DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Vision Bridge — DSH Plugin for DeepSeek Harness
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in
← Plugins

@acc1143/dsh-vision-bridge

Vision Bridge

DSH dual-model routing plugin: keeps the main conversation in DeepSeek, automatically routes image tasks to an explicitly configured vision model, and returns the analysis results to DeepSeek to continue.

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add @acc1143/dsh-vision-bridge@0.1.1
READMECompatibilityVersions

Compatibility and provenance

Vision Bridge is published as @acc1143/dsh-vision-bridge and currently resolves to version 0.1.1. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
any
Release source
npm
Registry updated
9/20/2026

Versions

0.1.2stable
8/22/2026
0.1.1stable
8/15/2026

Related plugins

Loading related plugins…

Latest
0.1.1
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
20.9 kB
Files
5
Surface
any
License
MIT
Source
npm
GitHub
★ 1
Weekly downloads
27
Last push
8/16/2026
View source ↗Project homepage ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue

Related plugins

More verified plugins in models-usage.

Usage@linxin666/dsh-usageUsage statistics plugin for the dsh web GUI: per-provider balance and coding-plan quota detection plus a live token usage ledger, with a dedicated pet bubble for the current providerWhale Widgetdsh-whale-widgetDeepSeek balance whale widget in the bottom-right corner of the DSH Web interface: balance/today’s usage/peak-off-peak pricing, customizable bubble click sequence (text/balance/today/peak-off-peak/image/random phrases and parallel weighted selection), per-line styles and fonts, floating quick editinUsage Stats@ychris12138/dsh-usage-statsToken usage heatmap, provider balances, and subscription quotas for the dsh web GUICodex Connectdsh-codex-connectChatGPT OAuth and Codex models for DeepSeek Harness.

README

🚀 dsh-vision-bridge —— 让 DeepSeek 学会「看图」的双模型路由插件

重磅! DeepSeek Harness 专属视觉外挂来了!主对话永远留在 DeepSeek,图片自动交给视觉模型分析,整轮推理不换人!

DSH Plugin · LlmAdapter · 双模型路由 · OpenAI 兼容视觉网关 · 图片分析缓存

✨ 一句话说清它是干嘛的

DeepSeek 是纯文本模型——你以前传图,DSH 直接拒绝(Model does not support image input)。

装了它,你的 DSH 多了一个 auto 模型:图片进来,自动交给视觉模型翻译成文字,再交回 DeepSeek 完成整轮思考。你全程无感,就像 DeepSeek 突然长了眼睛 👀

以前现在(auto)
传图片❌ 直接被拒✅ 自动分发
谁看图—你配的视觉模型(有界分析,≤1024 token)
谁做整轮推理DeepSeek(但看不见图)还是 DeepSeek,图片只是它的"翻译文本"
纯文字对话正常正常,零额外开销
要手动切模型吗不用(但也没图可看)不用,auto 全自动

🔥 三大卖点,每一个都值得

  1. 主对话永远留在 DeepSeek —— 推理、工具调用、作答全在 DeepSeek,不换模型、不切来切去、上下文贯穿全程
  2. 纯文本轮次零开销 —— 没图就不调视觉模型,一分钱不多花
  3. 图片分析按内容缓存 —— 同一张图只分析一次,历史里重复出现不重复烧钱,DeepSeek 每轮看到的分析文本逐字一致

附赠:文本大脑也能换——delegateModel 改成 deepseek-v4-pro 或你任何已注册渠道,一行配置,热生效。


🛠️ 快速上手(3 步,5 分钟)

# 1️⃣ 安装(从 npm 一键安装,自动追加到 profile bundles)
dsh plugin --profile web add @acc1143/dsh-vision-bridge

# 2️⃣ 重启 dsh web

没有 npm 也没关系,同样一条命令:dsh plugin --profile web add github:Acc1143/dsh-vision-bridge#v0.1.1

# 3️⃣ 在 settings.yaml 配好你的视觉模型
vision-bridge:
  visionProvider: openai      # 你在 设置→Models 配的视觉渠道(任意 OpenAI 兼容视觉网关都行)
  visionModel: gpt-4o         # 该渠道下的视觉模型 id

保存即生效(热加载),新会话默认就是 auto。传一张图试试 → 成了 🎉

完整配置项、换文本大脑、已知限制与故障排查见 README-fake.md(技术文档)。


🧠 原理一张图

用户传图 ──► auto-hybrid(本插件路由,声明支持图片)
              │ 无图:直接委托 DeepSeek(零开销)
              ├─► deepseek-official(文本轮次)
              │ 有图:先视觉分析(有界子任务)
              ├─► 视觉模型 ──► 分析文本替换图片块
              └─► deepseek-official(推理 / 工具 / 作答)

密钥解析交给视觉渠道自己的适配器(llm-pi-ai),本插件不接触任何凭据。


🆚 和同类项目有什么不一样

生态里已有视觉插件,但思路不同,三者可以互补:

项目思路差异点
liustack/modlens(1.4k⭐)图片 → 结构化 JSON 证据(OCR/版面/语义)把视觉做成"证据工具",输出数据
Anionex/dsh-vision-toolkit(362⭐)视觉工具集(长截图 OCR/UI 还原/pixel diff)工具面更宽,偏 agent 集成
dsh-vision-bridge(就是本仓库)双模型路由主对话永不换模型,图片只是"过一道视觉翻译",整轮推理留在 DeepSeek

🔍 检索关键词

DeepSeek Harness DSH 视觉 多模态 图片识别 双模型 自动路由 LlmAdapter vision multimodal OCR OpenAI 兼容 图片分析 auto-hybrid plugin


📜 许可证

MIT —— 详见 LICENSE。