dsh-vision-plugin
Configurable image recognition for text-only DSH models: image messages are first transcribed by any OpenAI-compatible vision model (Base URL, model ID and key configured in a Settings section), then passed to the main model as text; image-input support is advertised while enabled.
为纯文本 DSH 模型增加可配置识图能力：开启后图片消息先由任意 OpenAI 兼容视觉模型转写为文字描述，再交给主模型处理；设置页可配置 Base URL、Model ID 与密钥。
How to install
dsh plugin add @lp181818/dsh-vision-plugin About
dsh-vision-plugin DSH 图片识别插件 — 为 AI 会话添加图片理解能力。 在 **识图模型配置** 页面启用识别功能，并配置视觉模型的 Base URL、Model ID 和 API Key。 安装（通过 DSH 市场） 1. 打开 DSH Web GUI → **设置 → 插件** → **市场** 2. 添加仓库源 https://github.com/bug-huntter/dsh-vision-plugin 3. 扫描并安装 手动安装 bash dsh plugin add @lp181818/dsh-vision-plugin 效果展示 插件配置页（设置 → 识图模型配置） 在设置面板左侧导航的「识图模型配置」中，您可以配置： **启用图片识别** — 主开关，开启后 AI 可以识别并理解用户上传的图片内容 **Base URL** — 视觉模型 API 的基础地址 **Model ID** — 用于视觉任务的模型标识符 **密钥来源（变量名）** — 环境变量或凭证名，留空自动复用同路由已配置的密钥 **API Key** — 实际 API 密钥（敏感信息，妥善保管） 支持的服务 插件内部只做一次标准的 **OpenAI 兼容** POST {baseUrl}/chat/completions 请求（Authorization: Bearer <k…
Recommendation signals
Meta
- License
- MIT
- Language
- TypeScript
- GitHub stars
- 1
- mo. downloads
- 605
- Last push
- 2026-09-04
- Created
- 2026-09-03
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- dsh-plugin
Related plugins
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.
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-design-qa
sunxin-ai/dsh-design-qa
Design-fidelity QA for text-only models: a `deepseek_vision` tool borrows an eye from any OpenAI-compatible vision route, so the model can judge whether an implementation matches its mock — shipped with the benchmark behind that judgement (four fixtures, 23 injected defects, raw transcripts) and the questioning discipline it depends on.