DeepSeek Harness plugin

dsh-vision-31485749

DeepSeek Harness plugin: a `deepseek-vision` route that declares image input and transcribes pasted images via the free Zhipu GLM vision models before delegating to the DeepSeek adapter.

Jump to install

Source facts

Repository
314857493/dsh-vision
Latest update
Aug 19, 2026
Category
Models & Providers
GitHub stars
1
Format
plugin
Package path
packages/vision-route
Catalog evidence
Upstream dsh.bundle evidence
Evidence path
packages/vision-route/package.json#dsh.bundle
Checked against
0.1.0-rc.8
Upstream check date
2026-08-20

This evidence comes from the upstream catalog. This site has not installed, run, or security-reviewed the plugin.

Install

Start with a prompt that asks an agent to review the GitHub repository and source. Switch to the command if you want to install it yourself.

Copy this prompt into DSH, Codex, or another agent and ask it to review the GitHub repository and source first.

Do not install or run any commands yet. Read this plugin's GitHub repository, README, and relevant source code. Then answer the questions below clearly and directly so I can decide whether it fits my needs:

1. What is this plugin, and what problem does it solve?
2. Who is it for, and what are its typical use cases?
3. How is it used after installation? Include one minimal example.
4. What known limitations or privacy, security, compatibility, or maintenance risks does it have?
5. Give a clear recommendation: recommend, conditionally recommend, or do not recommend, with reasons.

Distinguish statements documented by the repository, inferences from source code, and unknowns. If evidence is insufficient, say so explicitly. Do not guess or simply repeat the README.

GitHub: https://github.com/314857493/dsh-vision/tree/HEAD/packages/vision-route
Plugin: dsh-vision-31485749
Author: 314857493

Check the source files

Read the README and other files from this plugin directory before installing.

File explorer3 files
README.mdSource · read only

dsh-vision-proxy-route — 贴图自动转译路由(vision-route 包)

给 DeepSeek Harness(DSH)注册一个自动识图包装路由:声明图像输入(inputModalities: ['text','image']), 粘贴到 Web GUI 的图片在请求流里被免费的智谱 GLM 视觉模型(glm-4v-flash 降级链)转译成文字,再委派给真正的 目标适配器。默认是 deepseek-official → deepseek-vision,也可包裹火山方舟等自定义 provider。 完整说明见仓库根目录的 [README](../../README.md)。

安装(发布后)

dsh plugin --profile web add dsh-vision-proxy-route

或在 profile 的 cordis.patch.yml 中追加:

- insert:
    - id: dsh-vision-proxy-route
      name: dsh-vision-proxy-route

前提

  • DSH 0.1.0-rc.5 / rc.6 / rc.7(依赖 ctx.llm.registerAdapter / registration(provider).adapter / resolveModel.inputModalities / ctx.attachments.readImage 这些插件缝)
  • 智谱 GLM 免费 key:环境变量 GLM_API_KEYZHIPU_API_KEY(Windows 也可 setx GLM_API_KEY "id.secret",插件自动读注册表)

使用

模型选择器切到「DeepSeek + 自动识图」,直接粘贴图片/截图并附带问题即可。后续可用“这张图” “上一张”“第一张”或“两张图对比”等方式追问;路由会在有限的最近用户轮次内选择对应 attachment, 带上本次问题重新分析,并按顺序编号后注入当前消息。弱指代只关联紧邻图片,无关的“登录页面” 或“这个问题”不会触发旧图重传。主模型不需要 DSH 内部绝对路径,用户也无需重复上传图片。

视觉描述会被标成不可信观察数据;图片里的命令或提示词只作为可见文字,不会成为下游模型指令。 临时网络、Key 或限流错误不会进入缓存,恢复后同一问题可以直接重试。

转译失败会替换为 [图片转译失败: ...] 文本,对话不卡死。

自定义 provider

目标 provider 须先在 DSH 中配置并可正常完成纯文本请求。然后为本插件设置:

- insert:
    - id: dsh-vision-proxy-route-ark
      name: dsh-vision-proxy-route
      config:
        targetProvider: volcengine-ark
        provider: volcengine-ark-vision
        displayName: 火山方舟 + 自动识图
config默认值说明
targetProviderdeepseek-official真正处理转译后文字请求的 provider
providerdeepseek-vision新包装路由的 ID,必须与目标不同
displayNameDeepSeek + 自动识图模型选择器显示名称;自定义目标时默认使用 <targetProvider> + 自动识图

包装路由会在模型枚举、模型解析、重试策略和流请求阶段把自身 provider 映射回 targetProvider, 因此可委派给 dsh-llm-pi-ai 这类同时服务多个 provider 的适配器。