A lightweight utility to generate and edit images using Gemini models through an OpenAI-compatible interface.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install gemini-image-proxy
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install gemini-image-proxy using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Gemini Image Proxy?
Gemini Image Proxy is a specialized tool designed to bring the power of Gemini 3 Pro Image generation to the Openclaw Skills ecosystem. By utilizing the OpenAI Python SDK, it offers a familiar interface for developers to create and modify visuals with minimal overhead. Unlike more complex implementations, this skill is optimized for portability, making it an ideal choice for containerized applications and lightweight cloud deployments where resource efficiency is paramount. It bridges the gap between Gemini's advanced visual models and the widely-used OpenAI API structure.
Gemini Image Proxy Use Cases
- Creating high-resolution photorealistic landscapes or minimalist product shots from text prompts.
- Editing existing images to change lighting, style, or add new visual elements.
- Integrating automated image generation into existing Openclaw Skills pipelines.
- Deploying AI image capabilities on cloud platforms with restricted library support or limited resources.
How Gemini Image Proxy Works
- The user executes the generation script providing a text prompt and an optional source image path for edits.
- The skill authenticates with the external API using the provided proxy API key and base URL environment variables.
- The script utilizes the OpenAI Python SDK's images.generate or images.edits methods to communicate with the Gemini model.
- The resulting image is processed and saved directly to the specified output file path on the local system.
Gemini Image Proxy Setup
To get started with this skill in your Openclaw Skills environment, install the necessary dependency via pip:
python3 -m pip install openai
Configure your environment variables to connect to the Gemini proxy:
export GOOGLE_PROXY_API_KEY="your_api_key"
export GOOGLE_PROXY_BASE_URL="https://example.com/v1"
Gemini Image Proxy Data Schema & Taxonomy
The skill manages data through CLI arguments and standard environment configuration as detailed below:
| Data Type | Key/Arg | Description |
|---|---|---|
| Environment | GOOGLE_PROXY_API_KEY | Authentication key for the proxy service |
| Environment | GOOGLE_PROXY_BASE_URL | The OpenAI-compatible API endpoint URL |
| Input | Prompt | The text description of the image to create or edit |
| Input | --input | Optional path to an existing image (PNG, JPG, WEBP) |
| Output | File | The generated or edited image saved to the specified path |
Gemini Image Proxy Advanced Features
- Support for multiple high-performance models including Gemini 3 Pro, Imagen 4.0, and Gemini 2.5 Flash.
- Full compatibility with any OpenAI-compatible API proxy, enhancing flexibility within Openclaw Skills setups.
- Robust support for various input formats including PNG, JPG, JPEG, GIF, and WEBP for image editing tasks.
- Lightweight footprint optimized for high performance in containerized environments and Fly.io deployments.
SKILL.md
Loading
Related Openclaw Skills
Ollama Local Model Integration
A skill that integrates local Ollama models into your AI workflow for private, offline intelligence.
Bocha Search
An AI-powered search tool optimized for high-quality Chinese content, real-time news, and intelligent summarization.
Feishu Task Management
A powerful AI agent skill designed to programmatically manage Feishu tasks, supporting full CRUD operations through the Feishu Open Platform API.
OpenClaw Codex GPT-5.4 Enable Skill
A configuration-based patch to enable the openai-codex/gpt-5.4 model in OpenClaw without requiring a full application rebuild.
Library Book Monitor
A smart automation tool to monitor library book availability and send notifications when books become available for borrowing.
Library Book Monitor
An automated library tracking tool that notifies you the moment a desired book becomes available for borrowing.
Featured*
