A versatile text-to-image skill supporting multiple AI models like MiniMax, DALL-E 3, and Stability AI for seamless visual content creation.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install publish-skill
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install publish-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Xiaoxia Image Gen skill is a powerful extension for Openclaw Skills designed to provide seamless text-to-image generation capabilities. By integrating leading models like MiniMax, OpenAI's DALL-E 3, and Stability AI, it allows users to create stunning visuals directly through natural language prompts.
This skill is ideal for developers, designers, and content creators who need high-quality AI-generated art, mascots, or realistic photography without switching between multiple platforms. It simplifies the creative workflow by providing a unified interface for the industry's most popular image generation engines.
To use this skill within Openclaw Skills, you must configure your preferred AI model by setting the corresponding environment variable in your terminal:
# MiniMax (Recommended for cost-performance)
set MINIMAX_API_KEY=your_key_here
# OpenAI DALL-E 3 (High quality)
set OPENAI_API_KEY=your_key_here
# Stability AI (Realistic style)
set STABILITY_API_KEY=your_key_here
The skill utilizes environment-based configuration and standard API response structures to manage image generation.
| Variable | Purpose | Model Support |
|---|---|---|
| MINIMAX_API_KEY | Primary authentication | MiniMax image-01 |
| OPENAI_API_KEY | Secondary authentication | OpenAI DALL-E 3 |
| STABILITY_API_KEY | Tertiary authentication | Stability AI |
All user prompts are handled as string inputs, and the resulting outputs are processed as image files or URL references depending on the model provider.
Loading
A systematic framework to identify cognitive blind spots in investment strategies through a structured four-question model.

A professional-grade skill for architecting and building modern Web3 decentralized applications using 2026 industry standards.

A streamlined utility for converting natural language messages into structured financial ledger entries for project tracking.

An automated financial reporting skill that generates CNY-based expense breakdowns and visual charts for ledger data.

A specialized migration and backup utility that packages an entire OpenClaw agent workspace into a single portable .soul archive.

A browser automation skill for performing interactive and creative tasks on Zhihu using structured and stable workflows.








































