Back to Technology

Model Context Protocol

AI-Native
Tool Architecture

YachtOS implements Anthropic's Model Context Protocol (MCP), the universal standard for AI tool integration. Our AI agents can discover and execute 60+ specialized tools in real-time.

How MCP WorksView All Tools

What is MCP?

Model Context Protocol is an open standard by Anthropic that enables AI models to discover and use external tools dynamically - without hardcoded integrations.

Universal Protocol

Works with Claude, ChatGPT, Gemini, and any future AI model. One integration, all platforms.

Dynamic Discovery

AI agents discover available tools at runtime. Add new capabilities without retraining.

Secure Execution

Tools run in sandboxed environments with strict permissions. User authorization required for sensitive actions.

How MCP Works in YachtOS

1

User Request

"Check the weather for our passage to Monaco"

2

Tool Discovery

AI discovers weather_check tool via MCP protocol

3

Tool Execution

MCP server executes tool with user context

4

Response

AI synthesizes data into natural response

60+ AI Tools

Specialized tools for every aspect of yacht operations

weather_check()

Real-time marine weather forecasts

vessel_tracker()

AIS vessel tracking and identification

find_suppliers()

Locate nearby marine suppliers

create_work_order()

Generate maintenance tickets

schedule_service()

Book service appointments

find_marina()

Search marina availability

berth_booking()

Reserve marina berths

crew_roster()

Manage crew schedules

provision_order()

Order supplies and provisions

fuel_status()

Check fuel levels and pricing

maintenance_log()

Access maintenance history

document_search()

Search vessel documents

+ 48 more tools for provisioning, crew management, compliance, and more

Why MCP Matters

Future-Proof

As new AI models emerge, YachtOS tools work immediately. No re-integration required.

Extensible

Third-party developers can create MCP tools that integrate cleanly with YachtOS AI agents.

Open Standard

MCP is open-source and community-driven. Join the ecosystem of AI-native applications.

Intelligent Routing

AI chooses the right tool for each task. Complex workflows executed autonomously.

Technical Implementation

MCP Server

  • JSON-RPC over HTTP/SSE
  • Tool schema auto-discovery
  • Session-based authentication
  • Rate limiting and quotas

Tool Capabilities

  • Typed parameters with validation
  • Async execution support
  • Error handling and retries
  • Audit logging

Experience MCP-Powered AI

Try the Command Center with full MCP tool integration

Try Command CenterBack to Technology