Video & Animation

Launches the evil-video-game-nerd subagent: an aggressive external reviewer who plays and stress-tests the Godot game as an outsider without access to the GDD, roadmap or project rules, hunts bugs, crashes, soft-locks, bad UI, bad controls and weak game decisions, keeps its own memory and a findings ledger across rounds, and rechecks earlier findings. Modes: player (black box, default), qa (also reads code and logs, runs the chaos runner), critic (compares the game with what the store text promi

Starts a new React web project or a new large web surface, or brings an existing one onto a written standard: product intake, rendering model and framework (Next.js App Router or React with Vite) chosen from the product, UI kit (shadcn/ui, Ant Design, Material UI or the existing system), styling approach chosen from the kit (Tailwind, CSS Modules or kit theme) with official sources, real i18n from the first second language (next-intl or i18next), motion foundation with reduced motion, theme, str

Audit a source repository for optional outbound requests, telemetry and promotional UI, then prepare reviewed offline-adaptation patches with the repo-airgap CLI. Use for privacy audits or air-gapped deployment adaptation.

Make flat-color line-art animated shorts with an image model and scripts, from story, character designs, drawings and mouth variants to a finished video with sound. Needs only Python and ffmpeg. Use for animated shorts, TV cartoon style, limited animation, dialogue films, keeping a character consistent across shots, or fixing mouths out of sync and flickering drawing swaps.

Write anything on a real narrative structure picked by emotion from 926 storytelling forms across 10 traditions (Kishōtenketsu, Noh, Propp, Hero's Journey, the Plea, the Quiet Pivot and hundreds more). Use this skill whenever the user asks to write, draft or rewrite something meant to move a reader (an email, post, essay, speech, story, script, pitch, apology, eulogy, letter, landing page, thread) or says "ar3stotle", "write it as a story", "give this a narrative", "which story structure fits",

Instagram skill: make, check and schedule Instagram posts from your AI agent, for one account or many. Turns an idea, a link, notes, photos or a video into finished posts: carousels and single images rendered locally (1080x1350 JPEG, no paid API), your own photos fitted and converted, Reels finished with a cover. Checks every file and caption against Instagram's current limits and fixes what it can (PNG/HEIC to JPEG, 4:5 and 9:16 fits, codecs, 5 hashtags), shows a preview of the feed, the grid a

Fully automated, self-improving Instagram Reels growth loop for any account. Claude studies the reels already on the page (and any reference reels the user likes) to learn the style, then every day plans, writes, voices, renders and quality-checks the next reel itself, posts it through vidIQ for free after the user says "post it", pulls the reel's own stats 24 h later, compares it with the account's typical reel, records what it learned, and makes the next reel better. Includes setup, a themeabl

Producir una composición editable de HyperFrames y verificar el vídeo exportado. En IAbot.io, asígnala a un empleado de producción audiovisual y aporta briefing, recursos autorizados, duración y formatos de entrega.

Turn a topic into a narrated two-host podcast video using local VoiceStudio voices and HyperFrames, rendered to 1080p MP4, after interviewing the user about what they want. Use when the user asks to make a podcast video, an AI-voiced conversation video, or a YouTube explainer in podcast format.

Use quando o criador mandar o mp3 da narração de um Short já roteirizado. Transcreve e nivela a voz, busca vídeos grátis (Pexels) e trilha (Pixabay sem Content ID), corta os clipes, escreve o generate.py das cenas com lib_short, aplica a fita VHS, confere quadros, renderiza com HyperFrames e entrega o vídeo na Área de Trabalho.

Turn a user-supplied emotion-tagged narration script (script.txt with tags like [excited], [curious]) and its voiceover into a consistent 2D educational animation in HyperFrames, using the project's visual references. Style: 2D flat, dominant white background with faded sketch lines, only the main subject and text in full color. Do not draft narration unless explicitly asked. Also covers sound effects: where they go, how to retrieve or synthesize them, and how to wire cues into the timeline.

Makes a narrated demo video of any project, where the person talks to camera and the screen shows exactly what they're saying. Researches the project, captures real footage (browser runs with Playwright, `claude -p` agent takes, terminal runs, live pages), writes a dialogue script the person reads in one sitting, then takes their single raw recording and turns it into the film - local transcription, retakes and fillers cut, the take split into sections, a cue sheet tying every visual to a spoken

Make story-driven videos with heavy motion graphics from real footage, photos and music, or from nothing but a brief, with a new theme chosen for each one. Asks for what it needs first: format, orientation, rhythm, motion intensity, visual identity. Use when the user wants to edit or make a reel, TikTok, Short, vlog series, aftermovie, event promo or reminder, progress update, animated poster or story, QR or projector loop, or says "make a video", "editează", "montează", "fă un reel", "fă un vid

Dalang — turn any topic into a finished animated video (narration, free local AI voice, music, sound effects, HyperFrames visuals): script → voice → per-word timing → visuals, each stage gated by the agent itself. Any topic, language and platform (Shorts, Reels, TikTok, YouTube, Instagram), stylized and true to the real thing. Use whenever the user asks for a video, animation, explainer or short from a topic, idea, article or notes ("make a video about X", "bikinin video tentang Z"), or to borro

Make a realistic 3D explainer short (30-60 s, vertical 1080x1920) in real Blender, a realistic human on a flat signature backdrop, camera dollying inside the body. Asks first for the voice (ElevenLabs, a clone of your own voice, your own recording, or captions only), the look, the main character (your own rigged file, one made in Blender, a free download, or none), the render engine and the length. Measures reference shorts frame by frame, writes an original fact-checked script, builds the scene

Verifica webs con Playwright CLI antes de darlas por terminadas: errores de consola, requests fallidas, overflow responsive a 375/414/768/1280/1440, accesibilidad axe, teclado y foco, reduced-motion, enlaces, SEO/meta y métricas de carga. Úsala al testear una web, tras cada cambio grande o para validar un deploy.

Make a 16:9 animated dashboard explainer video (mp4 + Python script + narration) for a section or result of a research paper, with a parameter sweep and live plots.

Implementa experiencias de scroll cinematográficas estilo Apple (hero pinneado, secuencia de frames en canvas, video scrubbing, capítulos narrados) con GSAP ScrollTrigger, sin scroll-jacking y con rendimiento medido. Úsala al construir heroes animados, secciones pinneadas, galerías horizontales o cualquier efecto "la web es como un vídeo que vas moviendo".

Use when turning supplied text, images, audio, video or product material into Chinese explainer, training or product videos, or browsing this package's visual and motion gallery.

Create, render, revise, inspect, or cancel narrated explainer videos from text, a topic, URL, PDF, document, notes, or a storyboard. Use for explainer video, whiteboard video, doodle video, educational video, diagram video, text-to-video, document-to-video, narrated video, "make a video from this", "做成解释视频", "生成白板视频", or "加配音和字幕" requests up to five minutes.

系列讲解视频全流程制作（策划→HTML动画页/文案→讲解稿→配音→字幕→1080p 成片→抖音封面横竖双尺寸），内置四道用户确认门禁（开工五问、讲解稿审阅、音色试听、合成确认）与发音修正能力（如「老子」读 lǎo zǐ）。This skill should be used when the user asks to 做一期讲解视频、制作讲解视频系列、道德经/经典解读视频、课件讲解视频、把PPT做成带配音和字幕的视频、生成讲解视频、新一期视频开工、继续做下一期讲解视频、给课件加语音讲解、带动画效果的讲解视频、给视频做抖音封面/竖版横版封面。触发典型场景：用户要为某个主题/经文/章节制作带旁白字幕的系列讲解视频，或要求沿用上期风格继续出下一期，或要为成片配抖音上传封面。

Layers AI images and video in depth (soft foreground, sharp subject, hazy background) and fixes flat frames. Prompts for H3, Seedance, Kling, FLUX, GPT Image, Nano Banana.

Use para desenhar as cenas animadas de um Short no generate.py - escolher a peça certa da lib_short para cada trecho (título do quadro 0, palavras, contadores, alvos, setas, bandeiras SVG, pavio de progresso, revelação com tela que se enche, final calmo ou energético, loop) e garantir uma interação por cena no tom do vídeo.

Edita vídeos con HyperFrames y FFmpeg a partir de crudos, referencias y kits de branding. Prepara un storyboard visual y espera aprobación antes de producir la edición; permite corregir versiones y guardar el estilo a petición. Úsala para editar o preparar el editor Makyd en Claude Code o Codex.