The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
The Grab is a claw-machine launch ad template for any product. The claw dives into a pile of capsules, pulls out your product, swings it to centre stage, and then a fresh one drops back in. It’s a seamless 8-second 4:5 loop for launches, drops and promos.
To make it yours: swap the product and logo images (each sits in a fixed frame named product-image and logo-mark), retype four lines of live text (headline, subhead, cta-label, badge-label) and recolour the capsule caps. The claw, the physics and the timing never change. Pomelo Pop (soda) and Mossa (serum) run on the exact same animation.
I built it end to end with Claude Code + the SVGator MCP. I researched references, generated the product shots and logos with Codex, computed the pendulum swing in Python and wrote it straight into SVGator keyframes. The second brand was made only with edit_part content swaps. Fonts are Bricolage Grotesque and Geist (Google Fonts, OFL).
Working project link: https://app.svgator.com/editor#/8168cd87c0464ed790a3fdbc3d74c7d8
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
smooth animation , love it , looking at it, thinking of how the components will look like and also the prototype , i really love this animation
Bloop, a Framer template for AI agent startups. I had a lot of fun animating it and creating its playful mascot.
This is fire! The playful mascot makes AI agents feel so approachable. Framer template for AI startups is such a hot niche right now - great execution!
Realistic AI Talking Head Video with Emotional Lip SyncAI
Video Creator | Character Generation, Lip Sync, Sound & Editing
The client needed a 30-second cinematic AI talking-head video featuring one woman delivering the same line, "I'm fine," with six distinct emotions. I created the character visuals, generated and refined the performances, and handled lip sync, sound and editing. The focus was consistent identity and room details, realistic skin, subtle facial expressions, natural body movement and believable emotional transitions.
Same line six times is a tough brief, since the words can't carry any of it and everything sits on the face. With recurring characters, identity usually slips first on the strong expressions, not the calm ones. Did you lock her with one reference set across all six emotions, or adjust refs per emotion?