Repolode
Submit
EBEBEEBE

EBEBEEBE/comfyui-lucid-vrp

A ComfyUI custom node pack for loading and editing images in a built-in canvas, plus visual regional prompting for Anima and MOSS-Image.

View on GitHub
Stars
40
Forks
3
Open issues
0
Last commit
7 days ago

What is it?

LuCiD VRP (Visual Regional Painter) is a set of ComfyUI custom nodes for image-to-image preparation and regional prompting. Its image loaders accept an imported image or a blank canvas, and the built-in LuCiD Canvas editor handles layers, painting, transforms, distortion and reversible cropping without leaving ComfyUI. Loaders output IMAGE and MASK values and can optionally encode latents with a connected VAE.

For composition control, users draw boxes over the canvas and assign each region its own prompt. Anima regional conditioning supports boxes or masks, positive and negative regional prompts, and adjustable influence. Experimental MOSS-Image support builds bbox layout prompts visually and can export native layout JSON. Regions can live inside a single node or be chained through separate VRP Region nodes.

The project targets classic ComfyUI node interfaces; Nodes 2.0 is stated as unsupported. Regional guidance is approximate, and inpainting, regional LoRA and built-in ControlNet are not included in this release.

Key features

  • 01Image loaders for empty latents or imported images, with resize, mask editing and undo/redo history.
  • 02LuCiD Canvas supports layers, painting, transforms, distortion and reversible cropping.
  • 03Visual box-based regional prompts for Anima with positive/negative prompts and adjustable influence.
  • 04Experimental MOSS-Image bbox layout prompting with native layout JSON export.
  • 05Regions can be kept in one node or chained through separate VRP Region nodes.

Use cases

  • Preparing an img2img source by compositing layers and mockups inside ComfyUI.
  • Laying out a scene with separate prompts for different regions before sampling.
  • Building a MOSS-Image layout mockup and reusing it as an img2img input for Anima.

In collections

AI image generators & Stable Diffusion tools

Topics

#comfyui#image-editing#img2img#regional-prompting#stable-diffusion#canvas#custom-nodes

FAQ

What is comfyui-lucid-vrp?
A ComfyUI custom node pack for loading and editing images in a built-in canvas, plus visual regional prompting for Anima and MOSS-Image. LuCiD VRP (Visual Regional Painter) is a set of ComfyUI custom nodes for image-to-image preparation and regional prompting. Its image loaders accept an imported image or a blank canvas, and the built-in LuCiD Canvas editor handles layers, painting, transforms, distortion and reversible cropping without leaving ComfyUI. Loaders output IMAGE and MASK values and can optionally encode latents with a connected VAE.
Is comfyui-lucid-vrp free to use? What license does it have?
comfyui-lucid-vrp is open source under the Other license. Check the license terms before using it in commercial products.
Is comfyui-lucid-vrp still maintained?
Actively maintained. The last commit to comfyui-lucid-vrp was on Oct 31, 2026.
What can I use comfyui-lucid-vrp for?
Preparing an img2img source by compositing layers and mockups inside ComfyUI. Laying out a scene with separate prompts for different regions before sampling. Building a MOSS-Image layout mockup and reusing it as an img2img input for Anima.
Where can I get comfyui-lucid-vrp?
The source code and installation instructions are on GitHub: https://github.com/EBEBEEBE/comfyui-lucid-vrp

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

Related projects