DeepSeek Harness Plugin Hub

发布与管理完整 Harness Profiles,发现适合你的插件。

探索

插件目录环境预设文档中心动态

社区

发布插件联系我们报告问题

相关链接

Plugin Hub GitHubDeepSeek Harness 官方项目系统状态隐私说明
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

独立、非官方社区项目,与 DeepSeek 官方无隶属、授权或背书关系。

Video To Notes — DeepSeek Harness 插件(DSH Plugin)
DeepSeek Harness Plugin Hub
ProfilesPlugins分类动态文档登录管理 Profiles
ProfilesPlugins分类动态文档登录
← Plugins

dsh-video-to-notes

Video To Notes

可选启用的 DeepSeek Harness 技能包,可将课程、讲座、教程、纪录片、会议和演讲视频转换为结构化学习笔记。

插件会安装到这里;不确定时保持 web。

npx -y @deepseek-ai/dsh plugin --profile web add dsh-video-to-notes@0.1.0
README兼容性版本

兼容性与来源证明

Video To Notes 以 dsh-video-to-notes 发布,当前版本为 0.1.0。Plugin Hub 会校验它的 manifest,并保存精确安装来源,便于复现安装结果。

DSH 兼容范围
*
运行环境
any
发布来源
npm
Registry 更新时间
2026/9/21

版本

0.1.0stable
2026/9/21

相关插件

正在加载相关插件…

最新版
0.1.0
DSH
*
HMR
重启进程
Tree shaking
未声明可安全裁剪
解包体积
未提供
文件数
未提供
Surface
any
许可证
MIT
发布源
npm
GitHub
★ 0
周下载
0
最近提交
2026/9/21
查看源码 ↗
README Badge

点击下方 Badge 复制 Markdown,粘贴到 README 即可。

这是你的 Plugin?认领权益 · 优先安全扫描

验证 package.json 声明的 GitHub 仓库,即可管理这个公开页面。认领后,Hub 会优先安排当前版本的安全扫描,并在通过后公开展示结果。

认领这个 Plugin →
报告问题

相关插件

继续浏览 vision-media 分类下经过校验的插件。

Tool Describe Image@linxin666/dsh-tool-describe-image面向模型的 describe_image 工具,用于 dsh Web GUI:通过在兼容 OpenAI 的端点调用视觉语言模型,为文本模型提供图像理解能力,以描述一张图像(本地路径、http(s) URL 或附件引用)。可热插拔 —Vision Routerdsh-vision-router为纯文本 DeepSeek Harness 代理提供视觉能力:内置免费视觉链(无需密钥)+ 像素级视觉工具(问答、定位、裁剪、像素差异、颜色、OCR、SVG 描摹、抠图、截图)。一条命令即可安装,无需 Python,图像回合的使用方式与普通工具相同Vision Toolkit@anionex/dsh-vision-toolkit面向 Harness 原生集成的 DeepSeek 与 agent-vision-toolkit:图像问答、OCR、定位、UI 还原、像素差异、Artifacts 和 Web UI。Deepseek Ivideodeepseek-ivideoiPolloWork HyperFrames Video Studio,以及 27 个可编辑视频模板,以原生 DeepSeek Harness 对话视图呈现。

README

dsh-video-to-notes

Turn course, lecture, tutorial, documentary, meeting, and talk videos into structured study notes, inside DeepSeek Harness.

将视频(课程 / 讲座 / 教程 / 纪录片 / 会议 / 演讲等)自动转为高质量结构化学习笔记的 DSH 技能包。

看视频 ≠ 学会。笔记才是真正属于你的知识。

Install

dsh plugin --profile web add dsh-video-to-notes

Restart DSH afterwards. The bundle applies one patch row and changes nothing else.

What the bundle does

This package ships no runtime plugin logic. Its substance is a single Skill:

skills/video-to-notes/
├── SKILL.md                  skill instructions and routing description
├── scripts/transcribe.py     ffmpeg + Whisper speech-to-text
├── scripts/download_video.py yt-dlp network download
└── references/               note templates and the detailed workflow

cordis.patch.yml inserts one @deepseek-ai/dsh-skill-filesystem row that mounts the packaged skills/ directory as an isolated provider (includeDefaultRoots: false), so this bundle contributes the video-to-notes skill without touching the project or user skill roots. The skill root resolves from the profile's own module resolution — the installed npm identity — never from a path concatenated onto baseUrl.

Requirements

The skill drives local tools; DSH does not install them for you.

ToolRoleInstall
python3runs the bundled scriptsbrew install python3
ffmpeg / ffprobeaudio extraction, silence trimming, speed changebrew install ffmpeg
openai-whisperspeech recognitionpip3 install openai-whisper
yt-dlpoptional, only for video URLspip3 install yt-dlp

The skill instructs the agent to detect missing dependencies, name them, and wait for your explicit confirmation before installing anything.

What the skill does

  1. Detect dependencies — reports exactly what is missing and asks before installing.
  2. Acquire the video — a local path, or a URL downloaded with yt-dlp after you confirm the output directory.
  3. Transcribe — Whisper with per-scenario defaults (base for ordinary videos, turbo for long ones, --language zh for Chinese, optional --strip-silence and --speed acceleration).
  4. Write the notes — semantic chunking into a chapter outline, an optional Mermaid knowledge map for long content, ⭐/🔧/⚠️ emphasis markers, a closing quick-reference card, and a style matched to the content type (technical tutorial, classroom, talk, documentary).

Model choice, device selection (CUDA / Apple MPS / CPU), speed, and the note format are all decided by the skill; you only supply the video and any preferences.

Privacy and copyright

Transcripts and notes can contain sensitive material — review and delete them when they are no longer needed. Downloading third-party video must respect the source platform's terms; use it for personal study.

License

MIT