skills.rest
jin-bo
jin-boCOMMUNITY
Explore all skills by this author →
GitHub

create_image

Generates custom visual assets from natural text prompts via multiple AI image APIs.

12|2|Updated May 14, 2026
Search Tags:#image generation#multi-model#text to image#prompt to image#ai image#aspect ratio#error retry
One-click install
npx skills add https://github.com/jin-bo/chahua --skill create-image-jin-bo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create_image
Source: https://github.com/jin-bo/chahua/tree/main/examples/personas/Yvonne/skills/create_image
Command: npx skills add https://github.com/jin-bo/chahua --skill create-image-jin-bo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python-dotenv, openai, google-genai, requests, dashscope, pillow, and includes scripts (resource) components.

What problem does it solve?

This Skill removes the friction of switching between disparate image generation tools and handling inconsistent parameter requirements across different AI models, enabling seamless creation of custom visual assets from natural language prompts.

Core Features & Use Cases

  • Multi-model support: Unified interface for 4 leading image generation backends (OpenAI gpt-image-2, Google Gemini 3.1 flash, TensorsLab Seedream v5, Alibaba Wan2.7-image-pro) with aligned core parameters.
  • Flexible customization: Control aspect ratio, output quality, resolution, file format, and number of generated images to match specific use case needs.
  • Robust error handling: Automatic retry for transient network or API errors, clear error messages for content moderation or invalid API keys, and context-aware output directory selection to keep generated assets organized.
  • Use case: When preparing a research report, generate a custom cover image directly saved to the report's asset folder without manual file management or switching between different image generation platforms.

Quick Start

Use the create_image skill to generate a 16:9 high-quality cyberpunk Tokyo street scene image and save it to the workspace/reports/cyberpunk-tokyo/ directory.

Frequently Asked Questions about create_image

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

FAQPage Schema
How do I generate images from text prompts using different AI models without switching platforms?▼

To generate images from text prompts across multiple AI models, you can use a unified interface that supports OpenAI, Google Gemini, TensorsLab, and Alibaba backends with aligned core parameters. This removes the friction of switching between disparate image generation tools.

Can I control the aspect ratio and output quality when generating AI images for a presentation?▼

You can control aspect ratio, output quality, resolution, file format, and the number of generated images for presentation design. This flexible customization ensures the generated visual assets match your specific use case needs.

What's the best way to automatically save generated images to a specific report directory?▼

The best way to save generated images to a specific report directory is using context-aware output directory selection. This feature automatically organizes generated visual assets into the correct workspace folders without manual file management.

Why do text to image API calls fail and how can I handle transient network errors?▼

Text to image API calls fail due to transient network issues, invalid API keys, or content moderation triggers. You can handle these using automatic retry for transient errors, while receiving clear error messages for authentication or content violations.

Do I need to manage API keys separately for each image generation backend?▼

You do not need to manage API keys separately for each backend during execution. The skill uses environment variable-based API key management to securely configure and authenticate requests across all supported image generation platforms.

Are there limitations when using multiple AI image generation models for batch asset creation?▼

Limitations when using multiple AI image generation models include encountering content moderation blocks or invalid API key errors. However, the system provides clear error messages for these issues and automatically retries transient network failures to ensure reliable batch asset creation.

Related Skills

Explore More Design & Creative Skills→
HuaiminHuangHuaiminHuang

image

Generate images via the MiniMax API from text prompts and reference images.

Community
Intermediate
gatiemegatieme
1

ppt-generation

Generate PowerPoint presentations with AI-generated images for each slide.

Community
Advanced
rmjamesrmjames
1

image-processor

Resize and crop raster and SVG images with ImageMagick and SVGO.

Community
Advanced
AICode-NexusAICode-Nexus
2

baoyu-image-gen

Generate images from text descriptions using multiple AI APIs.

Official
Advanced