MCP Integration

Model Context Protocol tool calling demonstrations and integrations

What is MCP?

Model Context Protocol enables AI agents to interact with applications and services

Standardized Communication

JSON-RPC protocol for AI agents to call tools and access resources

Tool Calling

Expose functions that AI can invoke with proper parameters and validation

Context Awareness

Provide AI with real-time data and capabilities beyond text generation

Connected MCP Servers

Rolldice MCP Server

connected

Simple random number generation demonstrating MCP pattern

1 tools available

Type: http

Endpoint: https://rolldice-mcpserver-theta.vercel.app/api/mcp

Person App MCP Server

connected

CRUD operations for person management

4 tools available

Type: http

Endpoint: https://person-search-neon.vercel.app/api/mcp

Digital Twin MCP Server

connected

RAG-powered AI assistant for interview preparation

4 tools available

Type: http

Endpoint: https://digital-twin-vert-nu.vercel.app/api/mcp

Interactive Tool Calling Demo

Test MCP tools from Rolldice MCP Server

Available Tools:

Tool Output:

Select a tool to see output...

HTTP MCP Server Configuration

Connect to remote MCP servers via HTTP protocol

Rolldice MCP Server

{
  "mcpServers": {
    "rolldice": {
      "url": "https://rolldice-mcpserver-theta.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "Simple random number generation demonstrating MCP pattern"
    }
  }
}

Digital Twin MCP Server

{
  "mcpServers": {
    "digital-twin": {
      "url": "https://digital-twin-vert-nu.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "RAG-powered AI assistant for interview preparation"
    }
  }
}

Person App MCP Server

{
  "mcpServers": {
    "person-search": {
      "url": "https://person-search-neon.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "CRUD operations for person management"
    }
  }
}

Connection Features

🌐 Remote Access

Connect to MCP servers hosted anywhere via HTTPS endpoints

🔒 Secure Communication

All communication encrypted with TLS/SSL certificates

⚡ Low Latency

Deployed on Vercel edge network for global performance

📊 JSON-RPC 2.0

Standardized protocol for tool calling and responses

MCP Architecture

How HTTP MCP servers communicate with AI agents

AI Agent

Claude, GPT

→

HTTPS

MCP Server

HTTP Endpoint

→

API

Data Layer

Database/RAG

Protocol Flow: AI Agent sends JSON-RPC request → MCP Server processes → Returns structured response → AI integrates result into conversation

Implementation Examples

Real-world MCP integrations in this portfolio

1. Rolldice MCP Server (HTTP)

Simple dice rolling demonstrating HTTP MCP pattern with remote tool calling

// HTTP Request to MCP Server POST https://rolldice-mcpserver-theta.vercel.app/api/mcp // JSON-RPC 2.0 Request { "jsonrpc": "2.0", "method": "tools/call", "params": { "name": "rolldice", "arguments": {} }, "id": 1 } // Response { "jsonrpc": "2.0", "result": { "content": [{ "type": "text", "text": "Rolled: 4" }] }, "id": 1 }

2. Person App CRUD Operations

Manage person records through natural language commands

// AI Agent prompt "Create a new person named Sarah with email sarah@example.com" // MCP Tool Call { "tool": "create_person", "arguments": { "name": "Sarah", "email": "sarah@example.com", "age": 28 } }

3. Digital Twin RAG Queries (HTTP)

RAG-powered queries to professional profile via HTTP MCP server

// HTTP Request POST https://digital-twin-vert-nu.vercel.app/api/mcp // JSON-RPC Request { "jsonrpc": "2.0", "method": "tools/call", "params": { "name": "query_profile", "arguments": { "query": "Tell me about .NET experience" } }, "id": 1 } // RAG-Enhanced Response { "jsonrpc": "2.0", "result": { "content": [{ "type": "text", "text": "I have 4+ years of .NET development experience..." }], "metadata": { "sources": ["experience.md", "skills.md"], "confidence": 0.95 } }, "id": 1 }

Benefits of MCP Integration

For Users

✓Natural language interaction with applications
✓AI-powered automation and workflows
✓Intelligent data access and manipulation

For Developers

✓Standardized protocol for AI integration
✓Reusable tool definitions across platforms
✓Enhanced AI capabilities beyond LLM knowledge

Connected MCP Servers

Rolldice MCP Server

connected

Simple random number generation demonstrating MCP pattern

1 tools available

Type: http

Endpoint: https://rolldice-mcpserver-theta.vercel.app/api/mcp

Person App MCP Server

connected

CRUD operations for person management

4 tools available

Type: http

Endpoint: https://person-search-neon.vercel.app/api/mcp

Digital Twin MCP Server

connected

RAG-powered AI assistant for interview preparation

4 tools available

Type: http

Endpoint: https://digital-twin-vert-nu.vercel.app/api/mcp

Interactive Tool Calling Demo

Test MCP tools from Rolldice MCP Server

Available Tools:

Tool Output:

Select a tool to see output...

HTTP MCP Server Configuration

Connect to remote MCP servers via HTTP protocol

Rolldice MCP Server

{
  "mcpServers": {
    "rolldice": {
      "url": "https://rolldice-mcpserver-theta.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "Simple random number generation demonstrating MCP pattern"
    }
  }
}

Digital Twin MCP Server

{
  "mcpServers": {
    "digital-twin": {
      "url": "https://digital-twin-vert-nu.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "RAG-powered AI assistant for interview preparation"
    }
  }
}

Person App MCP Server

{
  "mcpServers": {
    "person-search": {
      "url": "https://person-search-neon.vercel.app/api/mcp",
      "type": "http",
      "timeout": 30000,
      "description": "CRUD operations for person management"
    }
  }
}

Connection Features

🌐 Remote Access

Connect to MCP servers hosted anywhere via HTTPS endpoints

🔒 Secure Communication

All communication encrypted with TLS/SSL certificates

⚡ Low Latency

Deployed on Vercel edge network for global performance

📊 JSON-RPC 2.0

Standardized protocol for tool calling and responses

MCP Architecture

How HTTP MCP servers communicate with AI agents

AI Agent

Claude, GPT

→

HTTPS

MCP Server

HTTP Endpoint

→

API

Data Layer

Database/RAG

Protocol Flow: AI Agent sends JSON-RPC request → MCP Server processes → Returns structured response → AI integrates result into conversation

Implementation Examples

Real-world MCP integrations in this portfolio

1. Rolldice MCP Server (HTTP)

Simple dice rolling demonstrating HTTP MCP pattern with remote tool calling

2. Person App CRUD Operations

Manage person records through natural language commands

3. Digital Twin RAG Queries (HTTP)

RAG-powered queries to professional profile via HTTP MCP server

Benefits of MCP Integration

For Users

✓Natural language interaction with applications
✓AI-powered automation and workflows
✓Intelligent data access and manipulation

For Developers

✓Standardized protocol for AI integration
✓Reusable tool definitions across platforms
✓Enhanced AI capabilities beyond LLM knowledge