JochenYang/luma-mcp

View on GitHub ↗

多模型视觉理解 MCP 服务器，为不支持图片理解的 AI 编码模型提供视觉能力：分析截图、报错、UI 与文档，可接入多家主流视觉大模型。Multi-model vision MCP server that adds image understanding to AI coding models without native vision — analyze screenshots, errors, UI and documents via major vision LLM providers.

109 ★12 forksTypeScriptUpdated 26d ago

What you need to know

Multimodel visual-understanding MCP server with a unified image_understand tool (GLM-4.6V/DeepSeek-OCR/Qwen3-VL-Flash etc.)

Install

npx -y luma-mcp

Usage

  • •Point the image_understand tool at an image with a question

Key features

  • ✓unified image_understand tool
  • ✓multiple vision backends

Best for

Vision tasks in agents across several models

Caveats

  • ⚠Chinese docs; verify backend availability
Platforms: Node.jsClients: any MCP client

Reviewed 2026-08-11

Topics

image-understandingmcpmcp-servermodel-context-protocolvision
Stars
109★
Forks
12
Language
TypeScript
License
MIT
Created
2025-11-11
Last push
2026-08-09