Generate images on-the-fly from your coding workflow — powered by OpenAI or Google Gemini.

An agent plugin that gives your AI coding assistant the ability to generate images. Just ask — the skill handles provider selection, API key setup, and generation conversationally.

• Two providers: OpenAI (gpt-image-2) and Google Gemini (Nano Banana)
• Zero config up front — the skill onboards you the first time you use it
• Cross-tool compatible — works in VS Code, Copilot CLI, and Claude Code
• Agent plugin format — discoverable, installable, and manageable from the Extensions view

Open the Command Palette (Cmd+Shift+P) and run:

Chat: Install Plugin From Source

Enter this repo URL:

https://github.com/adamd9/skill-image-gen

The plugin and its skills will appear in your Extensions sidebar under Agent Plugins.

git clone https://github.com/adamd9/skill-image-gen.git
cd skill-image-gen
./install.sh

Just ask your AI assistant to generate an image:

• "Generate a hero image for my landing page"
• "Create a seamless grass texture for my game"
• "Make an app icon for a weather app"

The skill takes care of the rest — including walking you through API key setup if it's your first time.

skill-image-gen/
├── plugin.json              # Plugin manifest
├── skills/
│   └── generate-image/
│       └── SKILL.md         # The skill — all logic lives here
├── install.sh               # CLI installer
├── README.md
├── CONTRIBUTING.md
└── LICENSE

See CONTRIBUTING.md.

MIT — see LICENSE.