Video & Animation

Plan, make, edit, check and document sound effects for games, film, video and apps, keeping each file's source and licence on disk. A brief, shot list or game event list becomes a sound list; sounds come from procedural recipes with no model, any reachable ComfyUI, hosted APIs or CC0 libraries; variation sets, loops, layers and footstep sequences are built by script; measured checks catch clipping, clicks, DC, cut ends, padding and bad seams; `ship` answers 'can we ship this?'. Use whenever...

Produce a motion-design showreel or short animated video for a product, brand or story. Starts with an interview (conversation language and on-screen language first, English by default) that settles purpose, audience, story, voice, look, motion, music, length, formats, languages and assets before any production. Use whenever the user asks for a showreel, reel, animated video, motion graphics, kinetic typography, animated logo or intro, explainer or promo video, lyric video or product teaser,...

FilmForge director: runs an AI film production end to end (story, style lock, score, paintings, frame QA, motion in code, sound, delivery, archive) through stage gates, so no credit is spent before its gate passes. Keeps the production state file (production.yaml) and the credit ledger, prices the plan against the credits, and always names the one next action. Use this skill whenever the user wants to make, plan, budget, continue or check on a short film, painted film, animated story, music...

Product demo and release-notes videos with no screen recording. Rebuild the app's screens as static React from its source code, then drive them in Remotion with a scripted cursor and camera (lib/components Stage + Cursor). Use when making a product walkthrough, feature demo, release-notes or changelog video for an app whose source you can read, or when a Playwright recording would be flaky, need test data, or go stale. Triggers include synthetic demo, rebuild the UI, release video, feature...

This skill should be used when the user asks to make a promo, demo or explainer video for a product, website or open-source project from its URL, a GitHub or GitLab repository, a local repository or a folder of screenshots, producing a 9:16 bilingual Chinese-English short video with optional local voice narration.

FilmForge motion: plans each shot's life in code instead of credits (painted holds brought alive with drifting fog, ash, snow, rain, embers, glints, light sweeps, firelight flicker, heat shimmer, shake, lightning, glass cracks, transitions, golden threads, handwritten words and titles), maps every effect to the painted-film-pipeline renderer's exact parameters, applies frame-QA code fixes, exports the edit timeline and renders the preview; decides when paid video is worth its credits. Use...

Process video, audio, and images on the local CPU. Normalize video, transcribe speech, OCR images, and translate the extracted text when asked.

Create demo videos for CLI and TUI tools. Use when user asks to "record a demo", "make a demo GIF", "create a demo video", "record my CLI tool", "asciinema demo", "demo recording script". Covers asciinema+agg+ffmpeg pipeline, tmux recording, multi-shell, and dual-purpose demo+test harnesses in Python, Rust, TypeScript, and Bash.

Make videos with code (video-as-code / code2video) through the OpenVideoHarness workbench. Covers 3Blue1Brown-style math and science explainers, knowledge shorts (vertical or horizontal; 抖音, B站, 小红书, Shorts), product launch films and promos, lyric videos and music videos, data stories, paper explainers, hand-drawn or watercolour shorts, meme or fast-cut edits, long landscape explainers, recaps from the user's own photos and clips (weddings, trips, a year; 照片回顾), and cutting the user's own...

協助製作 AI 漫劇／動態漫／短劇，從小說改編、劇本與分鏡到提示詞和可選媒體生成。內建確定性驗收（時長量子、證據分級、segment 級依賴帳本、失敗目錄）與單一入口 h3d；圖片或影片生成取決於目前 Agent 工具及已連線的相容後端，本技能不假設 Agent 內建影片生成能力，也可只交付文字劇本、分鏡或提示詞 draft。

Comedy shorts about a group's own life - a team's launch, quarter end, office move or leaving do; a club's season; a class's year; a family's Christmas - or about a shared topic everyone recognises (video calls, the return to the office), told in a borrowed genre (80s action intro, creature feature, Dickens Christmas ad, heist, nature documentary) and starring the group's real, animated photos or an invented cast. Covers mining the group's own context for in-jokes (chat, email, calendars,...

Assembles finished vertical social video ads from clips plus a simple JSON shot list. It outputs a 9:16 master for Reels, Stories and TikTok plus a 4:5 feed cut, with brand-color dish labels, the brand on frame 0, hard cuts, a matched color grade, clean designed sound, safe loudness and an end card. It uses free ffmpeg only. Use when the user wants to edit, cut, assemble or render a short ad, Reel or TikTok from food or product clips, or make several ad variations to A/B test.

Improves a video ad through repeated blind A/B reviews. Two versions are shown as contact sheets to a fresh reviewer who doesn't know which is newer, and both are scored against benchmarks from winning food ads. The next version changes only what reviewers agree on, and the loop continues until the gains stop. Use when the user wants to improve, iterate, critique or A/B test an ad or Reel before spending money, or asks "which version is better".

Directs AI image-to-video food shots for restaurant ads (chopsticks lifting noodles, a spoon scooping, sauce poured, a piece lifted to the lens). It gives the prompt formula, which foods animate well and which morph, and a keep-or-reject checklist so the dish still looks like the real menu item. Works with any image-to-video model (Kling, Veo, Seedance, LTX, Runway). Use when the user wants to animate a food photo, make food B-roll or a hand-in-frame food shot, or turn menu photos into video...

Breaks a video ad down the way a creative director would. It measures length, cut speed and first cut; pulls frame strips of the first 3 seconds and of the whole ad; transcribes the speech; tags the hook, format, people, text, offer and ending; and scores it against benchmarks from winning food ads. Use when the user asks to analyze, break down, tear down, critique or compare a video ad, Reel or TikTok, especially for a restaurant or food brand.

The panel-free version of teardown. Turn one complex GPT Image concept into a real 3D film and an interactive 3D website, with no HUD, labels, charts or explanatory text anywhere. Use when someone wants to bring an image to life, make a cinematic 3D clip or short movie from code, a "living" picture, a 3D film with no text, a 3D hero or scroll experience built from an image, or says the teardown output has too much text on it. Use teardown instead when the point is explaining how a machine works.

Plan zoom together with the user in the plan-editor tool (click-a-word, see the plan-editor skill), then add dynamic punch-in zoom crops to an already-cut talking-head video at content-driven moments, correctly centered on the subject with proper headroom - so cuts between different zoom levels read as an intentional, cinematic camera choice instead of the subject visibly jumping around the frame.

Use with the core x-motion skill when a film should look and sound like ink-wash painting in motion (paper, carbon ink, water, mist, breath, one vermilion accent).

Art-direct and build original motion graphics, animated explainers, title sequences, product films, and kinetic typography. Use for motion briefs, storyboards, animatics, deterministic HTML animation, synchronized audio, and rendered-video critique. Start with the audience and a visual thesis, not a template.

Use with the core x-motion skill when a film should be about light, glass and colour at rest (refraction, dispersion, caustics, lenses) in a bright paper-white studio, rendered deterministically with a synthesized glassy score.

Use with the core x-motion skill when a short film needs the Orbit look of physically lit space, planets, eclipses, machined titanium and a quiet synthesized score.

End-to-end runbook for a YouTube video, Short or longform, built in Remotion and narrated by a cloned TTS voice, in one of three modes: a screen tutorial with simulated app screens, an illustrated explainer with drawn pictures (medical and science teaching, Pathology Notes), or a hybrid, or a motion-graphics news/incident explainer (built with remotion-motion-explainer).

End-to-end pipeline for producing Vox-style explainer videos from a single topic prompt. One topic in, finished film out — script, keyframes (Seedream 5.0 Pro), animation (Gemini Omni Flash), voiceover (xAI TTS), music (MiniMax Music 2.6), all assembled locally with ffmpeg via the Atlas Cloud API.

Direct any short video (about 6 seconds to 3 minutes) that mixes motion graphics with generated or supplied photos, footage and clips, such as a product launch film, an explainer, a brand film, a designed demo or tutorial, a vertical social short, an event, recap or data video, or a live-action story or comedy ad with a character and a product moment.