Prompt Motion
@AiMotionSiteS
Prompt library

Explainers

90-second factainment explainer drawn entirely in code

90-second factainment explainer drawn entirely in code is a 92.4-second Claude Opus motion design reference by @laurgrowth, built around Three.js, AI voice. GOAL: <role> You're a motion designer and a fact-checker. Copy this explainer motion prompt, keep the creator's frame counts and palette, and adapt the subject for your own launch film.

@laurgrowthView post

Prompt

GOAL: <role> You're a motion designer and a fact-checker. MOTION: Make a 60-90s vertical factainment explainer drawn entirely in code. No AI image or video models. </role> <inputs> Ask me for: the topic as a "what if" question, the length (default 90s), and whether I'll send an ElevenLabs voiceover or you should use a free local voice. </inputs> <research> Find 5-7 real events with real numbers. Every number needs a source. Show me the sources. </research> Structure: question hook in the first 2s -> "sounds impossible? it already happens" + one real example -> an escalation (H+1, H+3, H+6...) with one real event and one on-screen number per beat -> a twist that reframes everything -> end on a question with 3 answers for the comments. One line = one visual. Spell numbers out for the voice. Show me the script before you build anything. Low-poly 3D in three.js. Near-black navy background, glowing yellow and electric blue accents, red for bad news. Dot-matrix LED boards at the top for every key number. Counters count up, never cut. A globe made of glowing dots, real country shapes for maps. Light bloom, film grain, vignette. Captions 1-3 words in the bottom third, key word in yellow, rising in. Hard cut on every new line with a small zoom-out punch. <timing> When I send the voiceover, get word timestamps with forced alignment and fire every cut, counter and caption on its exact word. </timing> <build> One HTML file where https://t.co/IfGGJ8odPH(t) draws any frame as a pure function of time. Render every frame with Playwright at 1080x1920, 30fps. Encode with ffmpeg. </build> <sound> Low drone bed, a soft pulse during the escalation, a whoosh on every cut, blips on the LED boards, booms on the big reveals. Duck the bed under the voice. Master to -14 LUFS. </sound> <check> Before the full render, show me 12 stills on one contact sheet. Fix anything cropped, overlapping or too dark before you render. </check> <deliver> A captioned mp4, a clean mp4 with no captions for CapCut, and a music + sound effects track.
Author original
<role> You're a motion designer and a fact-checker. Make a 60-90s vertical factainment explainer drawn entirely in code. No AI image or video models. </role> <inputs> Ask me for: the topic as a "what if" question, the length (default 90s), and whether I'll send an ElevenLabs voiceover or you should use a free local voice. </inputs> <research> Find 5-7 real events with real numbers. Every number needs a source. Show me the sources. </research> Structure: question hook in the first 2s -> "sounds impossible? it already happens" + one real example -> an escalation (H+1, H+3, H+6...) with one real event and one on-screen number per beat -> a twist that reframes everything -> end on a question with 3 answers for the comments. One line = one visual. Spell numbers out for the voice. Show me the script before you build anything. Low-poly 3D in three.js. Near-black navy background, glowing yellow and electric blue accents, red for bad news. Dot-matrix LED boards at the top for every key number. Counters count up, never cut. A globe made of glowing dots, real country shapes for maps. Light bloom, film grain, vignette. Captions 1-3 words in the bottom third, key word in yellow, rising in. Hard cut on every new line with a small zoom-out punch. <timing> When I send the voiceover, get word timestamps with forced alignment and fire every cut, counter and caption on its exact word. </timing> <build> One HTML file where https://t.co/IfGGJ8odPH(t) draws any frame as a pure function of time. Render every frame with Playwright at 1080x1920, 30fps. Encode with ffmpeg. </build> <sound> Low drone bed, a soft pulse during the escalation, a whoosh on every cut, blips on the LED boards, booms on the big reveals. Duck the bed under the voice. Master to -14 LUFS. </sound> <check> Before the full render, show me 12 stills on one contact sheet. Fix anything cropped, overlapping or too dark before you render. </check> <deliver> A captioned mp4, a clean mp4 with no captions for CapCut, and a music + sound effects track.

Questions

What is this explainers prompt for?

It is a Claude Opus motion design brief for “90-second factainment explainer drawn entirely in code”. Paste it into Claude and keep the timings already written in the prompt.

Where is the original film?

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

More explainers prompts