nano-banana-pro
Verifiedby elizaOS · Updated Apr 16, 2026
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro). Use when the user asks to create an image, generate a picture, produce AI-generated artwork, edit a photo, compose multiple images, or upscale an image to higher resolution. Supports text-to-image generation, single-image editing, and multi-image composition using the Gemini API.
See It In Action
Now let me look at the input images to understand what we're working with:
Now I have a clear picture of the three source assets. Let me compose them into the hero mockup using the generate_image script:
The model name needs updating. Let me check available models and fix the script:
I see gemini-3-pro-image-preview is available. Let me update the model name and run the composition:
The image was generated successfully. Let me verify the output:
The hero mockup is done and saved. Here's a summary:
Saved → assets/2026-04-15-hero-mockup.png (2048 × 2048, ~1.9 MB)
The final image composites all three source assets:
ui-screenshot.png— the latest UI screen placed inside the devicedevice-frame.png— the device frame wrapping the screenshotbranded-background.png— the branded background behind the device
The result is a polished hero mockup with the device centered on the branded background with a subtle shadow and depth — ready for this afternoon's design review.
Note: I also updated
scripts/generate_image.pyto use the currentgemini-3-pro-image-previewmodel (the previousgemini-2.0-flash-preview-image-generationmodel was no longer available).
External Tools
| Tool | Type |
|---|---|
| uv | binary |
Permissions
| Scope | Description |
|---|---|
| filesystem:read | |
| filesystem:write | |
| network:outbound | |
| process:spawn | |
| env:read |
SKILL.md
Nano Banana Pro (Gemini 3 Pro Image)
Use the bundled script to generate or edit images.
Generate
uv run {baseDir}/scripts/generate_image.py --prompt "your image description" --filename "output.png" --resolution 1K
Edit (single image)
uv run {baseDir}/scripts/generate_image.py --prompt "edit instructions" --filename "output.png" -i "/path/in.png" --resolution 2K
Multi-image composition (up to 14 images)
uv run {baseDir}/scripts/generate_image.py --prompt "combine these into one scene" --filename "output.png" -i img1.png -i img2.png -i img3.png
API key
GEMINI_API_KEYenv var- Or set
skills."nano-banana-pro".apiKey/skills."nano-banana-pro".env.GEMINI_API_KEYin~/.otto/otto.json
Notes
- Resolutions:
1K(default),2K,4K. - Use timestamps in filenames:
yyyy-mm-dd-hh-mm-ss-name.png. - The script prints a
MEDIA:line for Otto to auto-attach on supported chat providers. - Do not read the image back; report the saved path only.
FAQ
What does nano-banana-pro do?
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro). Use when the user asks to create an image, generate a picture, produce AI-generated artwork, edit a photo, compose multiple images, or upscale an image to higher resolution. Supports text-to-image generation, single-image editing, and multi-image composition using the Gemini API.
When should I use nano-banana-pro?
Use it when you need a repeatable workflow that produces image output, code diff.
What does nano-banana-pro output?
In the evaluated run it produced image output, code diff.
How do I install or invoke nano-banana-pro?
Ask the agent to use this skill when the task matches its documented workflow.
Which agents does nano-banana-pro support?
Agent support is inferred from the source, but not explicitly declared.
What tools, channels, or permissions does nano-banana-pro need?
It uses uv; channels commonly include image, diff; permissions include filesystem:read, filesystem:write, network:outbound, process:spawn, env:read.
Is nano-banana-pro safe to install?
Static analysis marked this skill as medium risk; review side effects and permissions before enabling it.
How is nano-banana-pro different from an MCP or plugin?
A skill packages instructions and workflow conventions; tools, MCP servers, and plugins are dependencies the skill may call during execution.
Does nano-banana-pro outperform not using a skill?
About nano-banana-pro
When to use nano-banana-pro
When you need to create an AI-generated image from a text prompt. When you want to edit or upscale an existing image using Gemini. When you need to combine multiple input images into a single composed output.
When nano-banana-pro is not the right choice
When you need vector design editing or manual graphic design workflows. When network access or a Gemini API key is unavailable.
What it produces
Produces image output and code diff.