Skip to main content

MCP Server Introduction

Bundleport provides a Model Context Protocol (MCP) server that enables AI assistants and agents to interact with hotel booking services through a standardized, tool-based interface. The MCP server exposes hotel booking capabilities as tools that AI models can discover and use autonomously, making it easy to build conversational booking experiences.

What is MCP?​

The Model Context Protocol (MCP) is an open standard that enables AI assistants to securely interact with external systems through a structured tool interface. Instead of requiring custom integrations for each AI framework, MCP provides a universal protocol that works with any MCP-compatible client.

Bundleport MCP Server​

The Bundleport MCP server (hotel-booking-server) exposes hotel booking operations as tools that AI agents can use to:

  • Search for hotel availability across multiple suppliers
  • Get detailed quotes with pricing and cancellation policies
  • Confirm bookings with traveller and payment information
  • Retrieve booking details for existing reservations
  • Cancel bookings when needed
  • Access hotel content including descriptions, amenities, and images

Architecture Overview​

AI Agent / LLM
    ↓
MCP Client (Vercel AI SDK, LangChain, etc.)
    ↓
MCP Server (HTTP/SSE)
    ↓
Bundleport Booking Services
    ↓
Provider Integrations
  • MCP Tools translate natural language requests into structured booking operations
  • Adapters map MCP tool parameters to Bundleport's internal booking services
  • Shared Services ensure consistent business rules, pricing logic, and audit trails across all interfaces (REST, GraphQL, MCP)

Available Tools​

The MCP server provides six main tools:

  1. availability - Search hotel availability. Returns top offers in condensed AI-optimized format.
  2. quote - Get detailed pricing and cancellation policy for a selected option.
  3. book - Confirm a hotel booking with traveller and payment details.
  4. cancel - Cancel an existing hotel booking.
  5. bookingDetail - Get detailed information about an existing booking.
  6. getHotels - Retrieve hotel content (descriptions, amenities, location, images, rooms).

Key Features​

  • AI-Optimized Responses: Tool responses are formatted to be concise and context-aware, reducing token usage while maintaining essential information
  • Context Preservation: The server maintains context across tool calls, allowing agents to reference previous results (e.g., optionRefId from search used in quote)
  • Error Handling: Structured error responses help AI agents understand failures and suggest corrective actions
  • Stateless Design: The server is stateless, making it easy to scale and deploy

Use Cases​

  • Conversational Booking Assistants: Build chatbots that help users find and book hotels through natural conversation
  • AI-Powered Travel Agents: Create intelligent agents that can search, compare, and book hotels autonomously
  • Multi-Step Workflows: Enable complex booking flows where the AI agent orchestrates multiple tool calls (search → quote → book)
  • Content-Enhanced Experiences: Combine booking capabilities with hotel content retrieval for rich, informative responses

Getting Started​

To use the Bundleport MCP server, you'll need:

  1. MCP Server Endpoint: The server URL (https://api.connect.bundleport.com/mcp)
  2. MCP-Compatible Client: A client library that supports MCP (e.g., Vercel AI SDK, LangChain with mcp-use)
  3. Bundleport Credentials: API keys and access IDs configured for the suppliers you want to use

The server uses Server-Sent Events (SSE) transport for real-time communication and supports stateless operation, making it suitable for serverless deployments.