tutor
Allows retrieving Markdown documentation stored in MinIO S3-compatible storage, providing tools to fetch roadmap section content by roadmap, level, and section.
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., "@tutorShow me the Data Engineering level 0, section 1 content"
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.
Tutor — MCP server para Developer Docs de Z2H Academy
Qué es
tutor es un harness/agente que expone los Developer Docs (MDs de los roadmaps)
a TUIs agentic (OpenCode hoy; Codex y Claude Code en el futuro) vía el protocolo MCP.
Los MDs viven en un storage S3-compatible (MinIO en la VPS de producción,
https://minio.z2h-academy.com). El LLM de la TUI los consume directamente —
no las secciones TSX del web, que al contener componentes React consumen
más tokens y no son la fuente de verdad operativa.
Related MCP server: MCP Enhanced Data Retrieval System
Por qué existe
La academia tiene dos representaciones del contenido:
Representación | Consumidor | Fuente |
Secciones TSX ( | Estudiante en el web | Pipeline MD → TSX |
MD fuente ( | LLM en TUIs agentic | Storage S3-compatible |
El estudiante ejecuta labs en local siguiendo al pie de la letra las instrucciones
del MD. tutor es el puente que le da al LLM acceso a ese contenido sin pasar por
el pipeline de renderizado.
Estado actual
v0.3.0 — auth por estudiante en producción. Una sola tool (tutor_get_section),
MinIO en la VPS (https://minio.z2h-academy.com, bucket roadmaps), auth por
estudiante con device flow de GitHub + Edge Function de provisioning en
Supabase (service account read-only por estudiante). Flujo end-to-end validado
desde el entorno Sysbox canónico (SMOKE PASS).
Arquitectura
OpenCode / Codex / Claude Code
│
▼ MCP (stdio)
tutor
│ @aws-sdk/client-s3 (S3-compatible, forcePathStyle)
▼
MinIO (VPS producción, HTTPS vía Caddy)
│
▼
roadmaps/{roadmap}/{level}/{section}.mdDetalle completo en docs/ARCHITECTURE.md.
Stack
Node 20 + TypeScript 5 (ESM)
@modelcontextprotocol/sdk(protocolo MCP)@aws-sdk/client-s3(cliente S3-compatible para MinIO)zod(validación de input de tools)
Estructura
tutor/
├── scripts/
│ ├── seed.ts # crea bucket + sube un MD de prueba (contra VPS)
│ └── smoke.ts # test JSON-RPC end-to-end del server
└── src/
├── index.ts # entrypoint MCP (stdio transport)
├── server.ts # instancia McpServer
├── storage.ts # cliente S3/MinIO
└── tools/
└── get-section.ts # tool: tutor_get_sectionSetup (para un estudiante)
# 1. Dependencias
npm install
# 2. Autenticación (device flow de GitHub, una vez)
npm run build && node dist/cli/login.js
# → abre https://github.com/login/device, ingresá el código, autorizá
# → guarda credenciales en ~/.config/z2h/tutor.env (perms 600)
# 3. Correr el MCP server (stdio)
npm run devPara probar el loop completo sin OpenCode:
npm run smokeInfraestructura: MinIO corre en la VPS de producción (
~/tutor-minio/en62.238.103.174, verdocs/VPS.md). No hay MinIO local. Para el estudiante, ver docs/STUDENT-SETUP.md.
Tools
Tool | Input | Output |
|
| Contenido del MD de la sección |
Convención de paths en el bucket:
roadmaps/{roadmap}/{level}/{section}.mdEjemplo: roadmaps/data-engineering/nivel-0/1-section.md
Env vars
Var | Default | Nota |
|
| Endpoint de producción (VPS) |
| — | Se obtiene vía |
| — | Se obtiene vía |
|
| Bucket compartido |
Próximos pasos
Cómo extender el server: docs/DEVELOPMENT.md
Decisiones tomadas, diferidas y riesgos: docs/DECISIONS.md
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 Connectors
Team docs served to AI agents over MCP - search, Markdown reads, version pinning, read audit.
MCP-native collaborative markdown editor with real-time AI document editing
Connect AI agents to Filepad workspaces through OAuth MCP.
Give AI agents secure access to ZERNO project briefs, tasks, and context over remote MCP.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables automated GitHub login through browser automation using Playwright. Provides MCP tools for programmatic authentication to GitHub accounts with credential management.1
- FlicenseNot gradedqualityDmaintenanceEnables AI applications to access and contextualize organizational knowledge sources including GitHub repositories and internal documentation through standardized MCP protocol integration. Features OAuth 2.1 authentication, vector-based semantic search, and optimized context chunking for enterprise development workflows.
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to semantically search GitHub repository documentation by automatically fetching, vectorizing, and indexing content into an Upstash Vector database. It provides a standard MCP interface for agents to retrieve relevant documentation snippets through natural language queries.
- AlicenseNot gradedqualityBmaintenanceProvides AI agents with access to source code of all public GitHub repos, with integration for docs repos and compatibility with coding agents and MCP clients.61Apache 2.0
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/z2h-academy/tutor'
If you have feedback or need assistance with the MCP directory API, please join our Discord server