"author:mateuszpanagiotou0628" matching MCP servers:

Grants image-understanding capabilities to AI coding assistants that lack native vision by routing screenshots, error messages, UI mockups and documents through a single unified analysis tool. Supports multiple vision model providers such as GLM, DeepSeek-OCR, Qwen-VL, Doubao and Hunyuan, with automatic multi-crop, compression and retry handling for text-dense images.