skills.rest
norsizu
norsizuCOMMUNITY
Explore all skills by this author →
GitHub

comfyui-zimage

Generate manga images via ComfyUI with scripted prompts and LoRA overlays.

19|3|Updated Jan 13, 2026
Search Tags:#automation#image#generate#LoRA#manga#ComfyUI#Alma
One-click install
npx skills add https://github.com/norsizu/comfyui-zimage-skill --skill comfyui-zimage
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfyui-zimage
Source: https://github.com/norsizu/comfyui-zimage-skill/tree/main
Command: npx skills add https://github.com/norsizu/comfyui-zimage-skill --skill comfyui-zimage

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables local manga/illustration image generation via a ComfyUI server using scripted prompts and LoRA overlays.

Core Features & Use Cases

  • Fast, script-driven generation: Quick manga, high-res, and Riko-style outputs with Hojo and Satoshi styles.
  • LoRA-aware styling: Apply LoRA overlays for style transfer and outfit customization.
  • Local-first workflow: No ToolSearch; run scripts directly against your ComfyUI instance and save results locally.
  • Upscaling and high-res options: Generate production-ready images with high resolution and autoupscale features.

Quick Start

Run a quick manga generation example: python3 generate_manga_quick.py --lora "Z-image\z-image-hojo.safetensors" --prompt "Zanshou_kin_Hojo, 1girl, solo, office lady" Then view the image in downloads.

Frequently Asked Questions about comfyui-zimage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate local manga images using ComfyUI and LoRA models?▼

You generate local manga images by running Python scripts against a local ComfyUI server, composing prompts and applying LoRA model overlays to fetch rendered illustrations without relying on cloud tools.

What is the best way to apply style overlays for manga illustration generation?▼

The best way to apply style overlays is using LoRA-aware styling within your local ComfyUI workflow, enabling style transfer and outfit customization for high-resolution manga outputs.

Can I use my own LoRA files for outfit customization in ComfyUI?▼

Yes, you can use your own LoRA files for outfit customization. The skill applies LoRA overlays via script prompts, allowing you to specify local model paths like z-image-hojo.safetensors for tailored character styling.

Do I need a cloud service to run ComfyUI for high-res manga generation?▼

No, you do not need a cloud service. This local-first workflow runs Python scripts directly against your ComfyUI instance, utilizing high resolution and autoupscale features to save production-ready images locally.

What are the limitations of running script-driven manga generation locally?▼

A limitation of local script-driven manga generation is the requirement for a local ComfyUI server and specific LoRA model files. It operates without ToolSearch, meaning you must manage scripts and local image storage manually.

Related Skills

Explore More Design & Creative Skills→
erfianugraherfianugrah
1

lora-train

Automate SDXL or Flux LoRA training, captioning, and ComfyUI deployment via HTTP proxy.

Community
Intermediate
forcedotcomforcedotcom
803

generating-visual-diagrams

Generate rendered ERDs, UI mockups, and architecture diagrams for Salesforce.

Official
Intermediate
LamseyahEliasLamseyahElias

comfyui

Run ComfyUI workflows programmatically via REST API and WebSocket.

Community
Advanced
czh-ee-2023czh-ee-2023

run-experiment

Deploy and execute machine learning experiments across local and remote environments.

Community
Advanced