Utilizes jQuery for frontend interactions in the real-time chat interface.
Built on Laravel framework to power the backend of this e-commerce order management system with AI chat functionality.
Supports MySQL as an alternative database option for storing product and order information.
Integrates with OpenAI's GPT-3.5-turbo model to process natural language queries in Traditional Chinese, enabling intelligent responses about orders and products based on database information.
Uses SQLite as the default database for storing and querying product and order information.
Implements Tailwind CSS for the responsive design of the user interface.
MCP Demo Project
A Laravel-based Model Context Protocol (MCP) demonstration project featuring an e-commerce order management system with AI-powered chat functionality using OpenAI.
Features
- Order Management: Display and manage customer orders with pagination
- Product Catalog: Browse supermarket products (sodas, chips, ice cream, etc.)
- AI Chat Interface: Query order and product data using natural language in Traditional Chinese
- Database Integration: SQLite database with seeded sample data
- Modern UI: Responsive design using Tailwind CSS
Project Structure
Models
- Product: Represents supermarket items with name, price, description, stock, and category
- Order: Customer orders with transaction ID, customer name, amount, status, and product relationships
Database Schema
Products Table
id
- Primary keyname
- Product name (Traditional Chinese)description
- Product descriptionprice
- Product price (HKD)stock_quantity
- Available stockcategory
- Product category (飲料, 零食, 雪糕)created_at
,updated_at
- Timestamps
Orders Table
id
- Primary keytransaction_id
- Unique transaction identifier (TXN######)name
- Customer name (Traditional Chinese)amount
- Order total amountstatus
- Order status (pending, processing, completed, cancelled, refunded)product_id
- Foreign key to products tablequantity
- Quantity orderedcreated_at
,updated_at
- Timestamps
Sample Data
- 10 Products: Supermarket items including:
- 可口可樂 (Coca-Cola)
- 樂事薯片 (Lay's Chips)
- 哈根達斯雪糕 (Häagen-Dazs Ice Cream)
- 百事可樂 (Pepsi)
- And more...
- 500 Orders: Randomly generated orders with:
- Unique transaction IDs
- Chinese customer names
- Random products and quantities
- Various order statuses
- Realistic timestamps
AI Chat Functionality
The AI chat interface uses OpenAI's GPT-3.5-turbo model to answer queries about orders and products. The system:
- Processes natural language queries in Traditional Chinese
- Searches relevant data based on keywords and patterns
- Provides context to the AI model with retrieved data
- Returns intelligent responses about orders and products
Example Queries
- "顯示所有已完成的訂單" (Show all completed orders)
- "TXN000001 的訂單詳情" (Details for order TXN000001)
- "陳大明的所有訂單" (All orders for customer 陳大明)
- "有什麼飲料產品?" (What beverage products are available?)
Installation & Setup
Prerequisites
- PHP 8.1+
- Composer
- Node.js & npm (optional, for asset compilation)
Installation Steps
- Clone the repository
- Install dependencies
- Environment setup
- Configure OpenAI API
Add your OpenAI API key to
.env
: - Database setup
The project is configured to use SQLite by default:
- Start the server
- Access the application
Open your browser and navigate to
http://127.0.0.1:8000
Configuration
Database Configuration
SQLite (Default)
MySQL (Alternative)
For MySQL, create the database first:
OpenAI Configuration
Usage
Main Dashboard
- View paginated list of orders with details
- Browse product catalog
- Use AI chat to query data
AI Chat Interface
The chat interface supports various query types:
- Order lookup by transaction ID
- Customer order history
- Product searches
- Status-based filtering
- General inquiries about the data
API Endpoints
GET /
- Main dashboardPOST /chat
- AI chat endpoint
Technical Implementation
MCP Integration
The project demonstrates MCP concepts by:
- Data Retrieval: Structured database queries based on AI prompts
- Context Building: Formatting retrieved data for AI consumption
- Response Generation: Using OpenAI to generate intelligent responses
- User Interface: Real-time chat interface for natural language queries
Technologies Used
- Backend: Laravel 12, PHP 8.1+
- Database: SQLite/MySQL
- AI: OpenAI GPT-3.5-turbo
- Frontend: Blade templates, Tailwind CSS, jQuery
- HTTP Client: OpenAI PHP Client
Development
Adding New Features
- Create new models/controllers as needed
- Update database migrations and seeders
- Extend the chat functionality in
ChatController
- Add new UI components to the dashboard
Testing
Code Style
Troubleshooting
Database Issues
- Ensure SQLite is enabled in PHP
- For MySQL, check connection credentials
- Run
php artisan config:clear
after configuration changes
OpenAI Issues
- Verify API key is correct
- Check API quota and usage limits
- Ensure internet connectivity
Performance
- Consider adding database indexes for large datasets
- Implement caching for frequently accessed data
- Use Laravel queues for heavy AI operations
License
This project is open-sourced software licensed under the MIT license.
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Submit a pull request
Support
For issues or questions, please create an issue in the repository or contact the development team.
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
A Laravel-based Model Context Protocol demonstration that enables users to manage orders and query e-commerce data in Traditional Chinese through an AI-powered chat interface.
Related MCP Servers
- -securityFlicense-qualityA proof-of-concept Model Context Protocol server that enables LLM applications to interact with Uber Eats, allowing AI agents to browse and order food through natural language.Last updated -142Python
- -securityAlicense-qualityA Model Context Protocol server that enables interaction with the Tradovate API for managing trading contracts, positions, orders, and accounts.Last updated -JavaScriptMIT License
- -securityFlicense-qualityA Model Context Protocol server that enables Claude AI to interact with Paybyrd's payment processing API, allowing for creating payment links, processing refunds, and retrieving order information.Last updated -5TypeScript
- AsecurityAlicenseAqualityAllows AI assistants to seamlessly interact with Terminal.shop to browse products, manage shopping carts, place orders, and handle subscriptions through Terminal.shop's API.Last updated -19101JavaScriptMIT License