
面向教育科技工作流程的领域驱动MCP服务器
概览
面向教育技术工作流程的领域驱动MCP(模型上下文协议)服务器。该服务器展示经过验证的MCP工具,支持LangGraph代理、网页搜索和YouTube视频发现——所有这些都采用Clean Architecture和DDD组织。文档嵌入和检索实时运行在后端(教育技术系统后端嵌入服务);MCP服务器不再运行RAG。 外部MCP客户端调用Pydantic验证输入,委派给和,并通过基础设施层实现访问外部系统。 处理器在启用时使用,在需要时使用,并在协议边界处使用。 基于 src/mcp_server/ 的干净架构。域没有框架输入输出;基础设施实现域端口;接口和应用仅依赖于内部。 使用相同名称,加上测试、性能、代码健康、重构——详见 ARCHITECTURE.md § 变更日志层名称。 - Python(参见pyproject中的requires-python。
概述(AI 翻译)
领域驱动MCP(模型上下文协议)服务器,用于教育技术工作流。该服务器展示经过验证的MCP工具,支持LangGraph代理、网页搜索和YouTube视频发现——所有这些都采用Clean Architecture和DDD组织。文档嵌入和检索在后端(ed-tech-system-backend)嵌入服务中实时运行;MCP服务器不再运行RAG。
这个项目的作用
外部MCP客户端调用MCP工具,这些工具用Pydantic验证输入,委派给应用工作流和LangGraph代理,并通过在基础设施层实现的域端口访问外部系统。
MCP工具
| 工具 | 模块 | 目的 |
|---|---|---|
health_check |
custom_tools |
活性探针 |
search_youtube |
custom_tools |
教育YouTube搜索 |
build_lesson_enrichment_query |
custom_tools |
将课程元数据扩展为4–5个搜索词,用于文档/视频查询 |
research_article |
custom_tools_agent_workflows |
研究论文生成工作流程 |
content_generation |
custom_tools_agent_workflows |
课程/测验/项目内容生成 |
本页由 AI 翻译,技术细节请以原文为准。
README
以下为原文 README 节选,完整内容请查看原文。
ed-tech-system-mcp
Domain-Driven MCP (Model Context Protocol) server for ed-tech workflows. The server exposes validated MCP tools backed by LangGraph agents, web search, and YouTube video discovery — all organized with Clean Architecture and DDD. Document embedding and retrieval live in the backend (ed-tech-system-backend embedding service); the MCP server no longer runs RAG.
What this project does
External MCP clients call MCP tools that validate input with Pydantic, delegate to application workflows and LangGraph agents, and reach external systems through domain ports implemented in the infrastructure layer.
MCP tools
| Tool | Module | Purpose |
|---|---|---|
health_check |
custom_tools |
Liveness probe |
search_youtube |
custom_tools |
Educational YouTube search |
build_lesson_enrichment_query |
custom_tools |
Expand lesson metadata into 4–5 search terms for document/video lookup |
research_article |
custom_tools_agent_workflows |
Research article generation workflow |
content_generation |
custom_tools_agent_workflows |
Lesson/quiz/project content generation |
author_lesson_pipeline |
custom_tools_authoring |
Graph leaf → generate → validate → save (draft/publish) |
search_graph_nodes |
custom_tools_authoring |
Curriculum graph leaf search |
validate_lesson / validate_quiz / validate_project / validate_test_boilerplate |
custom_tools_authoring |
Content validation |
save_to_backend |
custom_tools_authoring |
Persist authored lesson tree via backend RPCs |
generate_mock_test_structure / validate_mock_test |
custom_tools_authoring |
Mock assessment scaffolding |
socratic_tutor |
custom_tools_socratic |
Socratic tutor turn |
collect_project_review_context / project_review |
custom_tools_project_review |
Project mentor review |
Handlers use MCP tool caching (when enabled), latency logging, privileged auth where required, and domain error mapping at the protocol boundary.
Integrations
| Capability | Integration |
|---|---|
| Curriculum graph + authoring RPCs | Supabase RPCs via authoring backend client (anon + manager JWT) |
| Web search | Tavily |
| Video discovery | YouTube Data API v3 |
| Agent orchestration | LangChain / LangGraph |
| Caching | Redis when CACHE_ENABLED=true |
| Document embedding / RAG | Backend ed-tech-system-backend embedding service + mcp-find-documents edge function |
Architecture
Clean Architecture under src/mcp_server/. Dependency rule: Domain has no framework I/O; Infrastructure implements Domain ports; Interface and Application depend inward only.
entrypoint → interface → application → domain ← infrastructure
| Layer | Path | Responsibility |
|---|---|---|
| domain | domain/ |
Entities, ports, validators, curriculum enums — no MCP/LangGraph/Supabase |
安装
This server does not publish a one-line install command.
Open the repository installation guide 新手?先看 mcpskill.org 教程 →配置
{
"mcpServers": {
"ed-tech-system": {
"command": "doppler",
"args": ["run", "--", "uv", "--directory", "/absolute/path/to/ed-tech-system-mcp", "run", "mcp-server"]
}
}
}