- Home
- Skills
- Image & Video
- motion-spot
Motion Spot
Agency-level motion graphics spots made 100% in code: product and app ads, launches, brand intros and showreels with real motion blur, voiceover and music.
Free
Works with the AI Agents you already use
motion-spot
Example session with this skill installed
Create a 15-second 9:16 vertical spot for a new task app called 'Flow'. Use a minimalist Apple-style aesthetic, ElevenLabs voice, and show a notification center UI.
- Read your context and instructions
- Compiled the motion-spot
I have initialized the 'flow-spot' project using the Apple-style archetype. The concept focuses on "The friction of mental load vs. the clarity of Flow." I've prepared the structure in escenas.js and generated the initial visual system.
Motion Spot Deliverable: Flow Task App
Visual Direction:
- Typography: SF Pro Display (System) with normal tracking; accent words in Instrument Serif italic.
- Color Palette:
#F5F5F7(Off-white),#0071E3(Apple Blue),#1D1D1F(Dark text). - Core Techniques: Camera drift, Gaussian blur transitions, and 3D voxel-style notification stacks.
// scenes.js snippet - Notification Reveal
spot.escena('intro', (capa, c) => {
const hero = capa.querySelector('.notification-stack');
const camara = spot.camara();
// 3D Stack Entry
gsap.fromTo(hero, {
z: -500,
rotationX: -45,
opacity: 0
}, {
z: 0,
rotationX: 0,
opacity: 1,
duration: 1.2,
ease: "expo.out"
});
// Sync with voiceover word "clutter"
spot.cuando('intro', 'clutter', () => {
gsap.to('.bg-blur', { backdropFilter: 'blur(20px)', duration: 0.8 });
});
});
Next steps
- Drop your app logo (SVG) and 3 interface screenshots into the
/activosfolder. - Run
node motor/preparar.mjs --borrador-vozto generate the timing file. - Run
node motor/render.mjs --fotosto review the 12-frame contact sheet.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop settling for generic, static AI-generated videos. This skill converts your product, app, or brand assets into studio-grade motion graphics spots (10 to 45 seconds) using a code-driven rendering engine.
What it does
- Code-based animation using GSAP, Three.js, and CSS 3D for frame-perfect, agency-level visuals.
- Dynamic UI recreation to flip, stretch, or orbit your actual app interfaces instead of using generic templates.
- Automated narration sync that matches kinetic typography and scene transitions to the exact millisecond of your voiceover.
- Physical camera motion with real motion blur, grain, and glow effects rendered frame-by-frame via headless Chrome and ffmpeg.
- Audio orchestration including ElevenLabs voice integration and rhythm-matched music composition.
How it works
- Init project: Run the setup script to scaffold the directory and install the rendering motor.
- Define concept: Select an archetype (e.g., Apple-style product reveal) and write a script in
guion.json. - Draft scenes: Build the visual system in
estilo.cssand define the 3D/2D animation layers inescenas.js. - Sync & Render: Align animations to voice cues and run the render script to produce a high-quality MP4.
Frameworks & tools
Built for Node.js environments utilizing GSAP for timelines, Three.js for 3D elements, ElevenLabs for high-fidelity voice, and ffmpeg for final video encoding.
Why this beats prompting it yourself
Standard LLM prompts result in "look Claude" videos—tight Inter fonts and generic layouts. This skill uses a specialized motor and 45 analyzed agency references to enforce professional camera paths, brand-specific typography, and frame-by-frame precision that standard video generation tools cannot replicate.
Use cases
- App Launch Promos: Recreate the mobile UI in 3D to show off a new feature flow.
- Brand Showreels: Animate logos and brand assets for social media or headers.
- Product Spec Ads: Create high-tension visuals for physical goods using photos and SVG assets.
- Web-to-Video: Turn a landing page URL into a fast-paced kinetic marketing spot.
Known limitations
Requires Node.js, Chrome, and ffmpeg installed on the runner. Real-time audio playback is only available via the local preview server, not during the rendering process.
Requires Node.js, headless Chrome, and ffmpeg. Does not support real-time CSS animations or Math.random; all motion must be a function of time (t).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 day ago
- Free with an Agensi account
Requires