent key, Bearer header, scopes), the platform model (apps, slugs, bundles, visibility and review, creator fee), every MCP tool, the SDK (sign-in, chat, images, audio, video, per-user files, IndexedDB sync, private prompts), the bundle rules (zip layout, dot files, entry page) and the safety rules. Use whenever the runapps MCP server is connected, or when a task mentions RunApps, runapps.dev, publishing an app there, the RunApps SDK, or an rk_ key.

Use SkillLens (this repository) to turn SKILL.md files into animated flowcharts, and to make skills chart well. Use it when someone wants to understand or present a skill's workflow, check whether the auto-extracted chart is right, add a skillflow block to a skill, export a web page / image / video of a skill, or change SkillLens itself.

PikPak cloud storage CLI (pikpak). Run every cloud-drive operation from the command line: browse and search files, cloud-download magnets/torrents/links, manage download tasks, stream videos, upload local files and folders, download to disk, share files and restore others' shares, move/copy/rename/delete, register accounts and manage access tokens. Use this skill when the user mentions PikPak or a cloud drive, or asks to: - cloud-download a magnet, torrent or link, or check task progress - upload a local file or folder, or download cloud files to disk - play or watch a video, movie or series stored in the cloud - share files, or save/restore someone else's share - browse, search, move, copy, rename or delete files, or create folders - log in, register an account, or save a personal access token - check account info, storage space or VIP status Trigger it even without the word "PikPak" (e.g. "download this link", "upload this file", "what's in my cloud drive", "play that video", "share this file").

yt-dlp 下载 → 上传 PikPak → 在线观看 → 清理本地 的一条龙管道：断点续跑、上传校验后才删本地、支持 URL 清单批量处理。覆盖 YouTube、TikTok、X(Twitter)、Facebook、B站 等 yt-dlp 支持的站点。支持 Windows / macOS / Linux。 One-shot pipeline: yt-dlp download -> upload to PikPak via WebDAV -> watch online -> clean local copies. Covers YouTube, TikTok, X(Twitter), Facebook, Bilibili and other yt-dlp-supported sites. Resume-safe; local files are deleted only after verified upload. Cross-platform. Must use this skill when: - 用户想「下载视频并存到 PikPak」「看完就删，本地不留」/ download a video straight into PikPak - 用户给了一堆 yt-dlp 支持的链接（YouTube、TikTok、X、Facebook、B站等），想批量转存云盘 / batch-save TikTok/X/YouTube links to cloud - 用户磁盘空间紧张，想把下载成果直接入云 / save disk space by downloading straight to cloud 关键词：yt-dlp / PikPak / 管道 / pipeline / 下载转存 / download to cloud / TikTok / Twitter / X 视频下载

Gestiona el Instagram @marcvallfisio (fisioterapeuta, Tonus Fisioterapia, Camp de Tarragona) para ganar seguidores locales y captar pacientes. Úsala para ideas de contenido, guiones de Reels, copys y hashtags, calendario semanal, respuestas a comentarios/DM y revisión de métricas.

>- Post-mortem on what the user has already posted - which reels actually worked, why, and what to stop making. Use when the user pastes their Instagram insights or past posts and asks "what's working", "why did this flop", "read my analytics", "audit my content", or wants to know what to do more of.

>- Write comments on other people's Instagram posts and reels that read as a person with an opinion, not a bot. Use when the user pastes a post or a reel and wants a comment, says "comment on this", "engage with this", "what do I say here", or wants a batch for their daily engagement round.

Plan a month of email campaigns with audience, purpose, angle, CTA, proof, timing, and frequency coordination. Use when users ask what to email each week, need a 30-day campaign calendar, or are sending too many promotions.

Record short clips of NPC conversations that look like real gameplay, ready to post. Covers scripted exchanges at chosen places and times, the gameplay camera held still behind the player, both lines rendered before the clip rolls so it plays back to back, the game's own HUD and subtitles with nothing of the mod on screen, timeline markers for cutting, screen and audio sync, 5-10 s cuts in 16:9 (and 9:16), pinning lines to re-voice an approved take, and re-taking clips that run long. Use when the user wants demo videos, clips or a montage of talking NPCs.

Use when the user wants a multi-perspective brainstorm on a feature, app, or product area — "spin up agents with different personalities/job titles", "look at it from different angles", "panel of experts", "come up with 30 ideas and a top 5", "how can we make X more useful", "brainstorm what to do next with X" — especially when they also want to DECIDE on the ideas (implement / hold / reject) rather than just read a list. Also use when they ask to re-run or extend a previous panel, or to "make the ideas decidable like dev-to-prod-promotion-review".

Use when collecting or revising source-backed content for an OMR video: собрать материалы по теме ролика, проверить факты и внешние источники, создать content/index.md и тематические Markdown-файлы. Not for requirements interviews, final narration or scene scripting, rendering, general research unrelated to a video, or development of OMR itself.