DSH插件商店

dsh-llm-vision-bridge

Einskyle

DeepSeek vision bridge for dsh: route image attachments to a vision model (Qwen3-VL via pi-ai/llama.cpp) and continue on a text-only LLM (DeepSeek)

dsh-attachments

LCYLYM

Cross-platform DSH WebUI multimedia input and workspace attachments

dsh-any-attachment

Zenjibad

dsh bundle: attach any file type in the DeepSeek Harness Web UI — text-likes inline, binaries as workspace path references, rasters via the built-in pipeline

dsh-vision-mcp

moton16

Give your text-only LLM eyes: zero-dependency MCP server that reads images via external OpenAI-compatible vision models with automatic provider fallback, plus a one-command patch for DSH image-attachment degradation.

dsh-file-fix

re-ITRT

DeepSeek Harness plugin: unified file import - byte-level upload store, file-list context injection, read_attachment/place_attachment tools, historical file bubbles