Zimage Skill

Name: Zimage Skill
Author: yizhiyanhua-ai

yizhiyanhua-ai/zimage-skill

58 installs
51 repo stars
Updated January 22, 2026
yizhiyanhua-ai/zimage-skill

Helps with ai & agent building tasks during AI-assisted development.

About

zimage-skill is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted coding.

zimage-skill
AI & Agent Building
AI-coding skill

Zimage Skill by the numbers

58 all-time installs (skills.sh)
+3 installs in the week ending Jul 28, 2026 (Skillselion tracking)
Ranked #6,589 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
Data as of Jul 28, 2026 (Skillselion catalog sync)

npx skills add https://github.com/yizhiyanhua-ai/zimage-skill --skill zimage-skill

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

[![Listed on Skillselion](https://skillselion.com/badge/skills/yizhiyanhua-ai/zimage-skill.svg)](https://skillselion.com/skills/yizhiyanhua-ai/zimage-skill)

Installs	58
repo stars	★ 51
Last updated	January 22, 2026
Repository	yizhiyanhua-ai/zimage-skill ↗

What it does

Helps with ai & agent building tasks during AI-assisted development.

Files

SKILL.mdMarkdownGitHub ↗

Z-Image Generation

Generate images using ModelScope's Tongyi-MAI/Z-Image-Turbo model.

Requirements

pip install requests Pillow

Usage

python3 ~/.claude/skills/zimage-skill/generate.py "prompt" [output_path]

Environment

Optional: Set MODELSCOPE_API_KEY to use your own API key.

Examples

python3 ~/.claude/skills/zimage-skill/generate.py "A golden cat" cat.jpg
python3 ~/.claude/skills/zimage-skill/generate.py "Sunset over mountains"

Notes

Default output: result_image.jpg
Timeout: 120 seconds max
Supports Chinese and English prompts

# Python
__pycache__/
*.py[cod]
*$py.class
*.so
.Python
*.egg-info/
dist/
build/
eggs/
*.egg

# Virtual environments
venv/
env/
.venv/

# IDE
.idea/
.vscode/
*.swp
*.swo

# OS
.DS_Store
Thumbs.db

# Generated images (optional - uncomment if you don't want to track generated images)
# *.jpg
# *.png
# result_image.*

#!/usr/bin/env python3
import requests, time, json, sys, os
from PIL import Image
from io import BytesIO

base_url = 'https://api-inference.modelscope.cn/'
api_key = os.environ.get("MODELSCOPE_API_KEY")
if not api_key:
    print("Error: MODELSCOPE_API_KEY environment variable is required", file=sys.stderr)
    print("Get your free API key at: https://modelscope.cn/my/myaccesstoken", file=sys.stderr)
    sys.exit(1)
headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
MAX_WAIT = 120  # seconds

prompt = sys.argv[1] if len(sys.argv) > 1 else "A golden cat"
output_path = sys.argv[2] if len(sys.argv) > 2 else "result_image.jpg"

response = requests.post(
    f"{base_url}v1/images/generations",
    headers={**headers, "X-ModelScope-Async-Mode": "true"},
    data=json.dumps({"model": "Tongyi-MAI/Z-Image-Turbo", "prompt": prompt}, ensure_ascii=False).encode('utf-8')
)
response.raise_for_status()
task_id = response.json()["task_id"]
print(f"Task started: {task_id}")

start = time.time()
while time.time() - start < MAX_WAIT:
    result = requests.get(f"{base_url}v1/tasks/{task_id}", headers={**headers, "X-ModelScope-Task-Type": "image_generation"})
    result.raise_for_status()
    data = result.json()
    if data["task_status"] == "SUCCEED":
        Image.open(BytesIO(requests.get(data["output_images"][0]).content)).save(output_path)
        print(f"Image saved to: {output_path}")
        sys.exit(0)
    elif data["task_status"] == "FAILED":
        print(f"Failed: {data.get('error', 'Unknown error')}", file=sys.stderr)
        sys.exit(1)
    time.sleep(5)
print("Timeout waiting for image generation", file=sys.stderr)
sys.exit(1)

MIT License

Copyright (c) 2024

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Z-Image Skill - AI 图片生成技能

让 Claude Code 帮你用自然语言生成精美图片！

English | 中文

这是什么？

Z-Image Skill 是一个 Claude Code 技能插件，让你可以用简单的中文或英文描述，自动生成 AI 图片。

你只需要这样说：

"帮我生成一张金色的猫咪图片"
"画一幅日落下的山脉"
"创建一个可爱的卡通头像"

Claude Code 就会自动帮你生成图片！

效果展示

你：帮我生成一张赛博朋克风格的城市夜景

Claude：好的，我来帮你生成这张图片...
        Task started: xxx
        Image saved to: cyberpunk_city.jpg

        图片已生成！保存在 cyberpunk_city.jpg

安装步骤

所有步骤都可以在 Claude Code 中用自然语言完成，无需手动输入命令！

第一步：获取免费 API Key

1. 访问 ModelScope 官网 2. 点击右上角「登录/注册」 3. 登录后，访问 API Token 页面 4. 点击「创建 Token」，复制生成的 API Key

第二步：安装技能

打开 Claude Code，直接说：

帮我安装 zimage-skill，仓库地址是 https://github.com/yizhiyanhua-ai/zimage-skill

Claude 会自动帮你完成：

下载技能到正确的目录
安装所需的 Python 依赖

第三步：配置 API Key

在 Claude Code 中说：

帮我配置 MODELSCOPE_API_KEY 环境变量，值是 ms-xxxxxxxx（替换成你的 API Key）

或者你也可以说：

帮我在 ~/.claude/settings.json 中添加环境变量 MODELSCOPE_API_KEY，值是 ms-xxxxxxxx

第四步：验证安装

重启 Claude Code 后，直接说：

帮我生成一张测试图片

如果看到图片生成成功，说明安装完成！

使用案例

基础用法

你：生成一张可爱的柴犬图片
你：画一个宇航员在月球上
你：创建一张抽象艺术画

指定保存路径

你：生成一张海边日落的图片，保存为 sunset.jpg
你：画一只熊猫，保存到 ~/Pictures/panda.png

详细描述获得更好效果

你：生成一张图片：一只橘色的猫咪坐在窗台上，窗外是下雨的城市夜景，
    温暖的室内灯光，电影感的构图，4K高清

你：画一幅画：中国水墨画风格的山水，远处有云雾缭绕的山峰，
    近处有一叶扁舟，意境悠远

不同风格示例

描述	效果
"赛博朋克风格的东京街头"	霓虹灯、未来感
"吉卜力动画风格的森林"	宫崎骏风格
"油画风格的向日葵"	梵高风格
"极简主义的几何图案"	现代设计感
"中国风的锦鲤戏水"	国画风格
"像素艺术的游戏角色"	复古游戏风

实用场景

1. 制作头像

你：生成一个可爱的卡通头像，女生，粉色头发，大眼睛，微笑

2. 文章配图

你：我在写一篇关于人工智能的文章，帮我生成一张科技感的配图

3. 社交媒体素材

你：生成一张适合发朋友圈的风景图，要有治愈感

4. 设计灵感

你：画一个 App 启动页的背景，渐变色，抽象，现代感

5. 表情包制作

你：生成一只惊讶的猫咪表情，卡通风格，夸张表情

常见问题

Q: 提示 "MODELSCOPE_API_KEY environment variable is required"

A: 你还没有配置 API Key。请按照「安装步骤」中的第一步和第二步操作。

Q: 图片生成失败或超时

A: 可能的原因：

网络连接问题，请检查网络
API 服务繁忙，请稍后重试
描述内容可能触发了内容审核，请修改描述

Q: 生成的图片质量不理想

A: 尝试以下方法：

使用更详细的描述
添加风格关键词（如：高清、4K、电影感、专业摄影）
指定艺术风格（如：油画风格、水彩风格、动漫风格）

Q: 支持哪些图片格式？

A: 默认保存为 JPG 格式，也支持 PNG 格式（在文件名中指定即可）。

技术信息

模型: Tongyi-MAI/Z-Image-Turbo
API 提供商: ModelScope (阿里云)
超时时间: 120 秒
支持语言: 中文、英文

获取帮助

如果遇到问题，可以： 1. 查看上方「常见问题」 2. 在 GitHub 提交 Issue 3. 检查 API Key 是否正确配置

许可证

MIT License - 可自由使用和修改

Z-Image Skill - AI Image Generation for Claude Code

Generate beautiful images with natural language in Claude Code!

English | 中文

What is this?

Z-Image Skill is a Claude Code skill plugin that lets you generate AI images using simple natural language descriptions.

Just say things like:

"Generate a picture of a golden cat"
"Draw a sunset over mountains"
"Create a cute cartoon avatar"

And Claude Code will automatically generate the image for you!

Demo

You: Generate a cyberpunk style city at night

Claude: I'll generate that image for you...
        Task started: xxx
        Image saved to: cyberpunk_city.jpg

        Done! The image has been saved to cyberpunk_city.jpg

Installation

All steps can be completed using natural language in Claude Code - no manual commands needed!

Step 1: Get Your Free API Key

1. Visit ModelScope 2. Click "Login/Register" in the top right corner 3. After logging in, go to API Token page 4. Click "Create Token" and copy your API Key

Step 2: Install the Skill

Open Claude Code and simply say:

Install zimage-skill from https://github.com/yizhiyanhua-ai/zimage-skill

Claude will automatically:

Download the skill to the correct directory
Install the required Python dependencies

Step 3: Configure API Key

In Claude Code, say:

Configure MODELSCOPE_API_KEY environment variable with value ms-xxxxxxxx (replace with your API Key)

Or you can say:

Add MODELSCOPE_API_KEY environment variable to ~/.claude/settings.json with value ms-xxxxxxxx

Step 4: Verify Installation

After restarting Claude Code, simply say:

Generate a test image for me

If you see an image generated successfully, you're all set!

Usage Examples

Basic Usage

You: Generate a cute Shiba Inu picture
You: Draw an astronaut on the moon
You: Create an abstract art painting

Specify Output Path

You: Generate a beach sunset image, save as sunset.jpg
You: Draw a panda, save to ~/Pictures/panda.png

Detailed Descriptions for Better Results

You: Generate an image: an orange cat sitting on a windowsill,
     rainy city night outside, warm indoor lighting,
     cinematic composition, 4K high definition

You: Create a painting: Chinese ink wash style landscape,
     misty mountains in the distance, a small boat nearby,
     serene and peaceful atmosphere

Style Examples

Description	Result
"Cyberpunk Tokyo street"	Neon lights, futuristic
"Ghibli-style forest"	Studio Ghibli aesthetic
"Oil painting of sunflowers"	Van Gogh style
"Minimalist geometric pattern"	Modern design
"Chinese style koi fish"	Traditional Chinese art
"Pixel art game character"	Retro gaming style

Practical Use Cases

1. Create Avatars

You: Generate a cute cartoon avatar, girl with pink hair, big eyes, smiling

2. Article Illustrations

You: I'm writing an article about AI, generate a tech-themed illustration

3. Social Media Content

You: Generate a scenic landscape photo suitable for Instagram, calming vibes

4. Design Inspiration

You: Draw an app splash screen background, gradient colors, abstract, modern

5. Meme Creation

You: Generate a surprised cat face, cartoon style, exaggerated expression

FAQ

Q: Getting "MODELSCOPE_API_KEY environment variable is required"

A: You haven't configured your API Key yet. Follow Steps 1 and 2 in the Installation section.

Q: Image generation fails or times out

A: Possible reasons:

Network connection issues - check your internet
API service is busy - try again later
Content may have triggered moderation - modify your description

Q: Generated image quality is not ideal

A: Try these tips:

Use more detailed descriptions
Add quality keywords (e.g., "high definition", "4K", "cinematic", "professional photography")
Specify art style (e.g., "oil painting style", "watercolor", "anime style")