dsh-vision-toolkit
Harness-native vision toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts.
Install
Forwards to pnpm inside the web profile. Confirm the README if the author documents a .dsh-plugin path.
dsh plugin --profile web add github:Anionex/dsh-vision-toolkit
What it is
A DeepSeek Harness-native integration of agent-vision-toolkit. Compared with generic vision skills, this one is wired for the Harness Web UI: image Q&A, long-screenshot OCR, frontend UI restoration, grounding, pixel diff, Artifacts, and an in-product viewer. Use it when you want the vision loop to live inside dsh, not in a sidecar process.
Source: Anionex/dsh-vision-toolkit. Official harness: deepseek-ai/deepseek-harness.
Listed from public GitHub topic dsh-plugin, snapshot 2026-08-13.
Related in Vision
First vision plugin for DeepSeek Harness. Paste an image, get structured JSON: OCR, layout, semantics.
Vision toolkit for text-only LLMs. Works with Codex, Claude Code, Pi, OpenCode — and tags dsh-plugin.