nano-banana-pro
Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K/2K/4K; use --input-image.
Why use this skill?
Generate, edit, and iterate on high-quality images using Gemini 3 Pro within OpenClaw. Supports 1K, 2K, 4K resolutions and image-to-image workflows.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/steipete/nano-banana-proWhat This Skill Does
The nano-banana-pro skill provides a direct, high-performance interface to Google's Gemini 3 Pro Image generation engine. It enables OpenClaw users to generate high-fidelity images from text prompts or perform sophisticated image-to-image transformations. Designed for professional-grade workflows, it supports multiple resolution tiers (1K, 2K, 4K) and allows for iterative refinement, making it an essential tool for developers, designers, and creative professionals who need to maintain control over visual assets directly from their CLI or AI agent environment.
Installation
To add this skill to your OpenClaw environment, execute the following command in your terminal:
clawhub install openclaw/skills/skills/steipete/nano-banana-pro
Ensure that you have uv installed in your path, as it is used to manage the execution environment for this skill. After installation, verify your API access by setting the GEMINI_API_KEY environment variable or keeping your key ready to pass as an argument during execution.
Use Cases
- Prototyping & Ideation: Generate rapid 1K drafts to validate prompt concepts and composition.
- Content Transformation: Use image-to-image capabilities to restyle existing assets, apply artistic filters, or modify specific elements in an image.
- Production Asset Creation: Finalize visual assets at 4K resolution once the creative direction is finalized through iteration.
- Automation: Incorporate image generation into larger development workflows or build pipelines where visual assets need to be programmatically updated or generated based on code changes.
Example Prompts
- "Generate a 1K draft of a futuristic cityscape at sunset, neon lights reflecting in the rain, save it as cityscape-draft.png."
- "Take my sketch at sketch.png and turn it into a high-res 4K professional architectural rendering, save as final-design.png."
- "Create a 2K resolution image of a minimalist workspace with a laptop and coffee mug, focus on clean lines and natural lighting."
Tips & Limitations
- Iterative Workflow: Always start with 1K resolution. This reduces costs, saves time, and provides faster feedback loops before you commit to a 4K render.
- Naming Conventions: Use descriptive names with the timestamp pattern (e.g.,
yyyy-mm-dd-hh-mm-ss-name.png) to keep your project directory organized. - Error Handling: If you encounter a 403 error, verify that your API key has the necessary permissions for Gemini 3 Pro Image generation. Always ensure the file path passed to
--input-imageis valid and the file is readable by the current user shell.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-steipete-nano-banana-pro": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read, external-api, code-execution
Related Skills
swiftui-liquid-glass
Implement, review, or improve SwiftUI features using the iOS 26+ Liquid Glass API. Use when asked to adopt Liquid Glass in new SwiftUI UI, refactor an existing feature to Liquid Glass, or review Liquid Glass usage for correctness, performance, and design alignment.
qmd
Local search/indexing CLI (BM25 + vectors + rerank) with MCP mode.
songsee
Generate spectrograms and feature-panel visualizations from audio with the songsee CLI.
summarize
Summarize URLs or files with the summarize CLI (web, PDFs, images, audio, YouTube).
bird
X/Twitter CLI for reading, searching, and posting via cookies or Sweetistics.