← all repositories
AHEKOT/ComfyUI_VNCCS

Four Button Presses to a Consistent Visual-Novel Cast

Creating a unique character that looks identical across dozens of poses and outfits is harder than generating one pretty picture; this ComfyUI suite treats that consistency problem as a full production pipeline.

ComfyUI_VNCCS
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history

What it does

VNCCS is a ComfyUI custom-node suite that turns diffuse image generation into a structured sprite factory for visual novels. It chains together character creation, pose generation, clothing design, and emotion variation into separate JSON workflows, then deposits the results into an organized output folder. The goal is to keep a single character’s face, proportions, and style locked across dozens of renders without manual Photoshop work.

The interesting bit

Instead of treating consistency as a prompting hack, the project breaks the pipeline into discrete assembly stages—create, clone, dress, emote—each with its own workflow file and a wizard that auto-fills tags from a plain description. It even includes an NSFW toggle for base sprite generation, acknowledging the reality of the asset packs people actually need.

Key highlights

  • Generates consistent character sprites across poses, outfits, and emotions using either the Illustrious or Anima models
  • Can clone existing characters from screenshots, downloaded art, or hand-drawn images into the same consistent pipeline
  • Pose Studio supports custom poses imported from reference images, with adjustable age, height, and body type to match the base character
  • Built-in background removal uses SAM3 detail recovery to avoid erasing eyes or clothing that match the chroma key color
  • Outputs everything to a structured ComfyUI/output/VNCCS/Characters/ hierarchy, ready for engine import

Caveats

  • Requires a separate VNCCS-Utils dependency to unlock the Pose Studio node
  • The newer Anima model is explicitly described as unstable for emotion generation and may drift from the base character design
  • The README leans heavily on anime-mascot narration, so technical details about node architecture are sparse

Verdict

Indie visual-novel developers who need a full sprite sheet but lack an art team will find a complete pipeline here. If you do not already live inside ComfyUI with a capable GPU, this is not your shortcut.

Frequently asked

What is AHEKOT/ComfyUI_VNCCS?
Creating a unique character that looks identical across dozens of poses and outfits is harder than generating one pretty picture; this ComfyUI suite treats that consistency problem as a full production pipeline.
Is ComfyUI_VNCCS open source?
Yes — AHEKOT/ComfyUI_VNCCS is open source, released under the MIT license.
What language is ComfyUI_VNCCS written in?
AHEKOT/ComfyUI_VNCCS is primarily written in Python.
How popular is ComfyUI_VNCCS?
AHEKOT/ComfyUI_VNCCS has 1.1k stars on GitHub.
Where can I find ComfyUI_VNCCS?
AHEKOT/ComfyUI_VNCCS is on GitHub at https://github.com/AHEKOT/ComfyUI_VNCCS.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.