dttxorg/deepseekeyes
可审计视觉与跨平台计算机使用运行时,适用于DeepSeek Harness——严格证据、健康检查故障转移、原始像素和Token核算。
Project Overview项目介绍
DeepSeekEyes is a native DeepSeek Harness plugin that adds auditable vision processing, MCP runtime support, and cross-platform computer use automation to DSH. DeepSeek’s top text models can reason about text and code but cannot process raw image pixels, and DeepSeekEyes fills this gap by adding a structured, auditable layer for visual input and automation control. All operations stay within the existing DSH conversation flow, so users never need to switch windows or manually transcribe content from images or automation outputs. To install the plugin, users can pull the package from npm and configure it directly within their DSH instance.
Common workflows supported by DeepSeekEyes include image understanding, browser automation, and structured MCP tool calls. For image understanding, a user pastes an image into the conversation, the configured multimodal model reads the raw pixels, and DeepSeek receives validated evidence to generate a final answer without leaving the task. For browser control, DeepSeek can issue commands to open web pages, scroll, click, and verify the result of each action from the fresh state returned by DeepSeekEyes. Users can also connect third-party MCP servers, with all tool outputs audited and bounded before being passed to DeepSeek for final reasoning.
DeepSeekEyes requires Node.js 22.19 or newer to run, and it is tested continuously on Ubuntu, macOS, and Windows. It is released under the permissive MIT open source license, so users can modify and redistribute it freely per the license terms. First-time users need to configure plugin settings in DSH, including model routing, token and call limits for automation, browser and desktop parameters, and MCP server connection details. The project maintains complete documentation for configuration, architecture, data retention, and troubleshooting to help users resolve any setup issues.
DeepSeekEyes 是专为 DeepSeek Harness (DSH) 打造的原生插件,提供可审计的视觉处理、MCP 应用层和跨平台计算机操作运行时。它为 DeepSeek 的纯文本大模型补充了视觉能力,让模型可以直接读取原始图片像素进行推理,同时支持浏览器自动化和原生桌面控制,整个流程都无需离开当前 DSH 对话,还会对所有生成的证据进行校验绑定,保障操作可追溯。
它支持多种典型工作流,包括粘贴图片后让多模态模型读取原始像素,生成可审计证据后返回 DeepSeek 做最终推理;也支持让 DeepSeek 控制浏览器打开页面、滚动点击,并在每一步操作后验证结果状态;还支持用户接入第三方 MCP 服务,让 DeepSeek 调用工具并审计返回的结果。适合需要给 DeepSeek 增加视觉理解和自动化操作能力的 DSH 开发者和普通用户使用。
它要求 Node.js 版本不低于 22.19,支持 Ubuntu、macOS 和 Windows 三大平台,用户可以通过 npm 安装,安装后需要在 DSH 配置中填写相关参数,比如视觉模型地址、自动化调用上限、MCP 服务配置等。项目采用 MIT 开源许可证,所有发布版本都经过持续集成测试,提供了完整的配置文档和故障排查指南。
请帮我安装这个 DSH 插件。安装前先完成【兼容性检查 + 安全性检查】,检查通过再动手。
插件:deepseekeyes(dttxorg/deepseekeyes)
仓库:https://github.com/dttxorg/deepseekeyes
本站详情页:https://www.yhbd.top/plugins/dttxorg-deepseekeyes/
本站登记:类型 client · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 7 · 最近提交 2026-09-22 · 主语言 JavaScript
按下面顺序执行,每步先把结论告诉我,再进入下一步:
【1 兼容性检查】
① 我这边:DSH 版本、Node 版本、操作系统、当前 profile(web / desktop)。
② 读它的 README、package.json、插件 manifest,列出它要求的 DSH 版本 / Node 版本 / 操作系统 / 外部依赖 / 需要另外先装的运行时。
③ 逐条比对,结论只写「满足 / 不满足 / 未知」三种;不满足的给出可行替代方案。
④ 检查是否和我已装的插件冲突:命令名重复、skill / tool 重名、端口占用、重复注册的 MCP server。
【2 安全性检查】
① 仓库可信度:和上面「本站登记」是否一致;star / fork 数、创建时间、最近提交,是否归档或长期停更。
② 安装脚本:逐行看 package.json 的 preinstall / install / postinstall,以及 install.sh、setup.ps1 之类脚本。出现 curl|bash、下载后直接执行、混淆代码、访问与插件功能无关的域名,立刻停下来告诉我,不要继续装。
③ 依赖:列出新增依赖,标出无人维护、或与知名包拼写近似的可疑包(typosquatting)。
④ 权限与副作用:它会读写哪些目录、访问哪些域名、需要哪些 DSH 权限(filesystem / network / shell / clipboard 等),以及怎么卸载和回滚。
⑤ 如果它要求 sudo / 管理员权限,或权限明显超出功能所需,先停下来问我。
【3 安装】
上面两步没有「不满足」和「高危项」时才执行;用官方推荐方式安装,不要自行提权。
【4 汇报】
用表格输出:检查项 / 结论 / 依据 / 是否需要我决策。拿不准的一律写「未知」并说明要我怎么确认——不要猜,也不要替我决定。
Send this message to DSH in your current session: it verifies compatibility and security first (answering met / not met / unknown item by item) and only installs once everything checks out — it will stop and ask you if it finds a high-risk item. The box scrolls; the copy is the full prompt. CLI install commands may not be accurate across systems, so DSH is the safer route.把上面这条消息直接发给当前会话里的 DSH:它会先核对兼容性与安全性(逐条给「满足 / 不满足 / 未知」),确认没问题再安装,有高危项会停下来问你。框内可滚动,复制到的是完整提示词;安装命令不一定准确,发给 DSH 更稳。
- Only 7 stars - very few users, little community feedback星标只有 7,几乎没人在用,遇到问题缺少社区反馈
- Desktop client: installation downloads an executable - verify the publisher and checksums桌面客户端:安装会下载可执行文件,请核对发布者与校验和
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:dttxorg/deepseekeyes
把 dttxorg/deepseekeyes 加入你的 DSH 配置(web profile)即可启用。
READMEREADME
DeepSeekEyes
Give DeepSeek sight without leaving the conversation.
An auditable vision, MCP and cross-platform Computer Use runtime for DeepSeek Harness.
简体中文 · Live screenshots · Quick start · How it works · Computer Use · MCP applications · Token accounting · X / @lucars2026
DeepSeek's strongest text models can reason about code, documents and interfaces, but they do not consume image pixels. DeepSeekEyes is the DSH runtime that makes those pixels auditable: it selects and health-checks visual routes, validates every nested evidence field, binds evidence to original bytes, records failover, and keeps DeepSeek as the reasoning model.
No window switching. No manual transcription. No lossy screenshot relay.
This is not another captioning window. It is the DSH auditable vision, Computer Use and MCP application runtime for image evidence, structured app calls, Browser automation and native Windows/macOS control.
Jev API control layer (0.9 candidate)
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 →
luobosibing2/dsh-jev-plugin
988hj7tczd-oss/dsh-computer-use
dsh-plugins/dsh-auxiliary