A DeepSeek Harness community plugin that builds auditable field-study knowledge graphs and a read-only teaching assistant for geography fieldwork — local-first, Windows 10/11, Cordis plugin system.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-08-30
License
MIT
01What can it help you accomplish?
把地理野外实习资料整理成可审核、可追溯的知识图谱
可审核、可追溯的知识图谱，用于现场导览、教学讲解和问答
地理野外实习的师生（现场导览、教学讲解与答疑）
以只读野外助教模式回答概念、邻接关系、位置、气象和行程查询
现场导览、教学讲解与问答；答不上来的问题自动登记为知识缺口待办
野外实习带队教师与学生，需要现场即时答疑
用知识演化模式补齐野外助教答不上来的知识缺口
按被问次数补资料、发布后逐条销掉缺口待办
资料整理与课程建设人员
02How to install into DeepSeek Harness
Prerequisites
- Node.js 24，或 Node.js 22.19 及以上版本
- pnpm
- Python 3.11
Installation steps
- 01
$ git clone <你的仓库地址>
- 02
cd <仓库目录>
- 03
.\安装问山.bat
- 04
.\启动.bat
Verify the integration
- 打开主界面：http://127.0.0.1:3080/
- 打开知识图谱：http://127.0.0.1:3080/wenshan/knowledge-graph
- 打开照片地图：http://127.0.0.1:3080/wenshan/photo-map
- 打开知识星系：http://127.0.0.1:3080/wenshan/knowledge-galaxy
03DSH integration and capability boundaries
Cordis plugin system (bundles a customized DeepSeek Harness source; DSH presets + wenshan.cordis.yml mount config; extensions dsh-georag / dsh-wenshan / dsh-wenshan-intake)
知识演化与资料收录
文字、PDF、Word、图片、录音和常见 GIS 文件→知识草稿，人工确认后再发布为正式知识
野外助教查询
只读正式知识库→概念检索、邻接关系、位置、气象和行程查询
知识缺口闭环
野外助教答不上来的问题→缺口待办；知识演化模式按被问次数补资料、发布后逐条销掉
本地持久化
原始资料、提取结果、草稿、发布记录、缺口记录→分开保存，重启后可恢复
原始资料、提取结果、草稿、发布记录、缺口记录写入本机磁盘并分开保存
04Who is it for? When not to use it?
Good for
- 地理野外实习的师生（现场导览、教学讲解与答疑）
- 野外实习带队教师与学生，需要现场即时答疑
- 资料整理与课程建设人员
Not for
- 提供的启动脚本仅面向 Windows 10/11，其他平台需自行适配。
- 非官方社区插件与二次开发项目，不代表 DeepSeek 或 DeepSeek Harness 官方发行版或上游项目；仓库内保留一份经过定制的 DeepSeek Harness 源码。
05Compatibility, maintenance and safety notes
- 提供的启动脚本仅面向 Windows 10/11，其他平台需自行适配。
- 野外助教没有资料写入和知识发布工具，只能登记缺口。
- 非官方社区插件与二次开发项目，不代表 DeepSeek 或 DeepSeek Harness 官方发行版或上游项目；仓库内保留一份经过定制的 DeepSeek Harness 源码。
MIT · DeepSeek Harness 社区插件（curated），无已发布的 GitHub Release
06Frequently asked questions
How do I install dsh-wenshan with DeepSeek Harness?
Clone the repo, then run `.\安装问山.bat` to install and build the bundled DeepSeek Harness plus knowledge-graph, document-extraction, audio-transcription and GIS dependencies, and `.\启动.bat` to start. Requires Windows 10/11, Node.js 24 (or 22.19+), pnpm and Python 3.11.
What are the prerequisites?
Windows 10/11, Node.js 24 or Node.js 22.19 and above, pnpm, and Python 3.11. Before installing, confirm `node`, `pnpm` and `py -3.11` are on your command line (install pnpm with `npm install -g pnpm`).
How does it connect to DeepSeek Harness?
It is built on deepseek-ai/deepseek-harness and reuses its Cordis plugin system. It ships a customized DeepSeek Harness source for a reproducible environment, plus presets and `wenshan.cordis.yml` for the DSH preset and product mount configuration. Extensions include `dsh-georag`, `dsh-wenshan` and `dsh-wenshan-intake`.
How is it different from alternatives?
It is a non-official community plugin and a secondary-development project — not an official DeepSeek / DeepSeek Harness release and not representing the upstream project. It is purpose-built for geography fieldwork: an auditable knowledge graph plus a read-only field-teaching assistant, not a general-purpose coding agent.
How do I verify it started, and what if something fails?
After starting, open `http://127.0.0.1:3080/` (main UI), `.../wenshan/knowledge-graph`, `.../wenshan/photo-map` and `.../wenshan/knowledge-galaxy`. Configure your model and API Key under Settings → Model. The startup scripts target Windows 10/11 only; there is no documented rollback step — reinstall or restore the repo to recover.
07Related DSH workflows
weknora
by tencent
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
mirage
by strukto-ai
The World's First Unified Virtual Filesystem For AI Agents
reme
by agentscope-ai
ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.
memsearch
by zilliztech
A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
08Data and sources
问山是面向地理野外实习的 DeepSeek Harness 社区插件与二次开发项目。
本项目基于 [deepseek-ai/deepseek-harness](https://github.com/deepseek-ai/deepseek-harness) 开发，沿用其 Cordis 插件体系。
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-08-30. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.