sjakdhasdh/dsh-vision 预览 preview

sjakdhasdh/dsh-vision

DeepSeek Harness(DSH)的视觉工具插件:为 deepseek-v4-flash 等纯文本模型提供图像识别能力,通过阿里云百炼或任何兼容 OpenAI 的视觉 API 实现。给 DeepSeek Harness 无识图能力模型加识图工具。

Project Overview项目介绍

dsh-vision is a native plugin built exclusively for DeepSeek Harness that adds image recognition capabilities to text-only LLMs that do not natively support vision, such as deepseek-v4-flash. It accepts both absolute paths to local images and URLs of remote images hosted online, then passes the image to an OpenAI-compatible vision LLM to generate a text description of the image content. By default, it uses the qwen3.7-flash model hosted on Alibaba Cloud Bailian, but it works with any OpenAI-compatible vision LLM from any provider.

To install the plugin, you first clone the repository to your local machine, run pnpm install to pull build dependencies, then run pnpm run build to compile the plugin code. After building, you run the DSH CLI command dsh plugin --profile web add ./dsh-vision from the parent directory of the repository to add the plugin to your DSH profile. Once the installation completes, you restart DSH and open a new session, and the vision tool will automatically be added to your model’s toolset. This plugin is ideal for DSH users who work with text-only DeepSeek models and need added vision capabilities.

The plugin supports two configuration methods: setting environment variables, or adding a config entry to DSH’s profile patch file. Configuration follows a clear precedence order: plugin-level config overrides environment variables, which override the plugin’s default values. The plugin has zero extra runtime dependencies, as it only uses Node.js’s built-in fetch API to make API requests. It is released under the permissive MIT open source license, and requires a small patch to the default dsh-llm-deepseek adapter to allow image uploads, which is documented in the plugin’s PATCHES.md file.

dsh-vision 是专为 DeepSeek Harness 开发的原生视觉识别插件,用于给 DSH 中不自带原生识图能力的大模型(比如 deepseek-v4-flash)添加图片识别能力。它支持传入本地图片绝对路径和网络图片 URL,将图片转交给兼容 OpenAI 格式的视觉大模型处理后,返回中文图片内容描述。默认使用阿里云百炼的 qwen3.7-flash 模型,也可切换到其他兼容服务商。

该插件需要通过 pnpm 安装依赖后编译构建,再通过 DSH 的 CLI 命令将插件添加到指定配置文件,重启 DSH 后即可在新会话中使用。模型会在需要识别图片时自动调用该工具,用户也可以额外指定识别要求,如果不指定则默认请求详细描述图片内容。适合使用 DSH 搭配纯文本模型、需要给模型添加识图能力的开发者使用。

配置支持环境变量和 DSH 配置补丁两种方式,配置优先级为插件配置 > 环境变量 > 默认值。该插件没有额外的运行时依赖,仅使用 Node.js 内置的 fetch 模块发起请求,遵循 MIT 许可开源。需要注意的是,原生 DSH 的 DeepSeek 适配器默认拦截图片上传,用户需要按照说明打一个小型补丁才能启用粘贴上传功能。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意
  • Only 2 stars - very few users, little community feedback星标只有 2,几乎没人在用,遇到问题缺少社区反馈
DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

  • DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
  • External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时(Electron / Python / Docker 等)是否齐备
  • Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突:命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

  • Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致,是否归档或长期停更
  • Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
  • curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
  • Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
  • Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需(filesystem / network / shell / clipboard)
  • Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限,以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照,不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装(适合开发者)

CLI Install命令行安装

dsh plugin --profile web add github:sjakdhasdh/dsh-vision

把 sjakdhasdh/dsh-vision 加入你的 DSH 配置(web profile)即可启用。

READMEREADME

dsh-vision 👁️

CI License: MIT

给 DeepSeek Harness 里没有原生识图能力的模型(如 deepseek-v4-flash)加上识图工具。 Give image-recognition ability to DeepSeek Harness models without native vision (e.g. deepseek-v4-flash).

把本地图片或网络图片 URL 交给视觉大模型(默认阿里云百炼 qwen3.7-flash),返回中文文字描述。 Delegates local image paths / URLs to a vision LLM (default: Alibaba Cloud Bailian qwen3.7-flash) and returns a Chinese description.

English | 中文

demo


中文

特性

  • 🖼️ 支持本地图片路径、网络图片 URL
  • 🔑 OpenAI 兼容格式,不绑定特定厂商(默认阿里云百炼)
  • ⚙️ 配置优先级:插件 config > 环境变量 > 默认值
  • 📦 零额外运行时依赖(只用 Node 内置 fetch)

安装

pnpm install && pnpm run build
# 在插件父目录执行:
dsh plugin --profile web add ./dsh-vision
# 重启 dsh,然后新建会话即可使用 vision 工具

配置

方式一:环境变量

export DASHSCOPE_API_KEY=sk-xxx
export VISION_MODEL=qwen3.7-flash-2026-07-15
export DASHSCOPE_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1

方式二:profile 补丁层 ~/.dsh/profiles/<name>/cordis.patch.yml

- id: dsh-vision
  config:
    apiKey: sk-xxx
    model: qwen3.7-flash-2026-07-15
    baseURL: https://dashscope.aliyuncs.com/compatible-mode/v1

使用

模型会自动调用 vision 工具,参数:

参数 必填 说明
image ✅ 本地图片绝对路径(如 C:\a.png)或 http(s) URL
prompt ❌ 识别要求,默认"请详细描述这张图片的内容"

提示:配合图片上传

DeepSeek Harness 默认的 DeepSeek adapter 声明模型纯文本,上传图片会被 MODEL_DOES_NOT_SUPPORT_IMAGES 拦截。 要让用户能直接粘贴图片(图片块渲染为 [图片附件: sha256:...] 标记),需要对 dsh-llm-deepseek 打一个小补丁(见 PATCHES.md)。


English

Features

  • 🖼️ Local image paths and remote http(s) URLs
  • 🔑 OpenAI-compatible API — vendor-agnostic (Bailian by default)
  • ⚙️ Config precedence: plugin config > environment > defaults
  • 📦 Zero extra runtime deps (built-in fetch only)

Install

pnpm install && pnpm run build
# from the PARENT directory:
dsh plugin --profile web add ./dsh-vision
# restart dsh, then open a NEW session — the `vision` tool appears in the model's toolset

Configuration

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要,完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

← 上一个 Prev dsh-completion-notifier 下一个 Next dsh-1bot →