TeacherAssist-MCP
AI-powered Teacher Assistance MCP Server for managing student records with PostgreSQL, enabling create and read operations via MCP tools.
README
TeacherAssist-MCP
AI-powered Teacher Assistant MCP Server built with FastMCP, FastAPI, PostgreSQL, SQLAlchemy, and Email Notifications.
This project demonstrates how to build a Model Context Protocol (MCP) server that allows AI assistants such as Claude Desktop to manage student records through MCP tools while automatically sending registration emails to students.
Features
- Create students
- Retrieve all students
- Email notification on student registration
- PostgreSQL database integration
- SQLAlchemy ORM
- FastAPI REST API
- Claude Desktop MCP integration
- MCP Inspector support
- Beginner-friendly project structure
Tech Stack
- Python 3.11+
- FastMCP
- FastAPI
- PostgreSQL
- SQLAlchemy
- Psycopg2
- SMTP (Gmail)
- UV Package Manager
- Claude Desktop
- MCP Inspector
Project Structure
TeacherAssist-MCP/
│
├── app/
│ ├── db/
│ │ └── database.py
│ │
│ ├── models/
│ │ └── student.py
│ │
│ ├── schemas/
│ │ └── student.py
│ │
│ ├── services/
│ │ ├── student_service.py
│ │ └── email_service.py
│ │
│ ├── mcp/
│ │ ├── server.py
│ │ └── tools/
│ │ └── student_tools.py
│ │
│ └── main.py
│
├── .env
├── pyproject.toml
├── uv.lock
├── README.md
└── .venv/
Prerequisites
Install:
- Python 3.11+
- PostgreSQL
- Git
- UV
Verify:
python --version
git --version
psql --version
uv --version
Clone Repository
git clone https://github.com/YOUR_USERNAME/TeacherAssist-MCP.git
cd TeacherAssist-MCP
Install Dependencies
Using UV:
uv sync
Activate environment:
Windows
.venv\Scripts\activate
Linux/macOS
source .venv/bin/activate
PostgreSQL Setup
Create database:
CREATE DATABASE teacherassist;
Example connection:
DATABASE_URL=postgresql://postgres:password@localhost:5432/teacherassist
Email Configuration
Create a .env file:
DATABASE_URL=postgresql://postgres:password@localhost:5432/teacherassist
EMAIL_USER=your_email@gmail.com
EMAIL_PASSWORD=your_gmail_app_password
Gmail Setup
- Enable 2-Step Verification
- Open Google Account Security
- Generate App Password
- Use the generated password as
EMAIL_PASSWORD
Do NOT use your Gmail login password.
Run FastAPI Server
uv run uvicorn app.main:app --reload
Open:
http://127.0.0.1:8000/docs
Swagger UI will be available for testing APIs.
Run MCP Server
uv run python -m app.mcp.server
Expected output:
Starting MCP server 'TeacherAssist MCP'
transport 'stdio'
Test Using MCP Inspector
Run:
npx @modelcontextprotocol/inspector uv run python -m app.mcp.server
Open the generated URL.
Available tools:
- create_student
- get_students
Claude Desktop Configuration
Open Claude Desktop MCP configuration file and add:
{
"mcpServers": {
"Teacher-Assistance": {
"command": "D:\\Projects\\TeacherAssist-MCP\\.venv\\Scripts\\python.exe",
"args": [
"-m",
"app.mcp.server"
],
"cwd": "D:\\Projects\\TeacherAssist-MCP"
}
}
}
Replace the paths with your own project location.
Restart Claude Desktop.
API Examples
Create Student
POST /students
{
"student_id": 101,
"name": "Monirul Islam",
"email": "student@example.com"
}
Response:
{
"student_id": 101,
"name": "Monirul Islam",
"email": "student@example.com"
}
A confirmation email will automatically be sent.
Get Students
GET /students
Response:
[
{
"student_id": 101,
"name": "Monirul Islam",
"email": "student@example.com"
}
]
Example Claude Desktop Prompts
Create a student with ID 1001, name Monirul Islam, email monirul@gmail.com
Show all students
Claude will invoke MCP tools automatically.
Common Issues
ModuleNotFoundError: app
Run from project root:
uv run python -m app.mcp.server
PostgreSQL Connection Error
Verify:
- PostgreSQL service is running
- Database exists
- Credentials are correct
- DATABASE_URL is correct
Claude Desktop Cannot Connect
Verify:
- Correct Python executable path
- Correct cwd path
- MCP server runs manually
- Claude Desktop restarted after config change
Gmail Authentication Failed
Use a Google App Password.
Do not use your normal Gmail password.
Duplicate Student ID
Student IDs must be unique.
Use a new ID when creating students.
Future Enhancements
- Update Student
- Delete Student
- Attendance Management
- Grade Calculation
- GPA Calculation
- Student Reports
- AI-powered Analytics
- Authentication & Authorization
- Multi-user Support
License
MIT License
Author
Monirul Islam
TeacherAssist-MCP demonstrates how AI assistants can interact with PostgreSQL databases through FastMCP and the Model Context Protocol (MCP), while integrating FastAPI APIs and automated email notifications.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。