为非多模态模型提供图像描述能力，并可查询可用视觉模型。
MCP server that provides image description capability using StepFun Step-3.7-flash multimodal model, enabling models to 'see' images by converting them into detailed text descriptions.
