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.

Intelhub — DSH Plugin for DeepSeek Harness
← Plugins
I

dsh-intelhub

Intelhub

The first zvec-native intelligence hub: automatically save information you come across (collections/folders/web pages/notes), with agent-powered hybrid semantic and keyword search including sources; zero daemons, zero API keys, and documents never leave your device; supports Obsidian feedback and sc

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

npx -y @deepseek-ai/dsh plugin --profile web add github:IKEASven69/dsh-intelhub#f31b7333407017bc9e494c264f42440b077c6bc4
READMECompatibilityVersions

Description

The first zvec-native intelligence hub: automatically save information you come across (collections/folders/web pages/notes), with agent-powered hybrid semantic and keyword search including sources; zero daemons, zero API keys, and documents never leave your device; supports Obsidian feedback and scheduled automation

Compatibility and provenance

Intelhub is published as dsh-intelhub and currently resolves to version 0.2.0. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
web
Release source
github
Registry updated
9/8/2026

Versions

0.2.0stable
9/8/2026
0.1.1stable
9/8/2026
0.1.0stable
9/7/2026

Related plugins

Loading related plugins…

Latest
0.2.0
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
web
License
MIT
Source
github
GitHub
★ 0
Weekly downloads
0
Last push
9/8/2026
View source ↗
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
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in

Related plugins

More verified plugins in search-research.

Browser Skill Dsh Plugin@wxg-prc-cpg/browser-skill-dsh-pluginDeepSeek Harness tool plugin that exposes BrowserSkill browser automation (browser_* tools) to the modelWeknora@wxg-prc-cpg/dsh-weknoraWeKnora knowledge retrieval tools for DeepSeek Harness (dsh): semantic search, document reading and RAG/agent answers over your own knowledge bases.Free Searchdsh-free-searchFree web search for DeepSeek Harness: 13 engines (Bing/DuckDuckGo/AnySearch/SearXNG/Exa/Tavily/Keenable/Firecrawl keyless; Parallel/Perplexity/SerpBase/DeepSeek with key) + time filtering + platform search + web_fetch, with web settings UI.Find Plugindsh-find-pluginFind DeepSeek Harness plugins inside the agent — live GitHub dsh-plugin topic search, ranked by stars.

README

dsh-intelhub · 情报站

第一个 zvec 原生情报站 —— 把刷到的信息,变成问得到的知识。

社媒采集 / 文件夹 / 网页 / 随手贴 → 自动沉淀 → 语义+关键词混合检索带出处 → Obsidian 反哺 零守护进程 · 零 API key · 文档不出本机

为什么需要它

已有的知识库插件是仓库:你手动拖文档,关键词搜索,用完留在插件里。 dsh-intelhub 是活水:信息自动流进来(采集脚本落盘即入库),按意思可搜(搜"怎么配置超时"能找到写"30 秒无响应则中断"的文档),元数据可过滤(只要宝玉xp 的精选层、赞过千),还能流出去(一键导出回 Obsidian)。

真实库实测(6300+ 篇微博采集,2.6 万块):换说法查询 6/6 第一名命中原帖,延迟 100-450ms。

用户故事

场景做法
"把 D:\knowledge-base 喂给你"kb_watch 注册常驻目录,采集脚本每 3 小时落盘的内容自动增量入库
"之前刷到过讲内容付费悖论的,说了啥"kb_search 语义命中,回答带 作者/赞数/文件#块号
调研做完,结论别丢kb_note 写入判断;下次问"我的判断是什么"直接召回
"把付费相关的导出到 Obsidian"kb_export 落盘普通 md,Obsidian 零插件直接索引
按发博节奏定时扫描kb_schedule 每天 09:00(或任意间隔),重启不丢

十个工具

kb_import(文件/文件夹,Obsidian 库直指) · kb_watch(常驻目录自动跟随) · kb_import_url(网页存档) · kb_note(直接写文本) · kb_search(混合检索+标量过滤) · kb_today(今日概览) · kb_evidence(判断找证据) · kb_export(反哺导出) · kb_schedule(定时任务) · kb_list/kb_delete

设置面板「情报站」:拖拽导入、常驻目录管理、定时任务、作者/阶段过滤检索、文件管理。

与同类插件的差别

dsh-intelhub通用文档库对话记忆类
数据从哪来自动长(采集管道+落盘监听)手动拖自动(但记对话)
检索语义+关键词+标量过滤关键词语义(记忆向)
数据形态带元数据的活信息静态文件会衰减的记忆
出口反哺 Obsidian 等工作流孤岛—

安装

dsh plugin add dsh-intelhub

首次导入自动下载本地向量化模型(约 30MB)到 ~/.dsh/dsh-intelhub/,之后全程离线。

架构

采集(opencli/rotate 定时) ─┐
文件夹 / Obsidian 库        ├→ 抽取(md/txt/pdf/docx)→ 标题边界分块
网页 URL / 对话粘贴        ─┘      → 本地向量化(e5-small,去语法嵌入)
                                   → zvec 进程内存储(WAL,标量字段)
                                   → RRF 混合检索(向量 0.75 + jieba FTS 0.25)
                                   → 带来源返回 ──→ kb_export 反哺
  • workspace 常驻目录:fs.watch 实时 + 5 分钟兜底轮询,双保险增量

  • frontmatter 标量(author/source/stage/tags/likes)进索引,可组合过滤

  • 引擎(skill 组合工作流)与数据(md 真相源)三层解耦,可持续演进

  • 安装提示(pnpm 用户):若 dsh plugin add 报 ERR_PNPM_IGNORED_BUILDS,在 profile 的 pnpm-workspace.yaml 追加 ignoredBuiltDependencies: [@zvec/zvec, onnxruntime-node, protobufjs, sharp];弱网环境可再加 supportedArchitectures: os/cpu/libc = current 跳过跨平台包下载

已知限制

  • 索引目录版本升级(schema 变更)会触发一次重建,旧目录自动弃用
  • 双机同时写同一索引目录不支持(索引目录随仓走的设计在路线图)
  • 极难例(运维近邻概念如"超时"vs"回滚")33M 本地模型区分度为零,嵌入层可配置升级

License

MIT