What Is an MCP Server? A Beginner's Guide

AI assistants are smart, but on their own they can't see your files, check your calendar, or look things up in your tools. MCP fixes that. This guide explains what MCP is, what an MCP server does, and how to start using one with Claude and other apps, with no jargon needed.

What is MCP?

MCP stands for Model Context Protocol. It's an open standard that lets AI applications connect to outside tools and data in one common way.

Think of it like a USB port for AI. Before USB, every device needed its own cable. With MCP, any AI app that "speaks MCP" can plug into any tool that "speaks MCP," so each new connection doesn't need its own custom code.

What is an MCP server?

An MCP server is a small program that gives an AI app access to one specific thing, like your files, a database, GitHub, or a search engine.

The AI app asks the server, "What can you do?" The server answers with a list. Then the AI can use those abilities when you ask it to. For example, a GitHub MCP server might let Claude list open pull requests or create an issue for you.

An "AI MCP server" and a "Claude MCP server" are the same thing: an MCP server. Because MCP is an open standard, one server can work with many different AI apps.

How does an MCP server work?

There are three parts:

  • Host: the AI app you use, such as Claude Desktop, Claude Code, or VS Code.
  • Client: a connector inside the host. The host makes one client for each server it connects to.
  • Server: the program that provides the tools or data.

A server can offer three kinds of things:

  • Tools: actions the AI can take, like "search the web," "run a database query," or "create a file."
  • Resources: information the AI can read, like a file's contents or a database schema.
  • Prompts: ready-made templates that help you ask for common tasks.

Local vs remote servers

MCP servers talk to apps in one of two ways:

  • Local (stdio): the server runs on your own computer, and the app talks to it directly. This is good for things like your files. A local server usually serves just one app.
  • Remote (Streamable HTTP): the server runs online at a web address, often ending in /mcp. It usually needs a sign-in (OAuth) or an API key, and it can serve many users at once.

Examples of MCP servers

Here are a few kinds of MCP servers people use:

  • Filesystem: lets the AI read, create, and organize files in folders you choose.
  • GitHub: review pull requests, open issues, and check code.
  • Databases: ask plain-English questions about your data (ideally with a read-only login).
  • Work tools: issue trackers, error monitoring, docs, and design tools, many of which now offer an official remote MCP server.

How to use an MCP server with Claude

Claude Desktop

  1. Open Claude Desktop, go to Settings → Developer, and click Edit Config.
  2. This opens claude_desktop_config.json. Add your server, for example the official filesystem server:
{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/Users/you/Desktop"]
    }
  }
}
  1. Quit and restart Claude Desktop. Your server will appear under Connectors.

Claude asks for your approval before it uses a tool, so you stay in control.

Claude Code

Claude Code uses the claude mcp add command.

For a remote server:

claude mcp add --transport http notion https://mcp.notion.com/mcp

For a local server (everything after -- is the command that starts it):

claude mcp add --transport stdio myserver -- npx -y some-mcp-server

Then run claude mcp list to check it's connected, or type /mcp inside Claude Code to sign in to servers that need it.

Other MCP clients

MCP isn't just for Claude. Other apps that can connect to MCP servers include:

  • Cursor: add servers in its MCP settings.
  • VS Code: connects to MCP servers for its AI features.
  • ChatGPT: can connect to remote MCP servers through developer mode, but this depends on your plan, and it can't connect to a server running only on your computer.

Setup steps vary by app, so check each server's instructions for your client.

How to choose a good MCP server

Not every server is equally good or equally safe. Before you connect one, check:

  • Is it online? A remote server that's often down will break your workflow.
  • Is it maintained? Look for recent updates and an active repository or publisher.
  • How does it sign in? OAuth or a scoped API key is better than pasting powerful passwords.
  • What permissions does it need? Give it the smallest access possible, for example read-only database users or only the folders it needs.
  • Do you trust it? Servers that fetch outside content can expose you to prompt-injection tricks, so stick to publishers you trust and review actions before approving them.

Find MCP servers on MCP Lantern

MCP Lantern is a directory of MCP servers. It runs regular automated status checks (every few hours), so you can see whether a server is up before you connect it. Server owners can list their servers for free, and AI agents can search the directory through MCP Lantern's own MCP endpoint.

If you used Smithery's hosted URLs, note that Smithery's hosted proxy shuts down on October 31, 2026, so it's worth finding each server's direct URL.

Browse MCP servers on MCP Lantern →

FAQ

What does MCP stand for?
Model Context Protocol, an open standard for connecting AI apps to outside tools and data.

Is an MCP server only for Claude?
No. Claude Desktop, Claude Code, Cursor, VS Code, and ChatGPT (in developer mode on some plans) can all use MCP servers.

Do I need to code to use an MCP server?
Not usually. Most need one command or a short block of settings. Building your own does take code.

Are MCP servers safe?
They can be. Connect only servers you trust, give them the fewest permissions they need, and review actions before approving.

What's the difference between a local and a remote MCP server?
A local server runs on your computer over stdio. A remote server runs online over Streamable HTTP, usually with a sign-in or API key.

Published 2026-10-21