Minimax Image Understanding

This hooks up MiniMax's vision API to analyze images through a simple understand_image call. You pass it a prompt and image URL (supports HTTP links, local paths, or base64), and it handles the conversion and API work. Best for screenshots with error messages, extracting code or text from images, analyzing UI designs, or general OCR tasks. It supports JPEG, PNG, and WebP. The skill is straightforward about when not to use it, like when you already have the context or for simple icons. One nice touch is it auto-converts images to base64 regardless of input format, so you don't need to think about encoding.

Install to Claude Code

npx -y skills add imsus/pi-extension-minimax-coding-plan-mcp --skill minimax-image-understanding --agent claude-code