"author:SakuraMartin750" matching MCP servers:

Enables AI clients and coding models that lack native vision to analyze images — screenshots, error captures, UI mockups, documents and OCR — through a single unified image-understanding tool. Routes requests to multiple vision model providers (GLM, DeepSeek-OCR, Qwen, Doubao, Hunyuan or any OpenAI-compatible endpoint) with automatic cropping, compression and retry handling.