nano-banana-pro

What This Skill Does

The Nano Banana Pro skill leverages the Gemini 3 Pro Image (Nano Banana Pro) AI model to generate and edit images. It provides a powerful command-line interface for creating novel imagery from text prompts, modifying existing images with specific instructions, and composing multiple images into a single scene. This skill is ideal for users who need to quickly create visual content for various purposes, from social media posts to concept art and design mockups.

Key functionalities include:

• Image Generation: Create entirely new images based on detailed textual descriptions.
• Image Editing: Modify existing images by providing instructions on how to alter them. This can include changing elements, styles, or adding new content.
• Multi-Image Composition: Seamlessly combine up to 14 existing images into a cohesive scene or collage, guided by a descriptive prompt.

Installation

To install the Nano Banana Pro skill, use the following command:

clawhub install openclaw/openclaw/skills/nano-banana-pro

This command fetches the necessary scripts and configurations from the openclaw/openclaw source repository.

Use Cases

• Content Creation: Generate unique visuals for blog posts, social media updates, presentations, or marketing materials.
• Design and Prototyping: Quickly visualize product ideas, design mockups, or explore different aesthetic concepts.
• Artistic Exploration: Create digital art, experiment with styles, and bring imaginative concepts to life.
• Image Enhancement: Edit and refine existing photographs or illustrations to better suit your needs.
• Collage and Compositing: Combine multiple images into a single, narrative piece or a professional-looking collage.

Example Prompts

Here are three examples of user prompts you could send to OpenClaw:

1. Generate an image of a cyberpunk city at sunset, neon lights reflecting on wet streets, digital art style, 4K resolution.
2. Edit the attached "my_logo.png" to have a subtle watercolor texture and a slightly darker color palette.
3. Combine "photo1.jpg", "photo2.jpg", and "photo3.jpg" into a single image depicting a serene forest scene with a stream.

Tips & Limitations

• API Key Configuration: Ensure your GEMINI_API_KEY is set either as an environment variable or within your ~/.openclaw/openclaw.json configuration file for authentication with the Gemini API.
• Resolution and Aspect Ratio: While default resolutions are 1K, you can specify 2K or 4K. For specific aspect ratios like avatars or banners, use the --aspect-ratio flag (e.g., 9:16, 16:9, 1:1). Without this flag, the model chooses the aspect ratio freely.
• Filenames: For better organization, use descriptive filenames with timestamps, such as yyyy-mm-dd-hh-mm-ss-yourimage.png.
• Output: The script will output a MEDIA: line which OpenClaw uses to automatically attach the generated image. The saved image path is reported, but the script does not read images back.
• Image Limits: For multi-image composition, you can include up to 14 images.
• Model Capabilities: Gemini 3 Pro Image (Nano Banana Pro) is a powerful model, but results can vary. Experiment with prompt phrasing to achieve desired outcomes. Complex edits or very specific artistic styles might require multiple iterations or more detailed prompts.