Prompt Motion
@AiMotionSiteS
Prompt library

Films

Vertical animated cartoon short about Casanova and a losing-proof betting system

Vertical animated cartoon short about Casanova and a losing-proof betting system is a 49.5-second Claude Opus motion design reference by @kirisando, built around Three.js, Canvas, AI voice. [Note: the creator shared a partial prompt. Copy this motion film prompt, keep the creator's frame counts and palette, and adapt the subject for your own launch film.

@kirisandoView post

Prompt

[Note: the creator shared a partial prompt. Wording below is theirs.] GOAL: Make a vertical animated cartoon short of MY story, built in code: three.js r160 world + canvas-drawn 2D character cards, rendered frame by frame on the GPU in a headless Chromium browser (macOS, Windows or Linux), voiced locally with the free Kokoro TTS voice bm_george, encoded with ffmpeg. 1080x1920, 30fps, exactly as long as the voiceover. Work start to finish without stopping or asking. Use a task list for progress.", MOTION: "my_inputs": { "story": "[your idea, true event, legend or full script]", "script": "[optional exact script; if empty, write it]", "length_seconds": 30, "captions": true, "notes": "[optional: era, setting, characters, jokes]" }, "stages": [ "A. Script (skip if I gave one): ~2.3 words/s for this voice (30s ~ 70 words, 45s ~ 100): 1) hook = absurd fact, no setup 2) setting in 1-2 lines with a concrete detail 3) 2-4 escalating beats, others doubt/mock, one exaggeration joke 4) pivot 'But...'/'You see,' revealing the real mechanism with an everyday analogy 5) short ironic twist = last words. Spoken English, contractions, numbers as words, short sentences, never describe what's on screen. If presented as true, web-check facts and keep only what holds up; label legend/disputed parts in the script itself. Save work/script.txt and show it in chat.", "B. Voiceover (free, local): Kokoro TTS in tts/: standalone Python 3.12 (https://t.co/3AS0rYiXUk, install_only build for this OS/CPU), venv, pip install kokoro-onnx soundfile; kokoro-v1.0.onnx + voices-v1.0.bin from https://t.co/Jb3SN4S0Vo releases (model-files-v1.0). Voice bm_george, speed 0.95, model loaded once. Split the script into short lines at sentence ends AND commas; voice each line to work/vo/sNN.wav. Never time-stretch (atempo/rate add echo): slow it with pauses. Trim line edges (-50 dB), join with gaps ~0.2s after commas, ~0.4s after sentences, ~0.55s before the pivot and the final line. Write voice.wav (48k mono) + work/phrases.json {text,start,end,words:[{w,start,end}]} (words spread by syllables). Tie every cut and reaction to these times, shot boundaries to phrase TEXT.", "C. Build everything below without stopping. Show the stills contact sheet in chat before the full render, fix what it shows, then render." ], "setup": [ "Works on macOS, Windows and Linux: detect the OS first and use its shell (PowerShell on Windows), paths (.venv/Scripts on Windows, .venv/bin elsewhere) and python command. Create web/lib work/vo stills frames out.", "Python venv .venv with numpy Pillow imageio-ffmpeg; ffmpeg = imageio_ffmpeg.get_ffmpeg_exe(). Nothing system-wide: no Node, no Playwright, no Homebrew. Kokoro gets its own Python 3.12 in tts/ (stage B).
Author original
Make a vertical animated cartoon short of MY story, built in code: three.js r160 world + canvas-drawn 2D character cards, rendered frame by frame on the GPU in a headless Chromium browser (macOS, Windows or Linux), voiced locally with the free Kokoro TTS voice bm_george, encoded with ffmpeg. 1080x1920, 30fps, exactly as long as the voiceover. Work start to finish without stopping or asking. Use a task list for progress.", "my_inputs": { "story": "[your idea, true event, legend or full script]", "script": "[optional exact script; if empty, write it]", "length_seconds": 30, "captions": true, "notes": "[optional: era, setting, characters, jokes]" }, "stages": [ "A. Script (skip if I gave one): ~2.3 words/s for this voice (30s ~ 70 words, 45s ~ 100): 1) hook = absurd fact, no setup 2) setting in 1-2 lines with a concrete detail 3) 2-4 escalating beats, others doubt/mock, one exaggeration joke 4) pivot 'But...'/'You see,' revealing the real mechanism with an everyday analogy 5) short ironic twist = last words. Spoken English, contractions, numbers as words, short sentences, never describe what's on screen. If presented as true, web-check facts and keep only what holds up; label legend/disputed parts in the script itself. Save work/script.txt and show it in chat.", "B. Voiceover (free, local): Kokoro TTS in tts/: standalone Python 3.12 (https://t.co/3AS0rYiXUk, install_only build for this OS/CPU), venv, pip install kokoro-onnx soundfile; kokoro-v1.0.onnx + voices-v1.0.bin from https://t.co/Jb3SN4S0Vo releases (model-files-v1.0). Voice bm_george, speed 0.95, model loaded once. Split the script into short lines at sentence ends AND commas; voice each line to work/vo/sNN.wav. Never time-stretch (atempo/rate add echo): slow it with pauses. Trim line edges (-50 dB), join with gaps ~0.2s after commas, ~0.4s after sentences, ~0.55s before the pivot and the final line. Write voice.wav (48k mono) + work/phrases.json {text,start,end,words:[{w,start,end}]} (words spread by syllables). Tie every cut and reaction to these times, shot boundaries to phrase TEXT.", "C. Build everything below without stopping. Show the stills contact sheet in chat before the full render, fix what it shows, then render." ], "setup": [ "Works on macOS, Windows and Linux: detect the OS first and use its shell (PowerShell on Windows), paths (.venv/Scripts on Windows, .venv/bin elsewhere) and python command. Create web/lib work/vo stills frames out.", "Python venv .venv with numpy Pillow imageio-ffmpeg; ffmpeg = imageio_ffmpeg.get_ffmpeg_exe(). Nothing system-wide: no Node, no Playwright, no Homebrew. Kokoro gets its own Python 3.12 in tts/ (stage B).

Questions

What is this films prompt for?

It is a Claude Opus motion design brief for “Vertical animated cartoon short about Casanova and a losing-proof betting system”. Paste it into Claude and keep the timings already written in the prompt.

Where is the original film?

The creator is @kirisando. Use View post to open the X video this prompt belongs to.

More films prompts