Core Concepts

Understanding the fundamental concepts behind MCP Forest and the Model Context Protocol

What is the Model Context Protocol (MCP)?

The Model Context Protocol (MCP) is an open standard that enables AI models to securely access external tools and data sources. It provides a standardized way for AI applications to interact with your services while maintaining security and control.

Think of MCP as a universal translator between AI models and your APIs. Instead of AI models needing to understand the specific details of every API they want to use, MCP provides a common language that both AI models and services can understand.

Without MCP

  • • Manual API integration for each service
  • • Custom code for every AI application
  • • Inconsistent security implementations
  • • Difficult to maintain and scale

With MCP

  • • Standardized protocol for all integrations
  • • Automatic AI model compatibility
  • • Built-in security and authentication
  • • Easy to deploy and maintain

MCP Architecture

AI Application

Claude Desktop, Cursor IDE, or custom applications

MCP Server

Generated by MCP Forest, handles protocol communication

Your API

WordPress, WooCommerce, or any REST API

AI Request
MCP Protocol
API Call

MCP Servers

An MCP server is a program that implements the Model Context Protocol to expose tools and resources to AI applications. MCP Forest automatically generates these servers for you.

Server Components

  • • Tools: Functions that AI can call
  • • Resources: Data sources AI can access
  • • Prompts: Pre-defined prompt templates
  • • Transport: HTTP or stdio communication

Auto-Generated Features

  • • Complete Node.js/Express application
  • • Docker containerization
  • • Authentication and security
  • • Error handling and logging

Tools

Tools are the core functionality that AI models can invoke through your MCP server. Each tool corresponds to a specific API endpoint or operation.

Example: WordPress Tools

get_postsGET /wp-json/wp/v2/posts

Retrieve WordPress posts with filtering options

Parameters: limit, status, author, search, date_range
create_postPOST /wp-json/wp/v2/posts

Create new WordPress posts

Parameters: title, content, status, author, categories, tags
update_postPUT /wp-json/wp/v2/posts/{id}

Update existing WordPress posts

Parameters: id, title, content, status, categories, tags

Transport Methods

MCP supports two transport methods for communication between AI applications and servers:

HTTP Transport

  • • Standard HTTP/HTTPS requests
  • • Works with any HTTP client
  • • Scalable and cacheable
  • • Easy to debug and monitor
https://api.mcpforest.dev/servers/your-server-id

Stdio Transport

  • • Direct process communication
  • • Lower latency
  • • Used by Claude Desktop
  • • Process-to-process messaging
node server.js

Security Model

Security First

MCP Forest implements multiple layers of security to protect your data and ensure only authorized access to your APIs.

Authentication

  • • API key authentication
  • • OAuth 2.0 support
  • • Basic authentication
  • • Custom authentication methods

Authorization

  • • Role-based access control
  • • Scope-limited permissions
  • • Rate limiting
  • • Audit logging

Why Use MCP Forest?

Automatic Generation

No need to write MCP server code manually. MCP Forest analyzes your API and generates a complete, production-ready server.

Production Ready

Generated servers include Docker containers, health checks, monitoring, and all the infrastructure needed for production deployment.

Platform Agnostic

Works with any AI application that supports MCP, including Claude Desktop, Cursor IDE, and custom applications.

Ready to Build Your First MCP Server?

Now that you understand the core concepts, let's put them into practice. Follow our quick start guide to create your first MCP server.