McpLearningPath
Before deployment done in generate path
Ask AI about McpLearningPath
Powered by Claude Β· Grounded in docs
I know everything about McpLearningPath. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
Learning Path Generator with Model Context Protocol (MCP)
This project is a Streamlit-based web application that generates personalized learning paths using the Model Context Protocol (MCP). It integrates with various services including YouTube, Google Drive, and Notion to create comprehensive learning experiences.
Features
- π― Generate personalized learning paths based on your goals
- π₯ Integration with YouTube for video content
- π Google Drive integration for document storage
- π Notion integration for note-taking and organization
- π Real-time progress tracking
- π¨ User-friendly Streamlit interface
- π NEW: Secure user authentication and account management
- πΎ NEW: Automatic saving of learning paths and interview sessions
- π€ NEW: User profiles with progress history
- π― NEW: Mock interview sessions with AI evaluation
- π€ NEW: Voice assistant for hands-free interaction
Prerequisites
- Python 3.10+
- Google ai Studio API Key
- Pipedream URLs for integrations (YouTube and either Drive or Notion)
Installation
- Clone the repository:
git clone <repository-url>
cd mcpLearningPath
- Create and activate a virtual environment:
python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate
- Install the required packages:
pip install -r requirements.txt
- Initialize the database:
python init_database.py
- (Optional) Test the authentication system:
python test_auth.py
Configuration
Before running the application, you'll need to set up:
- Google API Key
- Pipedream URLs for:
- YouTube (required)
- Google Drive or Notion (based on your preference)
Running the Application
To start the application, run:
streamlit run app.py
The application will be available at http://localhost:8501 by default.
Usage
Getting Started
- Register an Account: Click "Register" in the sidebar to create your account
- Login: Use your credentials to login and access all features
- Configure APIs: Enter your Google AI Studio API key and Pipedream URLs in the sidebar
Learning Path Generator
- Select your preferred secondary tool (Drive or Notion)
- Enter your learning goal (e.g., "I want to learn python basics in 3 days")
- Click "Generate Learning Path" to create your personalized learning plan
- Your learning paths are automatically saved to your profile
Mock Interview Simulator
- Configure interview settings (type, role, language, difficulty)
- Fill in your candidate profile for personalized questions
- Take the pre-test to unlock live interview features
- Complete the interview with AI evaluation and feedback
- View your final report and recommendations
- All interview sessions are saved to your profile
Voice Assistant
- Enable voice mode in the interview section
- Use voice commands like "start interview", "next question", "repeat question"
- Get hands-free interview experience with text-to-speech feedback
Project Structure
app.py- Main Streamlit application with authentication integrationauth.py- Authentication system and user managementutils.py- Utility functions and helper methodsmock_interview.py- Mock interview system with AI evaluationvoice_assistant.py- Voice assistant with three-stage processingprompt.py- Prompt templaterequirements.txt- Project dependenciesinit_database.py- Database initialization scripttest_auth.py- Authentication system test suiteAUTHENTICATION_GUIDE.md- Detailed authentication documentationusers.db- SQLite database (created after initialization)
Database Schema
The application uses SQLite with the following main tables:
- users - User accounts and authentication data
- user_sessions - Login sessions and tokens
- learning_progress - Saved learning paths and progress
- interview_sessions - Mock interview results and reports
- user_preferences - User-specific application settings
See AUTHENTICATION_GUIDE.md for detailed database documentation.
Security Features
- π Secure password hashing with PBKDF2 and random salts
- π« Token-based session management with expiration
- β Input validation and SQL injection prevention
- π§Ή Automatic cleanup of expired sessions
- π§ Email format validation
- πͺ Strong password requirements enforcement
Complete Tutorial: How to Connect MCP Server to Cursor
Prerequisites & Setup
Step 1: Install Node.js( ignore if already installed)
Before we begin, you need Node.js installed on your computer to run the setup commands.
For Windows Users:
-
Open your web browser and navigate to:
https://nodejs.org -
Download the LTS version (Long Term Support) - it will show something like "18.17.0 LTS"
-
Run the downloaded .msi file
-
Follow the installation wizard:
- Accept the license agreement
- Keep the default installation path
- Important: Make sure "Add to PATH" is checked
- Important: Check "Automatically install the necessary tools" when prompted
-
Restart your computer after installation
-
Verify installation: Open Command Prompt or PowerShell and type:
textnode --versionnpm --versionYou should see version numbers for both commands.
For Mac Users:
- Visit
https://nodejs.organd download the LTS version - Run the downloaded .pkg file and follow the installer
- Verify installation in Terminal with
node --version
For Linux Users:
- Use your package manager or download from nodejs.org
- Verify with
node --versionin terminal
Step 2: Ensure Cursor is Installed (ignore if already installed)
Make sure you have Cursor IDE installed on your computer. If not:
- Visit
https://cursor.sh - Download and install Cursor for your operating system
- Launch Cursor to ensure it's working properly
Accessing Composio MCP Platform
Step 3: Navigate to Composio MCP
- Open your web browser
- Navigate to:
https://mcp.composio.dev - You'll see the Composio MCP homepage with various integration options
Step 4: Search for LinkedIn Integration
- Look for the search bar at the top of the page
- Type "LinkedIn" in the search box
- Press Enter or click the search icon
- You should see LinkedIn appear under "Marketing & social media" category with description: "LinkedIn is a professional networking platform enabling job seekers, companies, and thought..."
Step 5: Access LinkedIn Integration
- Click on the LinkedIn card/option
- You'll see two buttons: "View Servers" and "Create Server"
- Click "Create Server" to begin the setup process
Creating LinkedIn MCP Server
Step 6: Server Name Configuration
- A dialog box "Create MCP Server" will appear
- Enter a server name in the "Server Name" field
- Example:
LinkedIn-wp90c4ormy-linkedin-server - You'll see a green "Valid server name" message when it's acceptable
- Example:
- Click "Next" to proceed
Step 7: LinkedIn Authentication
- A LinkedIn login page will appear in a new tab/window
- Enter your LinkedIn credentials:
- Email or Phone number
- Password
- Click "Sign in"
- Complete any security verification if prompted by LinkedIn
- Grant permissions when LinkedIn asks for authorization
Important
Security Note: This uses OAuth 2.0, so your password is not stored - only a secure token is created.
Authentication & Configuration
Step 8: Integration Confirmation
After successful login, you'll see:
- "Integration Already Active" message with a green checkmark β
- This confirms your LinkedIn account is properly connected
- Click "Continue to Action Setup"
Step 9: Configure LinkedIn Actions
You'll see "Configure Actions for LinkedIn" with available options:
Recommended Actions to Enable:
- βοΈ Create a LinkedIn post - For posting content directly from Cursor
- βοΈ Get company info - For researching companies
- βοΈ Get my info - For accessing your profile data
- βοΈ Delete LinkedIn Post - For content management (optional)
Instructions:
- Check the boxes for actions you want to enable
- You can see the count at the top (e.g., "Selected Actions: 4/30")
- Click "Create Server" when satisfied with your selection
Step 10: Get Your Setup Command
After creating the server, you'll see:
-
"MCP Server Created" with green checkmark β
-
"Choose your client and copy the setup command:"
-
Make sure "Cursor" tab is selected
-
Copy the command that appears in the blue box - it will look like:
textnpx @composio/mcp@latest setup "https://mcp.composio.dev/composio/server/[your-unique-id]/mcp?include_composio_helper_actions=true&agent=cursor" "Testing" --client cursor -
Copy this entire command - you'll need it in the next step
Terminal Setup Commands
Step 11: Run the Setup Command
-
Open your terminal/command prompt:
- Windows: Press
Win + R, typecmd, press Enter - Mac: Press
Cmd + Space, type "Terminal", press Enter - Linux: Use your preferred terminal application
- Windows: Press
-
Navigate to any directory (location doesn't matter for this command)
-
Paste and run your copied command:
bashnpx @composio/mcp@latest setup "https://mcp.composio.dev/composio/server/[your-unique-id]/mcp?include_composio_helper_actions=true&agent=cursor" "Testing" --client cursor -
Press Enter and wait for the process to complete
-
Expected output: You should see messages indicating successful setup and configuration file creation
What this command does:
- Downloads the Composio MCP package temporarily
- Sets up the LinkedIn server connection
- Configures it specifically for Cursor
- Creates necessary configuration files automatically
Cursor Integration
Step 12: Access Cursor MCP Settings
Step 12: Access Cursor MCP Settings
- Open Cursor IDE
- Open Command Palette: Press
Ctrl + Shift + P(Windows/Linux) orCmd + Shift + P(Mac) - Type "Cursor Settings" and select it from the dropdown
- Look for "MCP" or "Tools & Integrations" in the left sidebar
- Click on the MCP section
Step 13: Verify LinkedIn Server
In the MCP Tools section, you should see:
- Your server name (e.g., "Testing")
- Green dot indicator showing it's active and connected
- Available actions listed, including:
LINKEDIN_CREATE_LINKED_IN_POSTLINKEDIN_GET_COMPANY_INFOLINKEDIN_GET_MY_INFOCOMPOSIO_SEARCH_TOOLS- And other helper functions
Step 14: Configuration File Verification
The setup automatically creates a configuration file at:
- Global:
~/.cursor/mcp.json - Project-specific:
.cursor/mcp.jsonin your project folder
The file should contain something like:
json{ "mcpServers": { "Testing": { "url": "https://mcp.composio.dev/composio/server/[your-id]/mcp?include_composio_helper_actions=true&agent=cursor" } } }
Testing Your Integration
Step 15: Test Basic Functionality
- Open Cursor's AI Chat: Press
Ctrl + I(Windows/Linux) orCmd + I(Mac) - Try these test commands:
Test 1 - Get Your Profile:
textGet my LinkedIn profile information
Test 2 - Company Research:
textGet company information for Google
Test 3 - Create a Post (be careful with this one):
textCreate a LinkedIn post about learning AI development with the text "Excited to be learning AI development! #AI #Development"
- Accept tool usage when Cursor prompts you to allow MCP tool execution
- Verify responses - you should get actual data from your LinkedIn account
Thatβs itβyou nailed it! Wishing you all the best!
