OpenViSTAI · Agents · Blockchain · Crypto
Menu
Image & Video AI

Magic Copy

Magic Copy is a browser extension and Figma plugin that uses Meta’s Segment Anything Model to extract a foreground object from an image and copy it to the clipboard.

Independent OpenViST project profile, based on linked sources.

Magic Copy — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Magic Copy when you want to select an image’s foreground subject and transfer the extracted result through the clipboard, especially as part of a browser or Figma workflow.

Who it is for

  • People who need to isolate an image subject for reuse in another document or design workflow.
  • Figma users who want foreground extraction within their design environment.
  • Developers who want to build the extension or configure its image-embedding service.

Documented uses

  • Extract a foreground object from an image and copy the result to the clipboard.
  • Use the image-extraction workflow through a browser extension or Figma plugin.
  • Build the extension from source or configure it to use a self-hosted embedding service.
Before you choose

Requirements & limits

The README lists Chrome and Firefox extension distribution and a Figma plugin. It warns that the Firefox add-on may still be under review. It also documents manual installation of the Chrome extension from a release ZIP.

  • The documented default workflow uses the same embedding service as the Segment Anything demo; the README notes that some users may prefer not to send images to a third party.
  • The repository’s self-hosting example is explicitly not intended for production use.
  • The project is open source under the MIT license and its primary language is TypeScript.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Configure Magic Copy to use a local embedding service

Build and run the repository’s example vision-transformer service, then point the Chrome extension to it.

Before you start

  • The Magic Copy repository and its server-example directory
  • Docker and a GPU-capable setup, as indicated by the documented Docker run command
  • The Chrome extension

Steps

  1. Build the service container from the server-example directory with the command below.

    Documented command
    docker build -t segment-anything .
  2. Run it with the command below.

    Documented command
    docker run --gpus all -p 8000:8000 segment-anything
  3. In the Magic Copy Chrome extension, change the endpoint to http://localhost:8000/.

Expected outcome

The extension is configured to request image embeddings from the locally running example service instead of the demo service.

Guide limitations

  • The repository describes the server as a proof of concept and says it is not intended for production use.
  • The service must accept an image file and return the documented embedding format.
Evidence and transparency

Sources & review

Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
kevmo314/magic-copy · Official homepage title · retained in its source language