dsh-eyes: Attaching external "eyes" to DeepSeek Harness pure text chat
# Preface The philosophy behind DeepSeek Harness (hereinafter referred to as DSH) is "everything is a plugin," but its web chat interface is text-based: pasting a screenshot of an error or dragging in a design draft means the model cannot see them; you can only describe them in text or use other tools to perform OCR before copying the results back into the conversation. For agent developers, such operations are frequent and interrupt the workflow each time. qing9835's open-source plugin, dsh-eyes, addresses this problem: images pasted, dragged in, or imported are intercepted by the plugin, passed to OpenAI-compatible vision models for recognition as text, and automatically sent into the conversation. # What is dsh-eyes (package name dsh-vision-bridge) is a DSH web...
Read Moredsh-eyes: Enable text-only models in DeepSeek Harness to process images on the fly
# Introduction When using pure text models like DeepSeek as the main model in DeepSeek Harness (DSH), there is an unavoidable limitation: Harness does not allow sending messages with images to sessions where the current model does not support images. If you want the model to view an error screenshot, an interface layout, or a photo of a table, you must either convert the content in the image to text and send it first, or switch to a multimodal main model. dsh-eyes provides an alternative approach: images can be pasted and sent as usual, and are persisted in the background; the main model invokes the `view_image` tool when it needs to view the image, and the recognition part is handled by any OpenAI-compatible vision interface (defaulting to Bailian Qwen). In terms of usage experience, the effect is close to the model's native capabilities...
Read More