Intro to MCP

What is MCP?

MCP stands for Model Context Protocol. It is an open protocol that standardises how AI applications connect to external systems. Instead of every AI application building a custom integration for every API, database, or service, an MCP server can expose those functionalities through a common interface.

How does MCP work?

MCP has three main parts:

• MCP host: The AI application you interact with, such as Claude or Cursor.
• MCP client: Manages the connection between the host and an MCP server.
• MCP server: Exposes tools and data from an external system to the AI application.

A server can expose three main things:

• Tools: Actions the model can take, like creating a GitHub pull request.
• Resources: Data the application can read, like a file or database record.
• Prompts: Reusable templates for common tasks.

What does MCP solve?

Think of the time when you had different chargers for different devices. Then USB‑C came along. Now one cable works with your phone, your laptop, and your headphones.

This was the same problem with AI apps. Say you want Claude or Cursor to read your Google Drive. Before MCP, each of those apps would have to connect to Google Drive separately. Now imagine you also need to use the AI with Gmail, GitHub, or Notion - it's already too many cables.

Now with MCP, Google Drive gets one MCP connection, and any AI app that supports MCP can plug into it. Build the connection once, and it works everywhere.