name	nanobanana
description	Generate or edit images using NanoBanana. Use when users request image generation or edit, want to create visual content, or mention nanobanana.

NanoBanana Image Generation

CLI tool for generating and editing images using Google's Gemini 3.1 Flash and Gemini 3 Pro models.

When to use this skill

User requests image generation or editing
User wants to create visual content or visualize real-time info
User mentions "nanobanana" or image generation
User specifies image parameters like aspect ratio or resolution

Command

nanobanana "prompt" [options]

Options

Flag	Description	Default
`-o`	Output file path	`nanobanana_TIMESTAMP.png`
`-a`	Aspect ratio (`1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`, `1:4`, `4:1`, `1:8`, `8:1`)	`1:1` (gen) / keep (edit)
`-r`	Resolution (`0.5K`, `1K`, `2K`, `4K`)	`1K`
`-e`	Edit existing image (input path)	-
`-ref`	Reference image(s) for style guidance (max 14)	-
`-s`	Google Search grounding for real-time info	off
`-i`	Google Image Search grounding (Flash only)	off
`-p`	Use Gemini 3 Pro model	off
`-t`	Thinking level: `minimal`, `high` (Flash only)	-

Examples

# Generate
nanobanana "a cute cat sitting on a windowsill"
nanobanana "mountain landscape at sunset" -a 16:9 -r 2K -o landscape.png
nanobanana "cyberpunk cityscape" -r 4K -p -o cityscape.png

# Edit
nanobanana "add sunglasses and a hat" -e photo.png -o edited.png
nanobanana "change background to beach" -e input.png -o beach.png

# Reference images
nanobanana "a cute cat in this style" -ref style.png -o cat.png

# Search grounding (real-time info)
nanobanana "visualize today's weather in Tokyo" -s

# Complex: edit with grounding, reference, and style
nanobanana "turn this into a retro travel poster" -e input.png -ref style.png -s -a 16:9 -t high -o poster.png

Tips

Be specific: include style keywords, lighting, mood, composition
For edits, be clear about what to change vs. preserve
For -s search mode, start prompt with "visualize"
For multiple images, spawn parallel threads to generate/edit concurrently

NanoBanana Image Generation

CLI tool for generating and editing images using Google's Gemini 3.1 Flash and Gemini 3 Pro models.

When to use this skill

User requests image generation or editing

User wants to create visual content or visualize real-time info

User mentions "nanobanana" or image generation

User specifies image parameters like aspect ratio or resolution

Command

nanobanana "prompt" [options]

Options

Flag

Description

Default

-o

Output file path

nanobanana_TIMESTAMP.png

-a

Aspect ratio (1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, 1:4, 4:1, 1:8, 8:1)

1:1 (gen) / keep (edit)

-r

Resolution (0.5K, 1K, 2K, 4K)

1K

-e

Edit existing image (input path)

-ref

Reference image(s) for style guidance (max 14)

-s

Google Search grounding for real-time info

off

-i

Google Image Search grounding (Flash only)

off

-p

Use Gemini 3 Pro model

off

-t

Thinking level: minimal, high (Flash only)

Examples

# Generate nanobanana "a cute cat sitting on a windowsill" nanobanana "mountain landscape at sunset" -a 16:9 -r 2K -o landscape.png nanobanana "cyberpunk cityscape" -r 4K -p -o cityscape.png # Edit nanobanana "add sunglasses and a hat" -e photo.png -o edited.png nanobanana "change background to beach" -e input.png -o beach.png # Reference images nanobanana "a cute cat in this style" -ref style.png -o cat.png # Search grounding (real-time info) nanobanana "visualize today's weather in Tokyo" -s # Complex: edit with grounding, reference, and style nanobanana "turn this into a retro travel poster" -e input.png -ref style.png -s -a 16:9 -t high -o poster.png

Tips

Be specific: include style keywords, lighting, mood, composition

For edits, be clear about what to change vs. preserve

For -s search mode, start prompt with "visualize"

For multiple images, spawn parallel threads to generate/edit concurrently