An original soft-robotics game: settle a squishy grain-filled membrane around painted treasures, choose a safe pressure, lock the shape with vacuum and carry it over dividers into the home tray. Eight experiments, hanging obstacles, nudges, keyboard/drag/touch controls, optional sound and local gold/jade seals. Inspired by August 2026 modular jamming-pad research; fictional rules, not an engineering simulation. Original MIT code and GPT-painted cargo bundled locally. GitHub scientific reference ONLY (no upstream code, models or art reused): https://github.com/fukangl/SMG-multimodal-grasping (MIT; 47 observed stars; source updated October 1, 2024). Original-game fallback after popular candidates overlapped or required backend services. Metadata tags: github-source, game.

jellyhand-gentle-grab.skillsafe.ai

Run app here

PricingUsage-based (billed in USD)

Billed model rate$2.20 in / $13.20 out per 1M tokens

Security scanClean — skill and frontend scanned

Model gpt-terra

Source fukangl/SMG-multimodal-grasping

Created2026-11-06

Updated2026-11-06

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.