knowledge-workbench
Allows adding Evernote as a knowledge source by syncing notes via ENEX export or the Evernote Cloud API.
Allows adding Gitee repositories, including enterprise accounts, as a knowledge source and syncing Markdown/text files for unified search.
Allows adding GitHub repositories as a knowledge source, including self-hosted GitHub Enterprise instances, and syncing Markdown/text files for unified search.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@knowledge-workbenchsearch our knowledge base for the API authentication guide"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
知识库工作台 · Knowledge Workbench
一个链接 + 一个凭证,把分散在多处的知识与经验,收敛到一处、三端可用、本地或云上统一检索。 支持 ShowDoc / GitHub / Gitee / 禅道 / 印象笔记 / 腾讯文档,也支持 MCP 方式 与 通用 HTTP 接入任意系统。
为什么是它
接入极简:每个数据源只需要
baseUrl(一个链接)+credential(一个凭证),其余参数都有默认值。六 + 多平台:六大平台内置适配器,外加 MCP 客户端、本地目录、通用 OpenAPI 兜底。
双向 MCP:
作为 MCP 客户端 —— 把任意暴露 MCP 的服务器当成知识源接入;
作为 MCP 服务端 —— 让 Claude Desktop / 各类 IDE / Agent 直接检索与阅读你的知识库。
三端一致:Web / 移动(PWA,可加到主屏)/ 桌面(Electron)共用同一套前端与同一份后端数据。
本地优先,云上可选:服务端零第三方依赖(仅 Node ≥ 18 内置模块),不联网也能跑;一条
docker compose up即可上云。凭据不裸奔:数据库落地的凭证一律
AES-256-GCM加密,API 永不回显明文。
Related MCP server: Modular RAG MCP Server
功能矩阵
能力 | 支持 |
ShowDoc 接入 | ✅ 开放 API + Cookie |
GitHub 接入 | ✅ 仓库 Markdown/文本 |
Gitee 接入 | ✅ 含企业版 |
禅道接入 | ✅ REST v1 + 老版本 session,需求/任务/Bug/文档 |
印象笔记接入 | ✅ ENEX 导出(默认)/ Evernote Cloud API(可选依赖) |
腾讯文档接入 | ✅ 开放平台 API / 导出文件目录 |
MCP 方式整合 | ✅ 客户端 + 服务端 |
通用 HTTP 接入 | ✅ 可配置字段映射 |
本地部署 | ✅ |
云上部署 | ✅ Docker / Compose / 云托管 / GHCR |
三端可用 | ✅ Web / 移动 PWA / 桌面 Electron |
检索 | ✅ 中英文混合分词 + BM25 倒排 |
定时同步 | ✅ 增量、断点续传、失败隔离 |
快速开始(本地)
# 需要 Node.js ≥ 18
node server/src/index.js
# 浏览器打开 http://127.0.0.1:8787无需 npm install —— 服务端零依赖。
快速开始(Docker / 云)
docker compose up -d --build
# 打开 http://<host>:8787镜像构建与推送已在 .github/workflows/ci.yml 配置,推送 main 即自动构建并推送至
ghcr.io/zhangshifa/knowledge-workbench。
添加第一个数据源
点击左上角 +,选择平台(如 GitHub)。
填写:
链接:
https://api.github.com(GitHub.com)或自部署 GHE 地址凭证:Personal Access Token(裸串即可)
高级选项(JSON)(可选):
{"repos":"owner/repo","include":"\\.(md|txt)$"}
点「测试连接」→「保存并同步」。
回到顶部搜索框,跨库检索。
更详细的各平台参数见 docs/05-数据源接入手册.md。
MCP 双向
让 Agent 检索你的知识库(服务端 → 外部)
在支持 MCP 的客户端里这样配置:
{
"mcpServers": {
"knowledge-workbench": {
"command": "node",
"args": ["server/scripts/mcp-stdio.js"],
"env": { "KB_DATA_DIR": "./data" }
}
}
}提供工具:kb_search / kb_get_document / kb_list_sources / kb_list_documents / kb_sync_source;
资源:kb://doc/{id}。
把外部 MCP 服务器当作知识源(客户端 ← 外部)
新建数据源,平台选 MCP 服务器(客户端接入),高级选项填:
{ "transport": "stdio", "command": "npx", "args": ["-y", "@some/mcp-server"] }详情见 docs/04-MCP接入说明.md。
部署
形态 | 方式 | 说明 |
本地 |
| 默认绑定 127.0.0.1,数据落 |
桌面 |
| Electron 拉起本地服务 + 窗口 |
移动 | 浏览器打开同一地址 → 加到主屏 | PWA,离线可访问外壳 |
云(容器) |
| 挂载 |
云(托管) |
| 腾讯云 CloudBase 云托管 |
详见 docs/03-部署指南.md。
安全说明
凭证落地采用
AES-256-GCM加密,主密钥来自KB_MASTER_KEY或数据目录随机密钥文件。启用
KB_API_TOKEN后,所有/api/*需Authorization: Bearer。对外暴露服务时,请置于 HTTPS 反向代理之后(示例见
deploy/nginx.conf),并将KB_MASTER_KEY设为强随机值。
文档索引
docs/00-需求原文.md— 需求方原始表述(原封不动存档)docs/01-需求规格说明书.md— 工程化拆解docs/02-架构设计.md— 架构、统一模型、检索、同步、凭证安全docs/03-部署指南.md— 本地 / 桌面 / 移动 / 云docs/04-MCP接入说明.md— MCP 双向docs/05-数据源接入手册.md— 六平台 + 通用接入参数docs/06-API接口文档.md— REST API
许可证
MIT © zhangshifa
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to search, deep-read, and build knowledge bases from Markdown, PDF, DOCX, and PPTX documents via MCP tools for retrieval, document navigation, and ingestion.70616MIT
- AlicenseNot gradedqualityBmaintenanceConverts unstructured documents into a searchable knowledge base and exposes retrieval tools via MCP protocol for AI agents to query.MIT
- FlicenseNot gradedqualityCmaintenanceProvides RAG-based knowledge retrieval and document management as MCP tools, supporting hybrid search, reranking, and retrieval process visualization.
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to query and manage a document knowledge base via MCP, with RAG-powered search and grounded answers with citations.MIT
Related MCP Connectors
Your memory, everywhere AI goes. Build knowledge once, access it via MCP anywhere.
AgentDocs (agentdocs.eu) MCP: read, search, write, comment, share & attach images to Markdown docs.
Agentic search over your Dewey document collections from any MCP-compatible client.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/zhangshifa/knowledge-workbench'
If you have feedback or need assistance with the MCP directory API, please join our Discord server