A collection of specialized skill files designed to enhance AI coding assistants with domain-specific knowledge and capabilities.
Skills are structured instruction sets that give AI agents (like Claude Code and OpenAI Codex) specialized knowledge for specific domains. They address gaps in LLM training data by providing:
- Current API references and usage patterns
- Working code examples and best practices
- Domain-specific workflows and techniques
- Quality guidelines and error handling strategies
| Skill | Description | Install |
|---|---|---|
| nano-banana-pro | Image generation and editing using Google Gemini's Nano Banana Pro model. Supports text-to-image, image editing, multi-turn conversations, and transparency extraction via difference matting. | npx skills add EnzeD/skills --skill nano-banana-pro |
Install a specific skill:
npx skills add EnzeD/skills --skill <skill-name>Copy the skill folder to your project's skill directory:
Claude Code:
cp -r nano-banana-pro .claude/skills/OpenAI Codex:
cp -r nano-banana-pro .codex/skills/Each skill follows a consistent structure:
skill-name/
├── SKILL.md # Main skill instructions (required)
├── references/ # Additional documentation
│ └── *.md
└── scripts/ # Helper scripts and utilities
└── *.py
Contributions are welcome! When editing a skill:
- Update the
SKILL.mdwith improved instructions or new capabilities - Add or modify supporting files in
references/andscripts/as needed - Test your changes to ensure the skill works correctly
MIT License - feel free to use these skills in your projects.