ZhuXinAI/sidesight

Name: ZhuXinAI/sidesight
Author: ZhuXinAI

Plugin插件 Native原生 ⭐ 2 MIT Vision & Media视觉与多媒体 Dev Workflow开发与代码工作流 Data & Analysis数据与分析

面向命令行优先的文本编码代理的视觉辅助工具。使用兼容OpenAI的多模态模型分析截图、图表、图表、用户界面差异和视频。

Project Overview项目介绍

Edited by Zhiwei Agent (zhiwei)由知微Agent(zhiwei)编辑

SideSight is a native plugin built exclusively for DeepSeek Harness that adds support for vision-based analysis of attached images in DSH conversations. It supports two native DSH installation paths: as a configurable bundled plugin or a filesystem-based skill, and it does not require running a separate MCP server. To install the plugin via DSH CLI, you can run npx @deepseek-ai/dsh plugin --profile web add github:ZhuXinAI/sidesight, then restart your DSH profile and configure the plugin through DSH Web Settings. After configuration, a new DeepSeek (SideSight vision) model provider is added to DSH, letting you select it to enable image input support.

This plugin is designed for developers and data analysts who need to analyze visual content like screenshots and diagrams directly within DSH conversations. After installation and configuration, you simply select the SideSight vision model provider before attaching or pasting an image. The plugin automatically pulls the raw bytes of the DSH attachment, processes it in a private system temporary directory, sends it to your configured vision model, then forwards the analysis results back to your main text conversation. Once analysis is complete, the temporary files are automatically deleted, so no residual files are left in your workspace.

SideSight is released under the open-source MIT license, and it requires a working Node.js and npm environment to function. The video frame sampling feature requires an installed copy of ffmpeg, which you can either install system-wide or specify the path via the SIDESIGHT_FFMPEG_PATH environment variable. It includes built-in security controls that canonicalize local paths against an allowlist, block symlink escapes, restrict remote URL requests to public HTTP/HTTPS endpoints, and never expose or print configured API keys. Before first use, you must configure your vision provider’s base URL, API key, and model name; API keys are stored with your DSH profile, so you should treat your profile configuration as sensitive.

SideSight 是专为 DeepSeek Harness 开发的原生插件，用于为 DSH 启用附加图像的多模态视觉分析能力。它提供两种原生集成安装方式：可作为可配置插件包安装，也可作为文件系统技能安装，整个过程无需启动额外 MCP 服务器。安装配置完成后，会在 DSH 中新增「DeepSeek (SideSight vision)」模型提供商，用户选择该提供商后即可正常处理粘贴或附加的图像。

这个插件面向需要在 DSH 对话中处理图像、截图等视觉内容的开发人员和数据分析师。典型工作流程为：安装配置插件后，选择SideSight 视觉模型提供商，附加或粘贴需要分析的图像，插件会自动获取图像原始字节，在系统临时目录处理后调用配置的视觉模型分析，再将结果返回给原文本对话模型。分析完成后会自动清理临时文件，不会在工作区留下残留文件。

SideSight 采用 MIT 许可开源，依赖 Node.js 和 npm 环境，视频抽帧功能需要额外安装 ffmpeg。它内置安全机制，会规范本地路径访问、限制远程 URL 请求范围，API 密钥不会被暴露或打印。使用前需要先配置视觉模型的基础地址、API 密钥和模型名称，密钥会随 DSH 配置文件存储，需注意保护配置敏感信息。

View on GitHub在 GitHub 查看

Send to your DSH发给你的 DSH 安装

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：sidesight（ZhuXinAI/sidesight）
仓库：https://github.com/ZhuXinAI/sidesight
本站详情页：https://www.yhbd.top/plugins/zhuxinai-sidesight/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 2 · 最近提交 2026-08-14 · 主语言 TypeScript

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。

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

npx @deepseek-ai/dsh plugin --profile web add github:ZhuXinAI/sidesight

把 ZhuXinAI/sidesight 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

18.9 KB · 缓存于 2026-08-27 05:41:48 · 18.9 KB

SideSight

For DeepSeek Harness

SideSight supports both of DeepSeek Harness's native extension paths: a filesystem skill for the simplest installation, and a configurable plugin bundle. Neither path starts an MCP server.

Install as a configurable DSH plugin

In a DSH conversation, ask the agent:

Please install ZhuXinAI/sidesight dsh plugin

Or directly run:

npx @deepseek-ai/dsh plugin --profile web add github:ZhuXinAI/sidesight

Restart the profile after changing its configuration:

npx @deepseek-ai/dsh --profile web

Open DSH Web Settings and choose the SideSight entry to fill in the Base URL, Vision model, and API key.

SideSight DSH Plugin Setup

Important: after configuring SideSight, select a model from the DeepSeek (SideSight vision) provider before attaching or pasting an image. The regular DeepSeek-V4-Flash / deepseek-official route is text-only and will reject the attachment before SideSight or any tool can run.

SideSight DSH Model

The API key field is write-only: an existing key is shown only as configured, never prefilled or returned. You can also set the plugin fields in the profile's cordis.patch.yml:

- update:
    id: sidesight
    config:
      baseUrl: https://provider.example/v1
      apiKey: your-key
      model: your-vision-model

The plugin contributes the configured values to DSH-managed shell calls and the skill maps them to SideSight's SIDESIGHT_BASE_URL, SIDESIGHT_API_KEY, and SIDESIGHT_MODEL variables when it invokes npx -y sidesight <command>. The API key value is not interpolated into the command text or ordinary tool result, and the skill forbids inspecting or printing it. Treat the profile configuration as sensitive because a manually entered key is stored with that profile.

The pasted-image bridge uses the package's bundled dist/cli.js when available. A GitHub/source install does not contain that generated directory, so the plugin automatically falls back to npx -y sidesight; no separate manual CLI install in the DSH profile is required. The fallback needs npx and npm access or a cached SideSight package.

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 →

渲染基于仓库 ZhuXinAI/sidesight 的 README.md。在 GitHub 查看原文 →