User Registration Chatbot with Local MCP Server

A complete user registration system built with Model Context Protocol (MCP) that handles user registration through a conversational chatbot interface. The system stores data locally in CSV format, replacing the need for external services like Google Sheets or Zapier.

🎯 Project Overview

This project demonstrates how to:

• Build a local MCP server from scratch in Python
• Create conversational AI interfaces with Streamlit
• Handle data persistence with local CSV storage
• Implement data validation and error handling
• Provide rich user interactions through natural language

🏗️ System Architecture

┌─────────────────┐    HTTP/JSON-RPC    ┌─────────────────┐    CSV Operations    ┌─────────────────┐
│ Streamlit UI    │ ◄─────────────────► │   MCP Server    │ ◄─────────────────► │ user_registrations│
│ (Chat Interface)│                     │ (Python Server) │                     │     .csv         │
└─────────────────┘                     └─────────────────┘                     └─────────────────┘

Key Components:

1. MCP Server (registration_mcp_server.py) - Handles all data operations
2. Streamlit Chatbot (streamlit_chatbot.py) - Provides user interface
3. CSV Storage (user_registrations.csv) - Local data persistence

✨ Features

🤖 Conversational Registration

• Natural language interaction
• Step-by-step data collection (Name → Email → DOB)
• Real-time validation and feedback
• Confirmation before saving

📊 Data Management

• Add Registrations - Store new user data
• View All Registrations - List all registered users
• Search Functionality - Find users by name or email
• Statistics Dashboard - View registration analytics
• Data Validation - Email format, date validation, duplicate prevention

🛠️ Technical Features

• Local MCP Server - No external dependencies
• CSV Storage - Simple, portable data format
• Error Handling - Comprehensive error messages
• Real-time Validation - Immediate feedback on data entry
• Responsive UI - Clean, intuitive Streamlit interface

📋 Prerequisites

• Python 3.8+
• pip package manager

🚀 Installation & Setup

Step 1: Clone or Download Files

Create a new directory and save these files:

• registration_mcp_server.py - The MCP server
• streamlit_chatbot.py - The Streamlit interface

Step 2: Install Dependencies

# Install MCP SDK for the server
pip install mcp

# Install Streamlit for the chatbot interface
pip install streamlit requests

Step 3: Start the MCP Server

Open a terminal and run:

python registration_mcp_server.py

Expected Output:

🚀 Starting Registration MCP Server...
📁 Data file: user_registrations.csv

🛠️  Available tools:
   • add_registration - Add new user registration
   • get_all_registrations - Retrieve all registrations
   • search_registrations - Search by name or email
   • get_registration_statistics - View registration stats
   • validate_registration_data - Validate data before saving

📚 Available resources:
   • file://user_registrations.csv - Direct CSV file access

✅ Server ready for connections!
   Connect your AI model to use these tools...

Step 4: Start the Streamlit Chatbot

Open a new terminal and run:

streamlit run streamlit_chatbot.py

The chatbot will open in your browser at http://localhost:8501

💬 Usage Guide

Registration Process

1. Start Registration: Type register or click the "Register" button
2. Enter Name: Provide your full name (minimum 2 characters)
3. Enter Email: Provide a valid email address
4. Enter Date of Birth: Use YYYY-MM-DD format (e.g., 1990-05-15)
5. Confirm Details: Review and confirm your registration

Available Commands

Command	Description	Example
register	Start new registration	register
show registrations	View all registered users	show registrations
search [query]	Search by name or email	search john or search @gmail
statistics	View registration statistics	statistics
help	Show available commands	help

Example Conversation Flow

User: register
Bot: Great! Let's start your registration. 📝
     What's your full name?

User: John Doe
Bot: Nice to meet you, John Doe! 👋
     Now, please provide your email address:

User: john@example.com
Bot: Perfect! 📧
     Now please enter your date of birth.
     Format: YYYY-MM-DD (e.g., 1990-05-15)

User: 1990-05-15
Bot: 📋 Please confirm your registration details:
     • Name: John Doe
     • Email: john@example.com
     • Date of Birth: 1990-05-15
     
     ✅ Everything looks good!
     • Type 'confirm' to complete registration

User: confirm
Bot: 🎉 Registration Completed Successfully!
     ✅ Successfully registered John Doe
     
     Registration Details:
     • Name: John Doe
     • Email: john@example.com
     • Date of Birth: 1990-05-15
     • Registered: 2025-01-15 14:30:25

🗂️ Data Storage

CSV Structure

Data is stored in user_registrations.csv with these columns:

Column	Description	Example
Name	User's full name	John Doe
Email	Email address	john@example.com
Date_of_Birth	Birth date (YYYY-MM-DD)	1990-05-15
Registration_Date	When registered	2025-01-15 14:30:25

Sample CSV Content

Name,Email,Date_of_Birth,Registration_Date
John Doe,john@example.com,1990-05-15,2025-01-15 14:30:25
Jane Smith,jane@example.com,1985-03-20,2025-01-15 14:35:12
Mike Johnson,mike@example.com,1992-07-08,2025-01-15 14:40:33

🔧 MCP Server Details

Available Tools

1. add_registration

   • Purpose: Add new user registration
   • Parameters: name, email, dob
   • Validation: Name length, email format, date format, duplicate email check

2. get_all_registrations

   • Purpose: Retrieve all registered users
   • Parameters: None
   • Returns: List of all registrations with IDs

3. search_registrations

   • Purpose: Search users by name or email
   • Parameters: query (search term)
   • Returns: Matching registrations

4. get_registration_statistics

   • Purpose: Get registration analytics
   • Parameters: None
   • Returns: Total count, age demographics, registration dates

5. validate_registration_data

   • Purpose: Validate data without saving
   • Parameters: name, email, dob
   • Returns: Validation results for each field

Data Validation Rules

Field	Validation Rules
Name	• Minimum 2 characters • Maximum 100 characters • Cannot be empty
Email	• Valid email format (user@domain.com) • Must be unique (no duplicates) • Cannot be empty
Date of Birth	• YYYY-MM-DD format • Cannot be in the future • Reasonable age limits (0-150 years)

📊 Features Showcase

🎯 Registration Statistics

📊 Registration Statistics:

📋 Total Registrations: 25
🌐 Unique Email Domains: 12
📅 First Registration: 2025-01-10 09:15:30
🆕 Latest Registration: 2025-01-15 16:45:22
💾 File Size: 1,234 bytes

👥 Age Demographics:
   Average Age: 32.4 years
   Youngest User: 18 years
   Oldest User: 67 years

📁 Data File: user_registrations.csv

🔍 Search Results

🔍 Search Results for 'john' (3 matches):

1. John Doe
   📧 Email: john@example.com
   🎂 Date of Birth: 1990-05-15
   📅 Registered: 2025-01-15 14:30:25

2. Johnny Smith
   📧 Email: johnny.smith@gmail.com
   🎂 Date of Birth: 1988-11-22
   📅 Registered: 2025-01-14 11:20:15
``

### **Common Issues**

**1. "MCP Server: Disconnected"**
- **Cause**: MCP server not running
- **Solution**: Start the server with `python registration_mcp_server.py`

**2. "Invalid email format"**
- **Cause**: Incorrect email format
- **Solution**: Use format like `user@domain.com`

**3. "Invalid date format"**
- **Cause**: Wrong date format
- **Solution**: Use YYYY-MM-DD format (e.g., `1990-05-15`)

**4. "Email already registered"**
- **Cause**: Duplicate email address
- **Solution**: Use a different email address

**5. "Failed to call MCP tool"**
- **Cause**: Communication issue between client and server
- **Solution**: Restart both the MCP server and Streamlit app

### **File Permissions**
- Ensure the directory is writable (for CSV file creation)
- Check that `user_registrations.csv` isn't locked by another program

### **Port Conflicts**
- MCP server uses stdio communication (no port needed)
- Streamlit uses port 8501 (changeable with `--server.port` flag)

## 📁 **Project Structure**

registration-chatbot/ ├── registration_mcp_server.py # MCP server with all tools ├── streamlit_chatbot.py # Streamlit chat interface ├── user_registrations.csv # Generated data file ├── README.md # This documentation └── requirements.txt # Python dependencies (optional)

## 🔄 **Extending the System**

### **Adding New Fields**
1. Update `REQUIRED_FIELDS` in the MCP server
2. Add validation in `RegistrationValidator` class
3. Extend the chatbot conversation flow
4. Update CSV operations

### **Database Integration**
Replace CSV operations with database calls:
```python
# Instead of CSV operations
def add_registration(self, name, email, dob):
    # Database insert logic here
    cursor.execute("INSERT INTO users VALUES (?, ?, ?)", (name, email, dob))

API Integration

Add external API calls to the MCP server:

@app.list_tools()
async def list_tools():
    return [
        Tool(name="send_email", description="Send welcome email"),
        Tool(name="sync_to_crm", description="Sync to CRM system")
    ]

🎓 Learning Outcomes

After completing this project, you'll understand:

1. MCP Architecture - How to build and structure MCP servers
2. Data Validation - Implementing robust input validation
3. Conversational AI - Creating natural language interfaces
4. File Operations - CSV reading, writing, and manipulation
5. Error Handling - Comprehensive error management
6. UI/UX Design - Building user-friendly chat interfaces

📚 Additional Resources

• MCP Documentation: https://modelcontextprotocol.io/
• Streamlit Docs: https://docs.streamlit.io/
• Python CSV Module: https://docs.python.org/3/library/csv.html

---

Happy Coding! 🎉

Built with ❤️ using Model Context Protocol, Streamlit, and Python