A specialized technical skill for managing the nano-img and nano-image CLI tools to drive Gemini-powered image generation and configuration directly from the terminal.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install nano-img-cli
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install nano-img-cli using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is nano-img-cli?
The nano-img-cli skill is a robust integration designed for developers who need to interface with the nanobana npm package. It enables an AI agent to drive the local nano-img Gemini image CLI for high-fidelity image generation, model selection, and automated reference-image workflows. By leveraging this Openclaw Skills component, users can seamlessly manage local assets and configuration files without leaving their development environment.
This skill goes beyond simple generation by managing the persistent state of the tool. It handles the ~/.nano-img configuration directory, ensuring that saved defaults, instruction sets, and style guidelines are strictly followed. Whether you are automating bulk image creation or fine-tuning model parameters, this skill provides the necessary technical interface to interact with Gemini models via a structured command-line lifecycle.
nano-img-cli Use Cases
- Generating high-quality images using Gemini models directly through the terminal.
- Setting and persisting default AI models for consistent generation results.
- Managing local reference images and assets stored in the ~/.nano-img/assets directory.
- Automating image format conversion and specific output sizing for web assets.
- Inspecting and repairing local CLI configurations via machine-readable JSON outputs.
How nano-img-cli Works
- The skill identifies the command surface by executing help commands to verify available flags and capabilities.
- It inspects existing saved defaults by running configuration checks to ensure context-aware generation.
- The workflow utilizes dedicated commands like model or save-dir to modify persistent settings instead of risky manual file edits.
- It integrates global styles and instructions from Markdown files located in the user's home directory for tailored outputs.
- Upon a generation request, it executes the command while respecting overrides like aspect ratio, dimensions, and file formats.
- Finally, it validates the operation using JSON flags to confirm file paths or updated configuration states.
nano-img-cli Setup
To integrate this into your environment, ensure the nanobana package is installed and accessible. You can use the globally linked commands or run them through npm within the repository:
# Install the package globally
npm install -g nanobana
# Or use via npm scripts in the local repo
npm run nano-img -- help
# Initialize or check configuration
nano-img config --json
nano-img-cli Data Schema & Taxonomy
The nano-img-cli skill manages a specific directory structure and metadata taxonomy to ensure persistence across sessions:
| File/Folder | Purpose |
|---|---|
~/.nano-img/config.json |
Stores primary defaults like the selected model and output directory. |
~/.nano-img/assets/ |
The default repository for reference images used in generation. |
~/.nano-img/STYLE.md |
Global Markdown file defining reusable visual styles. |
~/.nano-img/INSTRUCTION.md |
Global prompts and behavioral constraints for the Gemini model. |
All generated outputs default to PNG format but can be converted to JPG, JPEG, or WebP during the local workflow.
nano-img-cli Advanced Features
- Multi-model support with interactive picking via the models command.
- Smart aspect ratio preservation during local resizing when only one dimension is specified.
- Full JSON output mode for all commands, enabling multi-agent orchestration and automated tool chaining.
- Support for custom reference image injection via the refs command and local asset mapping.
- Automated configuration clearing and environment resetting for clean-slate development tasks.
SKILL.md
Loading
Related Openclaw Skills
AI Skill Maintainer
A professional governance and security workflow for maintaining, patching, and deprecating Openclaw Skills using CTO and CISO standards.
Stakeholder Negotiation Planner
A specialized tool for architects to prepare persuasive, evidence-based negotiation strategies for business and technical stakeholders.
Didier.ai Research Workspace
Didier.ai is a collaborative research hub where autonomous AI agents publish findings, build citation graphs, and participate in structured peer review.
CTO x CISO Training & Compliance Skill
An automated training and assessment framework designed for AI agents to deliver technical standards and security compliance education.
Exec Tool
A security-focused command execution bridge for managing Openclaw Skills and system operations via controlled CLI triggers.
Karpathy Wiki
A framework for AI agents to build and maintain persistent markdown knowledge bases that bridge the gap between raw data and actionable intelligence.
Featured*
