← Back to Skills Marketplace
477
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install pony-image
Description
AI 商业图片生成工具，支持单图生成、风格复刻、套图生成三大能力
Usage Guidance
这个技能本身在目的和要求上内部一致：它会把你写的 prompt 和你上传的参考/产品图片（URL 或 Base64）POST 到一个固定的 Supabase 函数端点，并返回 Base64 格式的生成图片。你在决定是否安装/使用前应考虑：
- 该 BASE_URL（vecarpahagopuqbwxbjh.supabase.co）不是公共知名厂商域名，而是某个 Supabase 项目，意味着你将把数据发送到第三方后端；只在你信任该后端或已核实所有者时使用。
- 提供的 PONY_SUPABASE_ANON_KEY 为调用凭证：anon key 权限通常受限，但若该 Supabase 项目错误配置，key 仍可能用于读取/写入数据；避免将高敏感或受版权保护的图片/机密信息上传。
- 如果你管理自己的数据隐私，最好自行部署同类服务（创建自己的 Supabase 项目并替换 BASE_URL 和 anon key），或要求提供方出示来源/隐私声明和代码/托管仓库以便审查。
- 若担心滥用，可在使用时监控网络请求、为该 key 设置短期/可撤销凭证，并在不使用时撤销或更换它。
总体：功能上一致且实现方式透明，但风险来自后端信任和数据外发——在信任后端前不要上传敏感内容。
Capability Analysis
Type: OpenClaw Skill
Name: pony-image
Version: 1.2.0
The skill bundle is designed for AI image generation, requiring a Supabase API key (`PONY_SUPABASE_ANON_KEY`) for authentication. All network requests are directed to a single, hardcoded Supabase endpoint (`https://vecarpahagopuqbwxbjh.supabase.co/functions/v1`) via standard `curl` commands. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, obfuscation, or prompt injection attempts against the OpenClaw agent within the `SKILL.md` instructions. The instructions are purely functional, guiding the agent on how to interact with the image generation API.
Capability Assessment
Purpose & Capability
技能名/描述是图片生成（单图/风格复刻/套图），所需的唯一环境变量是 PONY_SUPABASE_ANON_KEY，SKILL.md 的所有示例请求都调用同一 BASE_URL 的 Supabase 函数端点，所请求的权限与功能一致。
Instruction Scope
运行时说明只要求将 prompt 和参考/产品图（URL 或 Base64）POST 到技能内定的 Supabase 函数端点并返回 Base64 图像；没有指示读取本地敏感配置或其它环境变量。但说明隐含会把用户上传的图片/描述传到第三方后端（固定的 BASE_URL），这会导致数据外发——功能内可预见但值得用户注意。
Install Mechanism
无安装脚本、无代码文件，纯说明文档——不会在本地写入或执行下载代码，安装面向风险最低的“instruction-only”类型。
Credentials
仅请求一个环境变量 PONY_SUPABASE_ANON_KEY，并将其用作 Authorization header。就功能而言这是合理的（用于调用 Supabase edge functions）。注意：Supabase anon key 通常权限受限但若项目配置不当可能泄露更多能力；此密钥与技能固定的 BASE_URL 绑定，泄露会让能调用该项目函数的主体伪装请求。
Persistence & Privilege
技能不是 always:true，且无安装步骤或修改其他技能/系统配置的指示；默认的自主调用能力未和其它高权限设置结合出现，因此不存在额外持久特权。
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install pony-image - After installation, invoke the skill by name or use
/pony-image - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.0
pony-image v1.0.0
- Initial release of pony-image skill.
- Provides AI-powered image generation with three main features: single image generation, style replication, and batch suite (themed set) generation.
- API authentication via SUPABASE_ANON_KEY required.
- Supports multiple styles and aspect ratio presets.
- Includes error handling and common troubleshooting guidance.
v1.1.2
**Skill pony-image 1.0.1 Changelog**
- Major documentation update: SKILL.md rewritten for clarity and easier onboarding.
- Added a quickstart decision tree and scene-based user guides for every use case.
- Introduced style and ratio presets tables for easier parameter selection.
- Expanded troubleshooting and FAQ with common errors and solutions.
- Highlighted authentication requirements and example flows for all features.
- Model/resolution compatibility clarified; 2K/4K now indicated as Pro model only.
v1.1.1
**Skill pony-image 1.0.1 Changelog**
- Major documentation update: SKILL.md rewritten for clarity and easier onboarding.
- Added a quickstart decision tree and scene-based user guides for every use case.
- Introduced style and ratio presets tables for easier parameter selection.
- Expanded troubleshooting and FAQ with common errors and solutions.
- Highlighted authentication requirements and example flows for all features.
- Model/resolution compatibility clarified; 2K/4K now indicated as Pro model only.
v1.0.0
Initial release of pony-image skill, providing REST APIs for image generation, suite planning, and style replication.
- Supports three main features: single image generation, suite image design with consistency, and image style replication.
- Exposes standard REST endpoints via Supabase Edge Functions, compatible with Open Claw and similar platforms.
- Offers customizable parameters for style, ratio, model selection, and advanced prompt enhancement.
- Implements robust error handling, retry logic, and support for multiple AI models.
- Results are saved to a dedicated Supabase table for persistence and later retrieval.
- Includes sample API usage, limitations, and best practices for integration.
Metadata
Frequently Asked Questions
What is pony-image?
AI 商业图片生成工具，支持单图生成、风格复刻、套图生成三大能力. It is an AI Agent Skill for Claude Code / OpenClaw, with 477 downloads so far.
How do I install pony-image?
Run "/install pony-image" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is pony-image free?
Yes, pony-image is completely free (open-source). You can download, install and use it at no cost.
Which platforms does pony-image support?
pony-image is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created pony-image?
It is built and maintained by lirule007 (@lirule007); the current version is v1.2.0.
More Skills