nano-banana_skill

This skill generates images with Google's Nano-Banana model, enhancing prompts for higher quality visuals and faster creative results.

Shell

2

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill dennisonbertram/claude-media-skills --skill nano-banana

PROMPTING.md5.3 KB
SKILL.md2.9 KB

Overview

This skill generates images using Google's Nano-Banana (Gemini 2.5 Flash Image) model. It converts natural language prompts into images, with options for resolution, aspect ratio, and output path. The skill runs in the background, not blocking other work, and auto-opens the final image with a macOS notification when complete.

How this skill works

Enhance the user's prompt first using a concise formula (Subject + Style + Lighting + Composition + Quality) to improve results. The skill maps user keywords to script flags (size, aspect, output, pro/flash) and calls a command-line generator to produce the image. Images are saved by default to ~/Pictures/nano-banana/ and the tool reports completion and errors via logs and notifications.

When to use it

When you ask to generate, create, or make an image, picture, or illustration
When you need control over resolution (1K, 2K, 4K) or aspect ratio (16:9, 9:16, 4:3, etc.)
When you want quick, background image generation with auto-open and macOS notification
When you need to save output to a specific file or folder
When you want a faster Flash model for quick drafts or a Pro model for higher quality

Best practices

Always enhance prompts before generating using: Subject + Style + Lighting + Composition + Quality
Specify resolution keywords (4K, 2K) and orientation words (landscape, portrait) to control flags
Use explicit save paths for production assets (e.g., --output ~/Desktop/name.png)
Use Flash for fast previews; use Pro for high-resolution, detailed images
Check the error log and API key when generation fails

Example use cases

Generate a photorealistic 4K landscape wallpaper: include '4K' and 'landscape' in the prompt
Create a vertical phone wallpaper of a character: specify 'portrait' or '9:16' and stylistic details
Produce a high-quality product mockup: add 'high quality, detailed, studio lighting' and use Pro
Quickly iterate concept art with Flash mode for faster drafts
Save a commissioned illustration directly to a project folder using --output

FAQ

Images are saved to ~/Pictures/nano-banana/ by default; use --output to change the path.

How do I request 4K or portrait orientation?

Include keywords like '4K' or 'portrait' (or '9:16') in your prompt; the skill maps them to --size and --aspect flags.