dsh-tool-vision
Local-first structured vision for text-only agents: images go to a local OpenAI-compatible VLM and come back as JSON evidence (summary, verbatim OCR, layout regions, entities/relations, colors, explicit uncertainty), with anti-hallucination fallback and an optional paste/upload bridge; zero cloud cost, images never leave the machine.
本地优先的结构化视觉插件：图片交给本地 OpenAI 兼容 VLM，返回 JSON 证据（摘要、逐字 OCR、版面区域、实体/关系、配色、显式不确定项），带反幻觉回退与可选粘贴/上传桥；零云端成本，图片不出本机。
How to install
dsh plugin add dsh-vision-local About
dsh-tool-vision **为纯文本的 DeepSeek Harness agent 提供本地优先的视觉能力。** npm 包：**dsh-vision-local** · 源码仓库：gloryxpnv/dsh-tool-vision · English README 让纯文本模型（DeepSeek、GLM 等不支持图片输入的模型）拥有"视觉"——使用**本地**视觉模型，**零 API 成本**，**图片数据完全不出本机**。 DeepSeek (纯文本) ──▶ vision 工具 ──▶ 本地 VLM（LM Studio / Ollama / 任意 OpenAI 兼容端点） ◀── 结构化 JSON 证据 ──▶ --- 亮点 🏠 **完全本地、完全私密。** 图片只发送给你自己的本地视觉模型（LM Studio、Ollama、vLLM 或任意 OpenAI 兼容端点）。无需云端 key、无单图成本、图片字节永不离开你的机器。 📋 **结构化证据，而非模糊转述。** VLM 被要求填写固定 JSON 模板——summary（摘要）、ocr（逐字全文+行）、layout（版面区域，含阅读顺序）、semantics（实体与关系）、visual（配色/风格），以及显式的 uncertainty（不确定项）列表。主模型引用具体证据作答，而不是凭空猜测。 🛡️ **反幻觉…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 1
- mo. downloads
- 230
- Last push
- 2026-08-15
- Created
- 2026-08-15
Links
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh-plugin, local-first, vision
Related plugins
modlens
liustack/modlens
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
dsh-web
zhu1090093659/dsh-web-ui/tree/main/packages/dsh-tool-describe-image
A `describe_image` vision tool for text-only models: images (local path, URL, attachment) go to a configurable OpenAI-compatible vision endpoint and only the returned text enters the session.
dsh-vision-router
ysr666/dsh-vision-router
Free vision for text-only agents: built-in keyless vision chain plus pixel tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots); paste an image to use it.
dsh-vision-toolkit
Anionex/dsh-vision-toolkit
Vision for text-only models: paste an image and the model switches to a Vision Toolkit variant for image Q&A, multi-image comparison, long-screenshot OCR, screenshot-to-UI reproduction, element grounding, and pixel diff. No API key by default — images are processed by the author-hosted free service, 100 per machine per day; configurable to your own provider.