Video & Animation

Glass shards fly in through fog and tile themselves into the headline. HyperFrames block, 1920×1080, 12.16s, 19 variables.

Build and render code-generated mathematics and science explainer videos with Remotion, React/SVG, and transparent Manim sequences. Use for formula-driven educational animations, waveform or spectrum explainers, and synchronized caption timelines that must produce a verified MP4.

Leaves sweep through the frame and part to reveal the headline. HyperFrames block, 1920×1080, 12s, 11 variables.

Use when installing, validating, or repairing the Boring Video planning skill set before its first use.

Use when a video must show terminal commands, CLI output, streamed execution, full-screen terminal applications, or code entered inside a terminal session.

Use when an approved content topic, research bundle, and evidence need to become a complete platform-independent article before video beats or publication adaptation.

Captions living in 3D space around a talking head: a camera flies between the speaker and the words (whip-in, parallax truck, push), caption groups sit at different depths so camera moves pull them apart, hero words hide BEHIND the speaker through an alpha matte, a ring of words wraps round the speaker and turns in front of them, focus racks between depths, and text steps at 15 fps with ghost motion blur. Works in any type style (editorial serif, bold sans) or hand-drawn (p5.brush write-on). Trigger on: "3D text", "3D camera captions", "text around / behind the speaker", "camera moves through the text", "depth captions on my avatar video". Covers prep (portion, padded plate, matte, word clock, font metrics), the shot grammar, depth rules, QC, and limits.

Use when a content-driven video has a candidate topic but still needs research, experiments, a defensible thesis, or user-approved scope before an article or video beats can be written.

Make an original, unconventional animated MP4 video with its own synthesised soundtrack: every frame drawn in code, beat-synced. Use when asked for a creative, unique or "crazy" video on any topic.

Use when developing a content-driven video from a topic, research bundle, article, or draft, or when routing an existing script or non-argument video into HyperFrames production.

List, create, edit, or delete the member's reel styles, set the default, or re-render the last reel in another style. Use for "/reels-engine:reel-style", "show my styles", "make a stories style", "switch the font", "make the captions bigger", "more zoom", "try that reel in my other style".

Edit 1-4 talking-head clips into a finished, styled Instagram reel. Use when someone says "new reel", "edit my reel", "/reels-engine:reel", "make a reel from these clips", adds video clips to the chat, or drops clips in their Reels inbox. Also handles follow-ups on the last reel ("tighter cuts", "different style", "move that line").

One-time setup for the reels engine, and repair when /reels-engine:reel says a tool is missing. Installs tools, creates the Reels folder, and sets up the member's first reel style through a short interview. Use for "/reels-engine:reel-setup", "set up reels", "reels engine isn't working", or "change my handle or reel length".

>- Build Remotion videos with RemotionUI: CLI install workflow, composition registration, and agent indexes. Use when the user wants captioned social clips, creator reels, data stories, or programmatic video.

Use whenever the user wants to start or run a VideoExpress project, asks for VideoExpress image, video or multishot prompts, asks the assistant to operate VideoExpress in a browser, or reports VideoExpress problems such as drifting characters or outfits, camera direction flips, stylised clips turning realistic, props changing shape, actions in the wrong place or order, dark scenes, text in footage, background hiss or unwanted music. Plans, scripts and produces multi-clip videos in VideoExpress (VE) 3.5 and later by PaulPonna.com, where each 3–10 second clip is a start image plus a motion prompt. Covers guided project intake from a concept or script, the Core Bible, the VE multishot prompt format, Create Mode and custom Creative mode styles, clip chaining, Consistent Character references, lip-synced dialogue (in the multishot prompt or the Lipsync HD dialog), production sound, and driving app.videoexpress.ai with a browser-capable agent.

用户提供自己的数字人视频（或先拿 Claude 写的口播稿去数字人工具生成），和 Claude 多轮沟通、逐轮打磨，按 combo 风格（博主A × 博主B 合并后的"暗底固定框架 + 数字人口播 + 少而准的动效"）做出 1–3 分钟横屏精品讲解视频，目标是成片和原片一样精美。用户说"按 combo 风格/博主A风格/博主B 风格做视频""我有数字人视频，帮我包装/加动效""帮我写数字人口播稿/脚本""出分镜表""给剪映剪辑说明""做 Remotion 动效视频/写 timeline.json/出样张/预览/渲染""这条视频不够精致，再改改"时使用。流程是协作轮次：素材接收 → 稿子 → 分镜 → 关键帧样张（与原片并排） → 低清预览 → 精修终版；每轮交给用户看、请用户确认、按反馈修改，可随时回到前面任一轮。

Use when the user wants a single VideoExpress image or clip prompt, a one-off shot, an image plus its clip, or help improving one existing prompt. For a multi-clip video, a script, a prompt pack, chaining across clips or operating VideoExpress in the browser, use videoexpress-scripting instead. A question-by-question wizard that writes one paste-ready VideoExpress (VE) 3.5+ prompt or a matched image-and-clip pair, walking through look and style (with links to the Creative mode style catalogue), what happens, dialogue, sound, camera, light and image, and negatives, with the same prompt rules, multishot format, styles, Consistent Character advice and lip-sync methods as the VideoExpress Scripting skill.

做口播 / 动效视频之前的备料：数字人（形象、绿幕生成、抠像、清晰度、口型校准）、克隆声音（读音规则、停顿处理）、沉淀动效积木（拆解参考视频→提取带数值的规律→做成带参数的组件→逐镜头复刻验证）、固定框架（字幕胶囊、背景、封面）。用户说「准备数字人」「学这个动效」「拆一下这个视频的动效」「沉淀成积木」「做我自己的动效库」「备料」「我还缺什么才能开始做视频」时使用。备齐后用 talking-video-oneshot 做片。

一句话需求 → 口播 / 动效讲解视频成片（含配音对齐字幕、数字人抠像融合、AE 级动效、打字机与音效、封面和一句话文案）。用户说「帮我做一条关于 X 的视频」「做个口播视频」「做一条讲 X 的动效视频」「把这段稿子做成成片」时使用。本 skill 是总入口：自己决定默认值、自动推进全流程，只在「只有人能做」的环节停下；细节分别读 combo-style（口播包装与协作）和 ae-motion-craft（动效规律与方法）。

Produz vídeo determinístico com HyperFrames — composição HTML/CSS capturada em Chrome/FFmpeg, com formato, fps, duração e resolução declarados e render autorizado. Use quando a entrega É um arquivo de mídia renderizado (MP4, WebM, MOV, PNG), quando o usuário pedir peça de vídeo, motion, animação exportada ou composição reproduzível. Não usar para vídeo já existente usado dentro de um site (isso é asset web) nem para UI animada interativa (isso é front-end web comum).

Produce a narrated, captioned training or walkthrough video of a software application by driving the real app in a browser and recording it - storyboard, offline AI voice-over, screen capture paced to the narration, chapter/title cards, burned-in captions, .srt subtitles, chapter list and poster, as a ready-to-publish 1080p MP4. Use this whenever someone wants a training video, tutorial video, product demo or walkthrough video, onboarding video, how-to video or "video showing how to use the app" for a web app, SaaS product, internal tool or desktop app with a web UI, or wants an existing training video re-recorded after the interface changed - even if they don't mention narration, captions or screen recording.

用代码做手绘风格的动画短片：水彩、水墨用 p5.js + p5.brush，蜡笔绘本用自写的 2D canvas 蜡笔引擎，皮影用自写的皮影引擎，打斗用自写的设色水墨引擎；逐帧画，无头 Chrome 渲染，ffmpeg 合成 MP4。用户说「做个短片 / 动画 / MV / 开场动画 / 十几秒的视频」时用这份；配整首歌的长片或 MV 另看同目录的 mv-workflow.md，蜡笔画风看 crayon.md，皮影看 piying.md，打斗和雾山五行风看 wushan.md，科普讲解片（原理科普、HTML 演示 + 代码合成声效 + edge-tts 配音，允许字幕和标注）看 explainer.md。版画系列（孔版、铜版、大理石纹、蓝晒、网点、植物图版，SVG 生成，每段一种工艺）看 prints.md；音符驱动的流体彩墨 MV（WebGL 流体、抽象、不画人）看 fluid.md；沙画（灯箱上的沙底下透光、一幅抹成下一幅，WebGL 高度场）看 sand.md；作业本涂鸦（圆珠笔勾线 + 彩铅排线 + 手机俯拍的作业本，涂鸦小东西活过来，带实物笔和影子）看 doodle.md；共点蒙太奇（同一个形状始终落在画面同一点，快切或画风接力）看 montage.md；要自己写配乐（Tone.js + 真乐器采样含 VSCO2 管弦乐和真鼓，离线导出，分轨加自写混音台，按拍对画面）看 music.md。

Create lyric videos with user-selected typography and reveal styles, word-level timing, an unclear-transcript review and correction loop, and verified exports. Includes an optional inverted-footage typography preset.

Add motion graphics on top of the user's own raw / talking-head video (Damiano 'Motion Graphics with Claude Code' workflow, PROMPT 11 + 12). Use when: the user gives a raw video, reel, clip or footage and wants it edited, animated, 'made into a masterpiece', cards/counters/checklists added, or a specific moment fixed ('when I say X, show Y').