Cloud MCP builds upon key MCP concepts and introduces enterprise-level management capabilities. Understanding these foundational concepts will help you effectively use the platform.

​
Enterprise Architecture Components

​
Organizations

Organizations are the top-level containers in Cloud MCP that group users, resources, and configurations together. Think of them as your company’s workspace where:
  • Users are managed with specific roles and permissions
  • MCP Servers are deployed and configured
  • Config Profiles are created for different use cases
  • Access Control is enforced across all resources
Organizations enable multi-tenant setups where different teams or departments can have isolated environments while sharing the same platform.

​
MCP Servers

MCP Servers are long-running services that expose tools, data, and workflows to AI clients. In Cloud MCP, these servers are:
  • Hosted and Managed: No need to deploy or maintain infrastructure
  • Scalable: Automatically handle traffic spikes and load
  • Secure: Built-in authentication and authorization
  • Monitored: Real-time health checks and logging
Common MCP server types include:
  • Database Connectors: Access to PostgreSQL, MongoDB, etc.
  • API Integrations: Stripe, GitHub, Slack, etc.
  • File Systems: Local and cloud storage access
  • Custom Tools: Your own specialized workflows

​
Config Profiles

Config Profiles are collections of MCP server configurations that define what tools and data are available to specific users or teams. They serve as:
  • Access Control: Determine which servers a user can connect to
  • Client Configuration: Provide ready-to-use settings for AI clients
  • Environment Management: Separate dev, staging, and production setups
Each profile contains connection details, authentication credentials, and server-specific parameters.

​
Clients

AI applications that connect to MCP servers to access external tools and data. Popular clients include:
  • Claude Desktop: Anthropic’s AI assistant
  • Cursor: AI-powered code editor
  • VS Code: With MCP extensions
  • Custom Applications: Your own AI integrations

​
Enterprise Integration Architecture

​
1. Organization Setup

Organization: Acme Corp
├── Users (developers, admins, analysts)
├── MCP Servers (database, APIs, tools)
├── Config Profiles (dev, staging, prod)
└── Access Policies

​
2. Server Deployment

When you deploy an MCP server in Cloud MCP:
  1. Provision: Infrastructure is automatically created
  2. Configure: Server parameters are set via the dashboard
  3. Secure: Authentication and encryption are applied
  4. Monitor: Health checks and logging begin

​
3. Client Connection

AI clients connect using config profiles: 
{
  "mcpServers": {
    "cloudmcp-stripe-01": {
      "command": "npx",
      "args": [
        "-y",
        "@cloudmcp/connect",
        "--url",
        "https://gateway.cloudmcp.dev/fa2e4261/sse",
        "--clientId",
        "abc123",
        "--clientSecret",
        "xyz789"
      ]
    }
  }
}

​
4. Tool Access

Once connected, AI clients can:
  • Query Databases: “Show me recent orders from the database”
  • Call APIs: “Create a new customer in Stripe”
  • Access Files: “Read the latest sales report”
  • Execute Workflows: “Run the monthly analytics pipeline”

​
Strategic Business Value

​
Centralized Management

  • Single Dashboard: Manage all MCP servers from one interface
  • User Management: Control who can access what resources
  • Audit Logs: Track all activities and changes
  • Compliance: Built-in security and governance features

​
Operational Efficiency

  • No Infrastructure: Cloud-hosted servers with automatic scaling
  • Quick Setup: Deploy servers in minutes, not days
  • Monitoring: Built-in health checks and alerting
  • Updates: Automatic security patches and feature updates

​
Developer Experience

  • Instant Access: Connect AI clients with provided configurations
  • Self-Service: Developers can deploy servers without IT involvement
  • Standardization: Consistent patterns across all integrations
  • Documentation: Auto-generated API docs and examples
Learn More: Explore the Model Context Protocol specification to understand the underlying technology that powers Cloud MCP.

​
Implementation Roadmap

Now that you understand the core concepts, you can:
  1. Sign up for Cloud MCP to create your first organization
  2. Deploy your first MCP server to see it in action
  3. Configure AI clients to start using your servers