This repository is a mixed practice workspace around:

• Java student-management domain modeling
• AI-assisted development workflow
• MCP server integration for Cursor / Claude
• Dockerized local services

It is best understood as an AI-native engineering practice repo, not a fully finished production application.

• docs/PROJECT_SCOPE.md
• docs/SETUP.md
• docs/ARCHITECTURE.md

The most complete runnable part is:

• mcp-server

It provides a lightweight MCP server that can:

• expose database connection status as a resource
• run read-only SQL queries
• audit local Java files with simple rule-based checks
• generate tutoring plans for at-risk students

Infrastructure available in this repo:

• MySQL
• Redis
• MCP server container

These parts are still practice code or showcase material rather than a complete deployable product:

• src
• frontend

These parts are learning and practice material:

• StudentSystem
• JavaLearningDay1
• JavaOOP_Lesson01
• ruanjian2ban

.
|-- src/                  Java domain code and service logic
|-- frontend/             React UI experiments
|-- mcp-server/           Python MCP server
|-- StudentSystem/        Java basic student system practice
|-- JavaLearningDay1/     Java beginner exercises
|-- JavaOOP_Lesson01/     Java OOP exercises
|-- docker-compose.yml    Local infrastructure for practice
`-- Dockerfile            MCP server image definition

cd mcp-server
powershell .\deploy.ps1
.\venv\Scripts\python.exe .\server.py

Copy-Item .env.example .env
docker compose up --build

This starts:

• MySQL on 3306
• Redis on 6379
• MCP server on stdio inside the container

Use .env.example as the template.

Important:

• never commit a real .env
• keep API keys and database passwords local only

If you are reading this repo to learn AI-native development, start here:

1. docs/PROJECT_SCOPE.md
2. mcp-server/server.py
3. mcp-server/deploy.ps1
4. src/main/java/com/student/service

What this repo demonstrates:

• humans define architecture and guardrails
• AI helps implement and inspect
• MCP gives AI structured access to local project context

This repo intentionally keeps learning materials and practice code together.

That makes it useful for study, but it also means:

• some directories are exercises
• some are prototypes
• only part of the repo is production-like

• keep learning directories stable and separate from product work
• add real tests for mcp-server
• add a proper frontend app shell
• turn Java backend code into a standalone buildable service