comfyui-zimage — for Claude Code comfyui-zimage, comfyui-zimage-skill, community, for Claude Code, ide skills, image generation, Bash scripting, high-resolution imaging, Riko route, manga style, Claude Code

v1.0.0
GitHub

About this Skill

Perfect for Image Generation Agents needing high-quality image creation capabilities using Bash scripts. comfyui-zimage is a cutting-edge AI agent skill that utilizes Bash scripts to generate high-quality images, ideal for developers seeking automated image creation solutions.

Features

Generate high-quality images using Bash scripts
Automate image creation with quick and high-resolution options
Utilize Riko route for consistent character appearance and style
Extract prompts from images for customized generation
Apply Hojo and Tsukasa Hojo styles for pure manga-inspired images

# Core Topics

for Claude Code image generation Bash scripting high-resolution imaging Riko route manga style
norsizu norsizu
[0]
[0]
Updated: 3/31/2026

Agent Capability Analysis

The comfyui-zimage skill by norsizu is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance. Optimized for for Claude Code, image generation, Bash scripting.

Ideal Agent Persona

Perfect for Image Generation Agents needing high-quality image creation capabilities using Bash scripts.

Core Value

Empowers agents to generate high-resolution images using Python scripts, supporting various styles such as manga and realistic, with options for custom outfit types and scene descriptions, utilizing libraries like safetensors for LoRA overlays and style triggers.

↓ Capabilities Granted for comfyui-zimage

Generating high-quality images for professional results
Automating image creation with custom scene descriptions and outfit types
Creating images in various styles, including manga and realistic, with options for LoRA overlays and style triggers

! Prerequisites & Limits

  • Requires Bash scripts to be run directly, avoiding ToolSearch
  • Limited to specific outfit options, such as school_uniform and maid
  • Riko's characteristics, such as hairstyle and facial features, are fixed and only exist in generate_riko_*.py scripts
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

comfyui-zimage

Unlock high-quality image generation with comfyui-zimage, an AI agent skill for developers. Automate image creation with Bash scripts and achieve...

SKILL.md
Readonly

Z-Image Image Generation System

⚠️ Do not use ToolSearch! Run scripts directly with Bash.

⚠️ Keyword Understanding (Must Read!)

User SaysMeaningScript to Use
high quality, masterpiece, professionalGood Prompt quality*_quick.py
high-res, upscale, 4K, large imageHigh resolution*_highres.py or upscale_*.py
default (no specification)Quick preview*_quick.py

🔀 Step 1: Identify Route (Important! Check Priority)

⚠️ Route Priority Rules

PriorityKeywordRouteDescription
1 (Highest)RikorikoAlways use riko route when "Riko" is mentioned!
2Hojo, Tsukasa HojomangaPure manga style
3Satoshi, UrushiharamangaPure manga style

Key Understanding

"Riko in Hojo style" = riko route + Hojo LoRA overlay

Riko's fixed appearance (hairstyle, facial features) and outfit definitions only exist in generate_riko_*.py. If you use generate_manga_*.py, you'll lose Riko's characteristics!

⚠️ Prompt Extraction Rules When Using Riko

When user requests "Extract prompt from this image, then generate in Riko style":

Only extract these elements for --scene parameter:

  • ✅ Actions, poses (sitting, standing, looking at viewer...)
  • ✅ Emotions, expressions (smile, serious, melancholy...)
  • ✅ Scene, background (classroom, window, sunset...)
  • ✅ Lighting, atmosphere (soft lighting, dramatic shadow...)
  • ✅ Composition, angle (close-up, full body, from below...)

Must discard these elements (Riko has fixed definitions):

  • ❌ Hair color, hairstyle (blonde, ponytail, long hair...)
  • ❌ Skin tone (tanned, pale...)
  • ❌ Body type (slim, curvy...)
  • ❌ Facial features (blue eyes, big eyes...)
  • ❌ Clothing (unless user specifies maid instead of school_uniform)

🎨 Step 2: Execute Generation

DIR=/path/to/your/comfyui-zimage-skill

2.1 Riko Route (Use this whenever Riko is mentioned!)

bash
1# Pure Riko (realistic style)
2python3 $DIR/generate_riko_quick.py \
3  --outfit_type school_uniform \
4  --scene "[scene description]"
5
6# Riko + Hojo style (black and white manga-style Riko)
7python3 $DIR/generate_riko_quick.py \
8  --outfit_type school_uniform \
9  --scene "[scene description]" \
10  --lora "Z-image\\z-image-hojo.safetensors" \
11  --style_trigger "Zanshou_kin_Hojo, monochrome, greyscale"
12
13# Riko + Satoshi style (90s anime-style Riko)
14python3 $DIR/generate_riko_quick.py \
15  --outfit_type school_uniform \
16  --scene "[scene description]" \
17  --lora "Z-image\\z-image-satoshi.safetensors" \
18  --style_trigger "sato, 90s anime style" \
19  --style_tags "shiny skin, sparkle"

Outfit options: school_uniform (JK uniform) | maid (maid outfit) View options: --view_type full (full body) | --view_type upper (upper body)

2.2 Pure Manga Route (When Riko is not involved)

bash
1# Hojo style
2python3 $DIR/generate_manga_quick.py \
3  --lora "Z-image\\z-image-hojo.safetensors" \
4  --prompt "[full prompt]"
5
6# Satoshi style
7python3 $DIR/generate_manga_quick.py \
8  --lora "Z-image\\z-image-satoshi.safetensors" \
9  --prompt "[full prompt]"

2.3 High-Resolution Version (When user explicitly requests "high-res")

Replace *_quick.py with *_highres.py, requires timeout=300000

📤 Step 3: Display Results

html
1<a href="[image_url]"><img src="[image_url]" width="350"></a>

Save image_url for later upscaling!

🔍 Step 4: Upscale Existing Images

When user says "upscale" for an already generated image:

bash
1# Riko image upscale (no LoRA needed)
2python3 $DIR/upscale_riko.py --image_url "[image_url]"
3
4# Manga image upscale (needs original LoRA)
5python3 $DIR/upscale_manga.py --image_url "[image_url]" --lora "[original LoRA]"

📋 Quick Decision Table

User RequestRouteScriptLoRA
Draw Rikorikogenerate_riko_quick.pyNone
Riko in Hojo stylerikogenerate_riko_quick.pyhojo + style_trigger
Riko in Satoshi stylerikogenerate_riko_quick.pysatoshi + style_trigger
Mecha girl in Hojo stylemangagenerate_manga_quick.pyhojo
Cyberpunk in Satoshi stylemangagenerate_manga_quick.pysatoshi

⚠️ Notes

  1. Riko first — Always use riko route when "Riko" is mentioned
  2. LoRA path uses double backslash \\
  3. highres/upscale needs timeout=300000
  4. Style guides are in data/ — Do not delete

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is comfyui-zimage?

Perfect for Image Generation Agents needing high-quality image creation capabilities using Bash scripts. comfyui-zimage is a cutting-edge AI agent skill that utilizes Bash scripts to generate high-quality images, ideal for developers seeking automated image creation solutions.

How do I install comfyui-zimage?

Run the command: npx killer-skills add norsizu/comfyui-zimage-skill/comfyui-zimage. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for comfyui-zimage?

Key use cases include: Generating high-quality images for professional results, Automating image creation with custom scene descriptions and outfit types, Creating images in various styles, including manga and realistic, with options for LoRA overlays and style triggers.

Which IDEs are compatible with comfyui-zimage?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for comfyui-zimage?

Requires Bash scripts to be run directly, avoiding ToolSearch. Limited to specific outfit options, such as school_uniform and maid. Riko's characteristics, such as hairstyle and facial features, are fixed and only exist in generate_riko_*.py scripts.

↓ How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add norsizu/comfyui-zimage-skill/comfyui-zimage. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use comfyui-zimage immediately in the current project.

Related Skills

Looking for an alternative to comfyui-zimage or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

★ 149.6k
⑂ 0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

★ 138.4k
⑂ 0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

★ 72.8k
⑂ 0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

★ 72.8k
⑂ 0
AI