"author:tuyenallen" matching MCP servers:

Adds image understanding to AI coding clients that lack native vision, letting them analyze screenshots, error messages, UI layouts, code and documents through a single unified image_understand tool. Routes requests across multiple vision LLM providers (GLM, DeepSeek-OCR, Qwen, Doubao, Hunyuan, or any OpenAI-compatible endpoint) with automatic multi-crop, compression, retry and SSRF-safe preprocessing.