skills.rest
portalshq
portalshqOFFICIAL
Explore all skills by this author →
GitHub

nano-banana-pro

Generate and edit images with the Gemini 3 Pro Image API at 1K, 2K, or 4K resolution.

5|Updated Dec 8, 2025
Search Tags:#png#api#image generation#image editing#gemini#text-to-image#prompt
One-click install
npx skills add https://github.com/portalshq/portals-studio --skill nano-banana-pro-portalshq
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nano-banana-pro
Source: https://github.com/portalshq/portals-studio/tree/main/.cursor/skills/nano-banana-pro
Command: npx skills add https://github.com/portalshq/portals-studio --skill nano-banana-pro-portalshq

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Creating or modifying images requires switching to separate design tools or web UIs. This Skill lets you generate new images from text prompts and edit existing images directly from your working directory using Google's Nano Banana Pro (Gemini 3 Pro Image) API. ## Core Features & Use Cases - Text-to-Image Generation: Create images from natural language prompts at 1K, 2K, or 4K resolution with a single command. - Image-to-Image Editing: Modify existing images via the --input-image flag with instructions like changing style, removing objects, or adjusting colors. - Draft-to-Final Workflow: Iterate quickly at 1K resolution, then render the locked prompt at 4K to control cost and time. - Use Case: A content creator drafts a blog header at 1K, refines the prompt over three iterations, then renders the final 4K version saved with a timestamped filename in the project folder. ## Quick Start Ask the AI to generate an image of your subject with the nano-banana-pro skill, optionally specifying a resolution like 4K or an input image to edit.

Frequently Asked Questions about nano-banana-pro

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an image with Gemini 3 Pro Image from the command line?▼

Run the generate_image.py script with uv, passing --prompt with your description, --filename for the output PNG, and optionally --resolution (1K, 2K, or 4K). The script requires a GEMINI_API_KEY environment variable or an --api-key argument.

How do I edit an existing image with AI instructions?▼

Pass the source image path via --input-image and put editing instructions in --prompt, such as "make the sky more dramatic" or "remove the person". The script sends both to the Gemini 3 Pro Image API and saves the edited result as a new PNG.

What resolutions does the Gemini 3 Pro Image API support?▼

The API supports three resolutions: 1K (~1024px, the default), 2K (~2048px), and 4K (~4096px). The recommended workflow drafts at 1K for fast iteration and renders the final locked prompt at 4K.

Why does image generation fail with a no API key error?▼

The script exits with "Error: No API key provided" when neither the --api-key argument nor the GEMINI_API_KEY environment variable is set. Set the environment variable or pass the key directly to resolve it.

What are the limitations of AI image editing with prompts?▼

Edits are governed by the prompt, so multi-change requests can drift from the original. The skill recommends changing one thing at a time and explicitly instructing the model to keep subject, composition, lighting, and text unchanged.

Related Skills

Explore More Design & Creative Skills→
AgriciDanielAgriciDaniel
7.9k

ads-generate

Generate ad creative images from campaign briefs and brand profiles.

Community
Advanced
indiearkindieark
1

video-gen-storyboard

Produces high-resolution grid, script-composed, and four-column layouts for video storyboards.

Official
Advanced
CleanExpoCleanExpo

Imagen 3

Generates, edits, and upscales images using Google's Imagen 3 API.

Community
Advanced
taracodlabstaracodlabs
799

stable-diffusion-image-generation

Generate images from text prompts using Stable Diffusion via HuggingFace Diffusers or Inference API.

Official
Intermediate