MCP Website Chatbot

A production-grade AI chatbot for srinivasanramanujam.sbs with live data retrieval via MCP (Model Context Protocol) and RAG (Retrieval-Augmented Generation).

🚀 Features

• Live Data Integration – MCP tools for real-time information retrieval

• RAG Support – Static knowledge base from website content, blogs, and FAQs

• Hallucination Prevention – Strict guardrails against fabrication and misinformation

• Beautiful UI – Modern, responsive chat interface

• Production-Ready – Scalable backend with proper error handling

• Health Monitoring – Built-in health checks and uptime tracking

📋 Requirements

• Node.js 16+

• npm or yarn

• OpenAI API key (for production use)

🛠️ Installation

📁 Project Structure

🔌 API Endpoints

POST /api/chat

Send a message and get a response.

Request:

Response:

GET /api/health

Check server health.

Response:

GET /api/system-prompt

Retrieve the system prompt (for debugging).

🎯 How It Works

1. User sends a message via the chat UI

2. Server analyzes if live data is needed (time-sensitive, external sources)

3. MCP tools are invoked if necessary to fetch real-time data

4. Response is generated using the system prompt guidelines

5. Assistant responds with proper citations and source attribution

🔐 Security Features

• ✅ No system prompt exposure to users

• ✅ Input validation and sanitization

• ✅ Rate limiting ready (add middleware as needed)

• ✅ Error handling without leaking internal details

• ✅ CORS headers (add if deploying to production)

🌐 Deployment

Option 1: Vercel (Recommended)

Option 2: Heroku

Option 3: Docker

Create a Dockerfile:

🎨 Customization

Update Website Info

Edit server.js and update the system prompt or knowledge base.

Change UI Theme

Modify the CSS in public/index.html gradient colors and styling.

Add Real API Integration

Replace mock MCP tools in server.js with real OpenAI/Claude API calls.

📝 System Prompt Highlights

• Live-first philosophy – Prioritizes current data over static knowledge

• Hallucination prevention – Refuses to guess or invent information

• Transparent reasoning – Cites sources and explains reasoning

• Professional tone – Clear, concise, helpful communication

• Safety guardrails – Rejects prompt injection and abuse

🚦 Next Steps for Production

1. Integrate OpenAI/Claude API – Replace mock responses

2. Add MCP server – Real connection to external tools

3. Set up database – Store conversations and user data securely

4. Add authentication – Protect sensitive endpoints

5. Configure CORS – Allow cross-origin requests from your domain

6. Enable logging – Monitor and debug in production

7. Add rate limiting – Prevent abuse and control costs

📧 Support

For questions or issues, contact the site owner at srinivasanramanujam.sbs

📄 License

MIT License – See LICENSE file for details