A high-performance image generation and editing skill powered by Gemini 3 Pro Image via OpenRouter.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install nanobananapro-openrouter
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 nanobananapro-openrouter 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 Nano Banana Pro?
Nano Banana Pro is a sophisticated technical skill designed for AI agents to interface with the Gemini 3 Pro Image model. It provides a robust framework for both creating high-fidelity original artwork and performing precise edits on existing images using natural language instructions. This tool is a cornerstone for developers looking to integrate professional-grade visual content creation into their Openclaw Skills ecosystem.
The skill is optimized for efficiency and quality, offering a tiered resolution system that scales from quick 1K drafts to production-ready 4K masterpieces. By leveraging OpenRouter, it ensures reliable access to state-of-the-art vision models while maintaining a developer-centric workflow through a streamlined Python and CLI interface.
Nano Banana Pro Use Cases
- Rapid prototyping of visual assets using a draft-to-final iterative workflow.
- Performing complex image-to-image modifications like adding objects, changing styles, or adjusting lighting.
- Automating the generation of SEO-friendly, timestamped media assets for technical documentation.
- Scaling image resolution dynamically based on project requirements from 1K to 4K.
How Nano Banana Pro Works
- The system receives a natural language prompt and optional input image reference from the user.
- The script validates the environment, checking for the UV package manager and OpenRouter API credentials.
- It maps user-specified quality levels to the appropriate API resolution parameters (1K, 2K, or 4K).
- A unique, timestamped filename is generated based on the prompt context to ensure organized storage.
- The request is dispatched to the Gemini 3 Pro Image model via OpenRouter's API.
- The resulting PNG is saved to the local output directory, and the absolute file path is returned to the agent.
Nano Banana Pro Setup
Ensure you have the uv package manager installed and an active OpenRouter API key. Set your API key as an environment variable:
export OPENROUTER_KEY="your_api_key_here"
To generate an image, run the script using the following structure:
uv run scripts/generate_image.py --prompt "a futuristic city" --filename "city.png" --resolution 4K
Nano Banana Pro Data Schema & Taxonomy
Nano Banana Pro utilizes a structured file management system to ensure data integrity across Openclaw Skills projects.
| Attribute | Description |
|---|---|
| Storage Path | Defaults to ./output_images/ unless --output-dir is specified. |
| Naming Convention | {yyyy-mm-dd-hh-mm-ss}-{slug}.png for precise version tracking. |
| Input Formats | Supports standard image formats for the --input-image parameter. |
| Metadata | Incorporates prompt context and resolution data into the generation lifecycle. |
Nano Banana Pro Advanced Features
- Iterative Workflow Support: Optimized logic for quick 1K iterations before committing to high-resource 4K renders.
- Intelligent Resolution Mapping: Automatically translates descriptive terms like "ultra" or "hi-res" into technical API parameters.
- Precision Editing Templates: Built-in prompt engineering templates to ensure the model preserves original subjects while applying specific changes.
- Flexible Directory Routing: Full support for custom output paths to integrate with existing project structures.
SKILL.md
Loading
Related Openclaw Skills
Edicts: Ground Truth for AI Agents
A lightweight ground-truth layer that injects verified facts directly into an AI agent's system context to eliminate hallucinations.
Humanizer
The humanizer skill refines AI-generated text by identifying and removing common algorithmic patterns to create more natural, authentic writing.
Knitting Pattern Search Skill
A specialized AI agent skill designed to programmatically find and retrieve knitting patterns for craft-focused workflows.
Stock Tools
A streamlined A-share market data tool for managing personal watchlists and fetching real-time price quotes through lightweight public APIs.
Google Postmaster Tools for AI Agents
A powerful integration for AI agents to monitor Gmail sender reputation and email deliverability metrics through the Membrane ecosystem.
Pointagram
A gamification platform skill that enables AI agents to manage employee engagement, challenges, and rewards via the Pointagram API.
Featured*
