MCP 服务器

MCP Academic Structure Server

Enables AI assistants to navigate, search, and analyze academic materials organized in cycles, semesters, and modules via Google Drive.

README

MCP Academic Structure Server

A powerful Model Context Protocol (MCP) server that provides intelligent access to your Google Drive-based academic structure. This server allows AI assistants like Claude to navigate, search, and analyze your academic materials organized in cycles, semesters, and modules.

Overview

The MCP Academic Structure Server transforms your Google Drive academic folders into an intelligent, searchable knowledge base that AI assistants can interact with. It provides:

Hierarchical Navigation: Browse through academic cycles, semesters, and modules
Intelligent Search: Find specific content across your entire academic structure
Path Validation: Verify and navigate to specific academic paths
Module Comparison: Compare different modules to understand relationships
Subject Discovery: Find all modules related to specific subjects

Project Structure

MCP-ENSIA/
├── src/
│   ├── clients/
│   │   └── drive_client.py          # Enhanced Google Drive API client
│   ├── core/
│   │   ├── academic_structure.py    # Main academic processor
│   │   ├── cycle_manager.py         # Cycle processing logic
│   │   ├── semester_manager.py      # Semester processing logic
│   │   ├── module_manager.py        # Module processing logic
│   │   └── category_manager.py      # Category processing logic
│   ├── models/
│   │   ├── academic_models.py       # Data models for academic structure
│   │   └── search_models.py         # Search-related models
│   ├── services/
│   │   ├── search_service.py        # Search functionality
│   │   ├── navigation_service.py    # Navigation services
│   │   └── comparison_service.py    # Module comparison services
│   ├── utils/
│   │   └── validation_utils.py      # Validation utilities
│   ├── server.py                    # Main MCP server
├── config/
│   ├── .env                        # Environment variables
│   ├── credentials.json            # Google Drive credentials (user-specific)
│   └── token.json                  # OAuth token (auto-generated)
├── tests/                          # Comprehensive test suite
├── requirements.txt                # Python dependencies
└── README.md                       # This file

MCP Tools Available

Tool	Description	Parameters
`get_academic_overview`	Complete overview with statistics	`max_depth` (optional)
`search_academic_content`	Search across academic structure	`query`, `scope`, filters
`navigate_to_path`	Navigate to specific academic path	`path`
`list_cycles`	List all academic cycles	None
`list_semesters`	List semesters in a cycle	`cycle`
`list_modules`	List modules in a semester	`cycle`, `semester`
`compare_modules`	Compare two modules	`module1_path`, `module2_path`
`find_modules_by_subject`	Find modules by subject	`subject`, `exact_match`
`get_structure_overview`	Detailed structure information	None
`validate_path`	Validate academic path format	`path`

Prerequisites

Python 3.8+
Google Drive API access: Access is exclusive for now. Help is welcomed in hosting this server for public access.
MCP-compatible AI assistant (like Claude Desktop)
Academic structure in Google Drive following the expected format

Installation & Setup

1. Clone the Repository

git clone <repository-url>
cd MCP-ENSIA

2. Install Dependencies

pip install -r requirements.txt

3. Test the Setup

# Test imports and basic functionality
python -c "
import sys
sys.path.append('src')
from src.clients.drive_client import EnhancedGoogleDriveClient
from src.core.academic_structure import AcademicStructureProcessor
print('All imports successful!')
"

Usage

Running the MCP Server

# Start the MCP server
python -m src.server

Integration with Claude Desktop

Add to your Claude Desktop MCP configuration:

{
  "mcpServers": {
    "academic-structure": {
      "command": "python",
      "args": ["-m", "src.server"],
      "cwd": "/path/to/MCP-ENSIA",
      "env": {
        "PYTHONPATH": "/path/to/MCP-ENSIA"
      }
    }
  }
}

Example Commands in Claude

Once integrated, you can ask Claude:

"Get an overview of my academic structure"
"List all my academic cycles" 
"Show me modules in First Superior Cycle, Semester 1"
"Search for automata theory content"
"Navigate to 3. First Superior Cycle/Semester 1/Theory of Computation"
"Compare Mathematics I from different semesters"
"Find all modules that teach machine learning"

Testing

Run Test Suite

# Install test dependencies
pip install pytest pytest-asyncio pytest-mock pytest-cov

# Run all tests
pytest tests/ -v

# Run with coverage
pytest tests/ --cov=src --cov-report=html

# Run specific test categories
pytest tests/test_imports.py -v           # Import tests
pytest tests/unit/ -v                     # Unit tests
pytest tests/integration/ -v              # Integration tests