describe-image

DeepSeek Harness plugin: describe_image — give a text-only model vision through an OpenAI-compatible VLM endpoint

A DeepSeek Harness plugin that lets DeepSeek see screenshots and local images via MindSee. DeepSeek Harness 插件：通过 MindSee 把截图和本地图片转成文字，让 DeepSeek 能看图作答。