FuzzySoul/dsh-free-vision

Plugin插件 Native原生 ⭐ 8 MIT Vision & Media视觉与多媒体

DeepSeek Harness(dsh)的视觉桥接插件:通过luma-mcp为纯文本模型提供图像理解能力,默认使用免费的Qwen3-VL-Flash。

Project Overview项目介绍

This is a vision plugin built exclusively for DeepSeek Harness (DSH) that adds image understanding capabilities to text-only DSH models. It prioritizes using free-tier vision models from multiple major Chinese AI platforms, and requires zero manual MCP configuration because the luma-mcp vision engine is bundled as an internal dependency. To install the plugin, you run the DSH CLI command dsh plugin --profile web add dsh-free-vision then restart your dsh web instance, after which the image_understand tool becomes available for use by DSH models. It supports multiple task types including OCR, UI analysis, debug screenshot description, and automatically processes large images with multi-cropping to preserve detail.

The plugin is designed for DSH users who need to add image recognition capability to their text-only model workflows, regardless of their technical skill level. After installation, you first get a free API key from one or more of the supported providers, then open the DSH settings menu and navigate to the Free Vision configuration panel. You enter your API key, select your preferred model provider, and save the configuration, which is stored locally at ~/.dsh/free-vision.json and takes effect on the next tool call. Any time your DSH model needs to analyze an image, it will call this tool with the image source and your text prompt, and return a textual description or analysis of the image content.

The plugin comes preconfigured for three providers with generous free quotas: Qwen3-VL-Flash from Alibaba Cloud Dashscope, Doubao Vision from Volcengine, and DeepSeek-OCR from Siliconflow. It also supports paid vision models from Zhipu AI and Tencent Hunyuan, plus custom OpenAI-compatible endpoints for any self-hosted or third-party vision model. All built-in providers allow overriding the default base URL, so you can route requests through a proxy or local API gateway. The plugin is released under the open-source MIT license, and free quota data is pulled from provider official pages, so users should verify current terms before relying on free access.

这是一款专为DeepSeek Harness开发的原生视觉插件,可为纯文本大模型赋予图片理解能力,支持截图分析、OCR识别、UI分析、报错排查等多种场景,优先调用各平台提供的免费额度视觉模型,无需额外配置MCP服务。安装方式为通过DSH CLI执行dsh plugin --profile web add dsh-free-vision,安装后重启dsh web即可启用通用的image_understand工具。插件支持多种任务模式,对大图会自动进行多裁剪处理以保证识别保真度。

面向需要给DSH纯文本模型添加视觉能力的开发者和普通用户,典型工作流程是:安装插件后,按需获取对应提供商的免费API Key,在DSH的设置界面打开「Free Vision」配置页面,填入API Key并选择想要使用的模型提供商,保存配置后即可调用。之后当模型需要分析图片时,会自动调用该工具,传入图片路径或链接以及问题描述,即可得到图片的文本分析结果。

插件默认使用通义千问Qwen3-VL-Flash,还有火山引擎豆包、硅基流动DeepSeek-OCR等多个免费额度选项,也支持智谱GLM、混元等付费模型以及自定义OpenAI兼容端点。所有提供商都支持覆盖API地址,可适配代理或本地服务,国内API默认直连避免代理导致的502错误。插件采用MIT许可证开源,免费额度来自各平台官方,可能会有变动,使用前需要自行核对。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意
  • Only 8 stars - very few users, little community feedback星标只有 8,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 dsh-free-vision

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

READMEREADME

dsh-free-vision

🌐 English | 中文

npm 版本 npm 下载 许可证 stars

DSH 免费视觉插件 — 让纯文本模型获得看图能力(截图、报错、UI 分析、OCR、文档),优先使用各平台免费视觉模型,零 MCP 配置。

Free vision plugin for DeepSeek Harness (dsh) — image understanding for text-only models using free-tier vision models, with zero MCP configuration.

为什么免费 / Why free

默认使用免费额度充足的提供商,无账单惊吓:

提供商 模型 免费额度 API Key 环境变量
qwen(默认) Qwen3-VL-Flash 阿里云百炼限免(激活送 50万 token) DASHSCOPE_API_KEY
volcengine 豆包视觉模型 火山引擎豆包免费 token(20万起,可申请 50万) VOLCENGINE_API_KEY
siliconflow DeepSeek-OCR 硅基流动 OCR 免费 SILICONFLOW_API_KEY
zhipu GLM-4.6V 按量 ZHIPU_API_KEY
hunyuan HY-Vision 按量 HUNYUAN_API_KEY
custom 任意 OpenAI 兼容 — CUSTOM_API_KEY + CUSTOM_BASE_URL + CUSTOM_MODEL_NAME

一张 1MB 截图 ≈ 2600 token,qwen 限免额度可分析约 19 万张图。 One 1MB screenshot ≈ 2,600 tokens ≈ $0.0006 on qwen; free quota covers ~190,000 images.

特性 / Features

  • 零 MCP 配置 — 不用改 cordis.patch.yml、运行时不用 npx:视觉引擎(luma-mcp)作为本包依赖内置,进程内启动
  • 单个通用工具 — image_understand(可用 config.toolName 改名)注册到 ctx.tools,每次请求模型都能看到
  • 免费优先、多提供商 — 千问 / 豆包 / 硅基流动免费档开箱即用;智谱 / 混元 / custom 可切换
  • 每个 Provider 可覆盖 API Base URL — 内置 Provider 可指向代理、API Gateway、本地服务或任意 OpenAI 兼容端点,无需改成 custom
  • 直连 — 子进程剥离代理环境变量,国内 API 直连(带代理会导致 502)
  • 任务模式 — auto | general | ocr | ui | debug | describe;大图自动多裁剪保真
  • 中英双语 — 工具描述与文档中英文都可用

安装 / Install

dsh plugin --profile web add dsh-free-vision

重启 dsh web 后,工具 image_understand 即可用。 Restart dsh web; the tool appears as image_understand.

设置界面 / Settings UI

重启 dsh web 后,打开 设置 → Free Vision 即可看到配置表单(API Key、提供商、 工具名等),由插件 schema 自动渲染。保存到 ~/.dsh/free-vision.json,下一次 调用立即生效,无需重启。

After restart, open Settings → Free Vision — a form for every config option, saved to ~/.dsh/free-vision.json, effective on the next tool call.

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-kimi-ppt 下一个 Next dsh-programming-mode →