Click on any object in a photo to cut it out: left-click adds a positive point, right-click adds a negative point, and a Segment-Anything model running entirely in your browser draws the mask. The cutout updates live after every click; add more objects, undo, then pick a background (transparent, white, blurred photo, colour, gradient or your own image) and download a PNG or JPG, or collage several cutouts on the design canvas. Two models to choose from: SlimSAM-77 (default, fast, about 17 MB) or the full SAM ViT-Base (higher quality, about 190 MB). Photos never leave your device. Free, no sign-up. Interface in 11 languages (EN, 简体中文, 日本語, 한국어, ES, PT, FR, DE, RU, ID, VI). Built on Meta's Segment Anything and SlimSAM (Apache-2.0, Xenova's ONNX exports) and onnxruntime-web (MIT), with weights served from SkillSafe's shared model registry.

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.