$npx skillfedfor your agent

nano-banana-pro

Create images from text descriptions or edit existing images with AI-powered generation. Supports resolutions from 512px to 4K, with a recommended workflow that starts at 1K for quick iteration before finalizing at higher resolutions. Requires a Gemini API key and the uv package manager.

Nano Banana Pro generates new images or edits existing ones from text prompts using Google's Gemini API.

AI-generated summary based on this skill's SKILL.md

★ 6,476  537 MITupdated by steipete

Decision gist · record as of 2026-07-23

Nano Banana Pro generates new images or edits existing ones from text prompts using Google's Gemini API. Create images from text descriptions or edit existing images with AI-powered generation. Supports resolutions from 512px to 4K, with a recommended workflow that starts at 1K for quick iteration before finalizing at higher resolutions. Requires a Gemini API key and the uv package manager.

manual: git clone https://github.com/steipete/agent-scripts → cp -r agent-scripts/skills/nano-banana-pro ~/.claude/skills/nano-banana-pro
skills/nano-banana-pro/SKILL.md · version d73fbb99

Use it when

  • Yes, nano-banana-pro can edit and modify existing images using natural language instructions.
  • nano-banana-pro creates images at multiple resolutions ranging from 512px up to 4K.

Verify before relying

Read SKILL.md below before installing (2 files). Open directory: indexed for reading, not audited.

Same gist for agents: .md · .json

Install

steipete/agent-scripts/nano-banana-pro · repository language: Shell

Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.

Frequently asked questions

AI-generated answers based on this skill's SKILL.md and metadata

How do I generate an image from a text prompt using nano-banana-pro?

nano-banana-pro generates images from text descriptions by leveraging AI-powered image synthesis. You provide a text prompt describing what you want to create, and the tool produces an image matching your description. The skill supports multiple resolution outputs from 512px up to 4K, allowing you to start with lower resolutions for quick iteration before finalizing at higher quality.

Can nano-banana-pro edit existing images with AI instructions?

Yes, nano-banana-pro can edit and modify existing images using natural language instructions. Beyond generating new images from scratch, the tool lets you take an existing image and apply AI-powered edits by describing the changes you want in plain text, enabling seamless image refinement workflows.

What resolutions does nano-banana-pro support for image creation?

nano-banana-pro creates images at multiple resolutions ranging from 512px up to 4K. The recommended workflow starts at 1K resolution for quick iteration and draft exploration, then finalizes at higher resolutions once you're satisfied with your composition and content.

How can I iterate quickly on image prompts with nano-banana-pro?

nano-banana-pro supports a draft-to-final workflow designed for rapid iteration. Start by generating drafts at lower resolutions like 512px or 1K to explore ideas quickly and cheaply, then refine your prompts based on results before creating final versions at higher resolutions like 4K.

Does nano-banana-pro offer command-line API automation?

Yes, nano-banana-pro can be automated via command-line API, enabling batch image generation and editing workflows. This allows you to integrate image creation and modification into scripts and automated pipelines for efficient processing of multiple images.

What are the requirements to use nano-banana-pro?

nano-banana-pro requires a Gemini API key for accessing the AI image generation and editing capabilities. You'll also need the uv package manager installed. The tool is released under the MIT license, making it free to use and modify.

SKILL.md

Rendered from the published skill. Quoted content, verbatim.

Nano Banana 2 Image Generation & Editing

Generate new images or edit existing ones using Google's Nano Banana 2 API (Gemini 3.1 Flash Image).

Usage

Run the script using absolute path (do NOT cd to skill directory first):

Generate new image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "your image description" --filename "output-name.png" [--resolution 512|1K|2K|4K] [--api-key KEY]

Edit existing image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "editing instructions" --filename "output-name.png" --input-image "path/to/input.png" [--resolution 512|1K|2K|4K] [--api-key KEY]

Important: Always run from the user's current working directory so images are saved where the user is working, not in the skill directory.

Default Workflow (draft → iterate → final)

Goal: fast iteration without burning time on 4K until the prompt is correct.

  • Draft (1K): quick feedback loop
  • `uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "<draft prompt>"

(truncated - see the full file via the links below)

File tree — 2 files
skills/nano-banana-pro/SKILL.md
skills/nano-banana-pro/scripts/generate_image.py

Let your AI agent find skills like this

Example. Real query, live index.

You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.

wish › “Generate new images from text descriptions using AI”

Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →

Related skills

Nano Banana Pro
by intellectronica · intellectronica/agent-skills

Nano Banana Pro taps Google's Gemini 3 Pro Image API to create new images from text descriptions or modify existing ones. Pick your resolution—1K, 2K, or 4K—and let the skill handle generation or editing tasks like style changes, element removal, or color adjustments.

CC0-1.0updated Apr 2026
★ 279repo stars
nano-banana-pro
by TheSyart · TheSyart/emperor-agent

Nano Banana Pro lets you create new images from text prompts or edit existing ones using Google's Gemini 3 Pro Image API. Choose from 1K, 2K, or 4K resolution depending on your needs, and iterate quickly with a draft-to-final workflow before committing to high-resolution renders.

MITupdated Jul 2026
★ 160repo stars
Inno Figure Gen
by OpenLAIR · OpenLAIR/dr-claw

Create or modify images by specifying a prompt and choosing your resolution tier. The skill defaults to OpenAI's image model when available, seamlessly switching to Gemini if needed. Control output size, pick your provider, and iterate quickly from draft to final.

no license declared → metadata onlyupdated Jul 2026
★ 1,038repo stars
nano-banana
by kingbootoshi · kingbootoshi/nano-banana-2-skill

nano-banana is a CLI tool for generating AI images from text prompts, defaulting to Gemini 3.1 Flash with optional Pro model support. It handles multi-resolution output (512 to 4K), custom aspect ratios, reference images for style transfer, and automated green-screen transparency for assets. Perfect for UI mockups, game sprites, marketing materials, and video production.

MITupdated Apr 2026
★ 401repo stars
Gpt Image 1 5
by intellectronica · intellectronica/agent-skills

This skill harnesses OpenAI's GPT Image 1.5 model to generate new images from text descriptions or edit existing images with precise control. It supports multiple output sizes and quality levels, with optional mask-based inpainting for targeted edits.

CC0-1.0updated Apr 2026
★ 279repo stars
nano-banana-pro
by swarmclawai · swarmclawai/swarmclaw

Nano Banana Pro harnesses Gemini 3 Pro's image capabilities to generate, edit, and composite images through simple command-line scripts. Supports text-to-image generation at multiple resolutions, single-image editing, and multi-image composition with customizable aspect ratios.

MITupdated Jun 2026
★ 628repo stars

More skills nano-banana-pro (MIT) · nano-banana-pro (MIT) · Nano Banana Pro (NOASSERTION) · nano-banana-pro (MIT)

Tags
generative-aiimage-synthesisprompt-engineeringvisual-content-creationiterative-refinementresolution-scalingapi-integrationbatch-processingcreative-automationmulti-modal-generation