Repolode
Submit
lonely-dance-lab

lonely-dance-lab/cover-studio

A lightweight web workspace for AI image generation, splitting images into transparent layers, and layer-based editing for covers, posters and social graphics.

Open-source alternative to:CanvaAdobe Photoshop

View on GitHub
Stars
56
Forks
8
Open issues
0
Last commit
4 days ago

What is it?

Cover Studio is a browser-based tool for making covers and posters. You write a prompt to generate an image, then split that image into separate transparent layers, move and scale each layer, adjust stacking order, and export the result. It covers four creation modes: cover images, Xiaohongshu covers, e-commerce posters, and free-form work.

The project runs either as a local Python server with only the standard library, or on Cloudflare Workers with Static Assets. The frontend is plain HTML, CSS and JavaScript with no build step. Image generation and layer splitting call the MaaS image models, so users supply their own API key in the page. A bundled example poster with seven layers can be loaded without a key to try the editor.

The editor supports dragging, proportional scaling, show/hide, locking, renaming, drag-to-reorder, undo and redo. Export options include the original image, a composited PNG, single-layer PNGs, and a ZIP of the raw layers. Generated works are kept in a local output folder or the browser's IndexedDB, depending on the mode.

Key features

  • 01Generates images from text prompts with preset Punk cover and PONYO Xiaohongshu templates.
  • 02Splits uploaded PNG, JPEG or WebP images into editable transparent layers.
  • 03Supports dragging, scaling, hiding, locking and reordering layers with undo and redo.
  • 04Exports original images, composited PNGs, single-layer PNGs or a ZIP of raw layers.
  • 05Runs as a local Python server or on Cloudflare Workers without a frontend build.

Use cases

  • Designing cover images for articles or videos with editable text and graphic layers.
  • Producing Xiaohongshu-style covers from a template and prompt.
  • Creating e-commerce product posters and adjusting individual elements after splitting.
  • Self-hosting a private image generation and layer editing workspace with your own API key.

Topics

#ai-image-generation#image-editing#layers#poster-design#cloudflare-workers#python#web-app

FAQ

What is cover-studio?
A lightweight web workspace for AI image generation, splitting images into transparent layers, and layer-based editing for covers, posters and social graphics. Cover Studio is a browser-based tool for making covers and posters. You write a prompt to generate an image, then split that image into separate transparent layers, move and scale each layer, adjust stacking order, and export the result. It covers four creation modes: cover images, Xiaohongshu covers, e-commerce posters, and free-form work.
Is cover-studio free to use? What license does it have?
cover-studio has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
Is cover-studio still maintained?
Actively maintained. The last commit to cover-studio was on Oct 15, 2026.
What can I use cover-studio for?
Designing cover images for articles or videos with editable text and graphic layers. Producing Xiaohongshu-style covers from a template and prompt. Creating e-commerce product posters and adjusting individual elements after splitting. Self-hosting a private image generation and layer editing workspace with your own API key.
Where can I get cover-studio?
The source code and installation instructions are on GitHub: https://github.com/lonely-dance-lab/cover-studio

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects