nano-banana
Generate Images with Google Gemini
Also available from: inference-sh-9,inference-shell,toolshell,tul-sh
Quickly create images using Google Gemini native image models through a simple CLI interface. This skill provides a convenient wrapper for the inference.sh platform to generate photorealistic images, illustrations, and artwork from text prompts.
Download the skill ZIP
Upload in Claude
Go to Settings → Capabilities → Skills → Upload skill
Toggle on and start using
Test it
Using "nano-banana". a futuristic city at sunset
Expected outcome:
A high-quality image of a futuristic city at sunset with flying cars and neon lights
Using "nano-banana". logo design for a coffee shop, minimalist
Expected outcome:
Four minimalist logo concepts for a coffee shop in various black and white styles
Using "nano-banana". Edit this image to add autumn colors
Expected outcome:
The original image modified with autumn foliage, warm orange and red tones, and seasonal atmosphere
Security Audit
SafeStatic analysis flagged external commands and network patterns in the SKILL.md file, but these are false positives. The 'external_commands' findings are bash command examples in code blocks (documentation), not actual code execution. The 'network' findings are legitimate URLs to the inference.sh service documentation. No cryptographic code or malicious patterns exist. This is a legitimate documentation file for an image generation skill.
Risk Factors
⚙️ External commands (31)
🌐 Network access (6)
Quality Score
What You Can Build
Quick Image Generation
Generate images from text prompts using Gemini models for immediate visual content
Custom Image Creation
Create images with specific aspect ratios, resolutions, and artistic styles
AI-Powered Image Editing
Edit existing images using Gemini's image understanding and generation capabilities
Try These Prompts
Generate a {style} image of {subject}Create {num_images} images with {aspect_ratio} aspect ratio showing {description}Edit this image to {modification}Generate {resolution} image with {style} style showing {subject}. Include {details}Best Practices
- Provide detailed prompts with specific styles, lighting conditions, and composition instructions
- Use appropriate aspect ratios for your intended output format (social media, print, etc.)
- Start with lower resolution for quick previews, then upgrade to 4K for final output
Avoid
- Using vague prompts without specific details or style guidance
- Requesting extremely high resolutions for initial generations
- Not accounting for model availability or rate limits