Tuzi Image Gen

This is a multi-provider image generation skill that defaults to Tuzi API's nano-banana models but also supports Google, OpenAI, DashScope, and Replicate. It handles text-to-image with reference images, aspect ratios, and quality presets, plus OpenAI gateway dialect switching for compatibility. The setup is thorough, maybe overly so. It blocks execution until you configure API keys and preferences in EXTEND.md, which could frustrate quick tests but prevents half-configured failures. Tuzi offers a solid model lineup including VIP tiers and async polling for slower models like Midjourney. If you need one skill to route between providers without rewriting prompts, this does it, though you'll spend a few minutes on first-run configuration.

Install to Claude Code

npx -y skills add tuziapi/tuzi-skills --skill tuzi-image-gen --agent claude-code