Study Tools MCP Server
AI-powered study assistant built with Model Context Protocol (MCP) - Generate quizzes, flashcards, summaries, and concept explanations from your study materials. Integrates with Claude Desktop and provides a FastAPI web interface.
Installation
npx study-tools-mcp-serverAsk AI about Study Tools MCP Server
Powered by Claude Β· Grounded in docs
I know everything about Study Tools MCP Server. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
Study Tools MCP π
An AI-powered study assistant built with Model Context Protocol (MCP) that generates quizzes, flashcards, summaries, and concept explanations from your study materials.
π― Features
- Smart Summarization β Generate concise summaries from study materials
- Quiz Generation β Create customizable quizzes with difficulty levels
- Concept Explanation β Get beginner/intermediate/advanced explanations
- Flashcards β Auto-generate flashcard decks from documents
- Comparison Tool β Compare and contrast multiple concepts
- MCP Integration β Works directly with Claude Desktop
- Web UI β Standalone chat interface with FastAPI backend
π οΈ Tech Stack
- Backend: FastAPI + Python 3.10
- AI Framework: Model Context Protocol (MCP)
- AI: OpenAI API
- Document Parsing: PyPDF2, pdfplumber, python-docx
- Frontend: Vanilla JavaScript, HTML, CSS
- Cloud: AWS EC2 + S3 + Secrets Manager
- CI/CD: GitHub Actions
π Quick Start
Prerequisites
- Python 3.10+
- OpenAI API key
Installation
- Clone the repository:
git clone https://github.com/francis-rf/study-Tools-mcp-server.git
cd study-Tools-mcp-server
- Install dependencies:
pip install -r requirements.txt
- Create
.envfile:
cp .env.example .env
# Edit .env and add your OPENAI_API_KEY
- Add study materials:
Place PDF or Markdown files in data/notes/:
data/notes/
βββ Machine Learning.pdf
βββ Your Notes.md
- Run the application:
python app.py
- Open browser:
http://localhost:8080
π³ Docker Deployment
Build and Run
docker build -t study-tools-mcp .
docker run -p 8080:8080 --env-file .env study-tools-mcp
βοΈ AWS Deployment
Services Used
| Service | Purpose |
|---|---|
| EC2 (t2.micro) | Hosts the Docker container |
S3 (study-tools-mcp-materials) | Stores PDF study materials |
Secrets Manager (study-tools-mcp) | Stores OpenAI API key |
| IAM Role | Grants EC2 access to S3 and Secrets Manager |
Setup
- Store OpenAI API key in AWS Secrets Manager under secret name
study-tools-mcp - Upload PDFs to S3 bucket
study-tools-mcp-materials - Launch EC2 instance with IAM role attached (
study-tools-mcp-ec2-role) - SSH in, install Docker, clone repo and run container
βοΈ GitHub Actions CI/CD
Automated deployment is configured via .github/workflows/deploy.yml.
Workflow: Deploy to AWS EC2
On every push to main, the pipeline:
- Checks out the code
- SSHs into the EC2 instance
- Pulls latest code from GitHub
- Rebuilds the Docker image
- Restarts the container with zero downtime
Required GitHub Secrets
| Secret | Description |
|---|---|
EC2_HOST | EC2 instance public IP |
EC2_USER | ubuntu |
EC2_SSH_KEY | Contents of the .pem key file |
Workflow Status
π Project Structure
study-Tools-mcp-server/
βββ app.py # FastAPI web application
βββ src/study_tools_mcp/
β βββ server.py # MCP server entry point
β βββ config.py # Configuration (Secrets Manager + .env fallback)
β βββ tools/ # Quiz, flashcards, summarizer, explainer
β βββ parsers/ # PDF and Markdown parsers
β βββ utils/ # Logger
βββ static/ # Frontend assets
βββ templates/ # HTML templates
βββ data/notes/ # Study materials (local only β S3 on AWS)
βββ logs/ # Application logs
βββ .github/workflows/ # CI/CD
β βββ deploy.yml
βββ Dockerfile
βββ requirements.txt
βββ pyproject.toml
π‘ API Endpoints
| Method | Endpoint | Description |
|---|---|---|
| GET | / | Web UI |
| GET | /health | Health check |
| GET | /api/files | List available study materials |
| POST | /api/chat | Chat with streaming |
| POST | /api/chat/clear | Clear conversation history |
π Claude Desktop Integration
Add to %APPDATA%\Claude\claude_desktop_config.json:
{
"mcpServers": {
"study-tools-mcp": {
"command": "uv",
"args": ["--directory", "C:\\path\\to\\study-tools-mcp", "run", "study-tools-mcp"]
}
}
}
Restart Claude Desktop β the tools will be available automatically.
πΈ Screenshots
Study Tool AI Interface with quiz generation
Study Tool AI Integration with Claude Desktop
π License
MIT License
