Free explainer

What Is MCP? The Model Context Protocol in Plain English

MCP (Model Context Protocol) is an open standard that lets AI apps such as Claude, ChatGPT, Cursor and VS Code connect to outside tools and data. Think of it as a USB-C port for AI: any app that speaks MCP can plug into any MCP server, whether that server reads your files, queries a database or posts to Slack.

Below, step through what actually happens when an AI uses MCP, check whether your own task needs it, and browse 31 real MCP servers. No sign-up; everything runs in your browser.

How MCP works, step by step

Pick a task, then step through what happens between you, the AI app and the MCP server.

  1. Youask in plain English
  2. Host (AI app)Claude, ChatGPT, Cursor, VS Code
  3. MCP clientone per server, inside the host
  4. MCP serverFilesystem: offers tools, resources, prompts
  5. Real systemfiles, database, Slack

Step 1 of 7

You ask the AI app

You type: "Summarize notes.md in my Projects folder." The AI app (the host) cannot do this alone. It needs a tool.

Transport in this example: stdio (the server runs as a local program on your computer).

The three parts of MCP

Host

The AI app you use. It runs the model, shows you results and decides which tool calls need your approval.

Client

A connector inside the host. Each client keeps one connection to one server and passes requests back and forth as JSON-RPC messages.

Server

A program that offers tools (actions), resources (data to read) and prompts (templates). It runs locally over stdio or remotely over Streamable HTTP.

MCP or not? Check your task

Choose a common task or answer the four questions yourself.

Who decides when the action happens?
Where do the data or app live?
Does your AI app already have a built-in connector for it?
How many AI apps should use it?

Verdict

Use MCP: a remote server (Streamable HTTP)

The AI needs to choose when to read or change data in a cloud service. Many vendors host an official remote MCP server; you add its URL and sign in with OAuth, and every MCP-capable AI app can use the same server.

How to set it up: Add the server URL as a custom connector or in your MCP config, then sign in when the app opens the OAuth window.

Example: HubSpot MCP server

Official or reference servers you can add today. Local servers run on your computer; remote servers are hosted by the vendor and you sign in with OAuth.

31 servers

  • Read, write and search files in folders you allow

    Reference

  • GitLocal

    Read history, diffs and branches of a local repository

    Reference

  • FetchLocal

    Fetch a web page and convert it to Markdown

    Reference

  • MemoryLocal

    Keep a knowledge graph the AI can recall later

    Reference

  • TimeLocal

    Current time and time zone conversion

    Reference

  • Step-by-step reasoning scratchpad

    Reference

  • GitHubLocal or remote

    Issues, pull requests, code search and Actions

    Developer

  • Drive a real browser: click, type, read pages

    Developer

  • Inspect, debug and profile pages in Chrome

    Developer

  • SentryRemote

    Look up errors, stack traces and releases

    Developer

  • Context7Local or remote

    Current library docs pulled into the prompt

    Developer

  • LinearRemote

    Create, find and update issues and projects

    Work & docs

  • NotionRemote

    Search, read and edit pages and databases

    Work & docs

  • AtlassianRemote

    Jira issues and Confluence pages

    Work & docs

  • SlackRemote

    Search channels, read threads, send messages

    Work & docs

  • AsanaRemote

    Tasks, projects and goals

    Work & docs

  • BoxRemote

    Search and read files stored in Box

    Work & docs

  • SupabaseRemote

    Query tables, manage schema and projects

    Data

  • NeonRemote

    Serverless Postgres: run SQL, branch databases

    Data

  • MongoDBLocal

    Query collections and manage Atlas clusters

    Data

  • Google's server for Postgres, MySQL, BigQuery and more

    Data

  • StripeRemote

    Customers, payments, refunds and docs search

    Business apps

  • PayPalRemote

    Invoices, orders and transactions

    Business apps

  • SquareRemote

    Payments, catalog, orders and customers

    Business apps

  • HubSpotRemote

    Contacts, companies and deals in your CRM

    Business apps

  • ZapierRemote

    Actions in thousands of connected apps

    Business apps

  • Web, news and image search

    Web & research

  • ExaLocal or remote

    AI-oriented web search and page contents

    Web & research

  • Search official Microsoft and Azure docs

    Web & research

  • Workers, DNS, logs and docs for your account

    Cloud & design

  • FigmaLocal or remote

    Design context and code from Figma frames

    Cloud & design

MCP vs API vs function calling vs plugins

MCPDirect APIFunction callingPlugins
What it isAn open protocol that connects AI apps to tools and dataA service's own interface for programsA model feature: the model returns a structured tool callA product-specific add-on system (for example ChatGPT plugins, 2023 to 2024)
Who calls itThe AI app, when the model chooses a toolYour code, on a fixed pathYour code, after the model asks for a callOne AI product
Works across AI appsYes: build one server, use it in Claude, ChatGPT, Cursor, VS CodeEach app needs its own integration codeTool definitions are written per appNo, tied to one product
Tool discoveryBuilt in: tools/list at run timeRead the docsYou pass the tool list in each requestManifest file
AuthOAuth for remote servers; local servers use your machineAPI keys or OAuthWhatever your code usesSet by the product
Best forLetting assistants and agents use many tools safelyScripts, back ends and fixed automationsTools inside one app you are buildingLargely replaced by MCP and app connectors

They work together: an AI app often uses the model's function calling to decide on a tool, MCP to reach it, and the server uses the service's API to do the work.

MCP security checklist

  • Install servers from sources you trust

    A local server runs as a program with your user's rights. Prefer the vendor's official server or the reference servers, and check who publishes a server before you add it.

  • Grant the least access

    Limit folders, use read-only database logins and narrow OAuth scopes. Turn off tools you do not need.

  • Keep approval on for actions

    Let the AI app ask before any tool that sends, deletes, pays or posts. Read the arguments before you click Allow.

  • Watch for prompt injection

    Text a tool returns, such as a web page, email or issue, can contain instructions aimed at the model. Do not let untrusted content trigger tools with write access.

  • Keep secrets out of chats and configs you share

    Put API keys in environment variables or your app's secret store, not in prompts or committed config files.

  • Review and remove

    List your connected servers now and then, revoke OAuth grants you no longer use, and update servers like any other software.

How to start using an MCP server

  1. 1. Pick a server

    Choose an official server for the app you want to connect, such as GitHub, Notion or Slack, from the vendor docs or the MCP Registry.

  2. 2. Add it to your AI app

    For a remote server, paste its URL as a custom connector. For a local server, add its command to the app's MCP config file.

  3. 3. Sign in and limit access

    Complete the OAuth sign-in or set the API key, and grant only the scopes, folders or tools you need.

  4. 4. Ask for something

    Ask the AI to use it in plain English. Approve tool calls that change data after checking the arguments.

MCP glossary

Host
The AI application you use, such as Claude, ChatGPT, Cursor or VS Code. It runs the model and manages connections.
MCP client
The connector inside the host that talks to one MCP server. A host with three servers has three clients.
MCP server
A program that offers tools, resources and prompts to AI apps over MCP. It can run on your computer or on the web.
Tool
An action the model can call, such as search_issues or send_message, with a name, description and input schema.
Resource
Read-only data the server can share, such as a file, a database schema or a document, identified by a URI.
Prompt
A reusable prompt template the server offers, often shown to users as a slash command.
Elicitation
A server asking the user for more input or a confirmation through the host.
Sampling
A server asking the host's model for a completion. Deprecated in spec 2026-07-28.
Transport
How messages travel: stdio for a local program, Streamable HTTP for a remote server.
JSON-RPC 2.0
The message format MCP uses: requests with a method and params, and responses with a result or error.
Connector
What Claude and some other apps call a remote MCP server you add to the app.
MCP Registry
The official, open catalog where publishers list MCP servers so clients can find them.

MCP FAQ

What is MCP in AI?

MCP (Model Context Protocol) is an open standard for connecting AI applications to external tools and data. An AI app such as Claude, ChatGPT, Cursor or VS Code connects to MCP servers, and each server offers tools the model can call, such as reading a file, querying a database or posting to Slack.

Who created MCP?

Anthropic introduced MCP as an open-source protocol on November 25, 2024. On December 9, 2025, Anthropic donated it to the Agentic AI Foundation, a directed fund under the Linux Foundation co-founded by Anthropic, Block and OpenAI.

Is MCP only for Claude?

No. MCP is supported by many AI apps, including ChatGPT, Visual Studio Code, Cursor and Claude, and by agent frameworks. A server built once works in any MCP-capable app.

What is an MCP server?

An MCP server is a program that exposes tools, resources and prompts to AI apps using MCP. Local servers run on your computer and talk over stdio; remote servers run on the web and use Streamable HTTP, usually with OAuth sign-in.

Is MCP safe?

MCP itself is a protocol; safety depends on which servers you connect and what they can do. Use servers from trusted publishers, grant the least access, keep approval on for actions that change things, and be careful with untrusted content that could carry prompt injection.

Do I need to code to use MCP?

No. In most AI apps you add an MCP server by pasting its URL as a connector or by adding a short config entry, then signing in. You only need to code to build your own server.

What is the difference between MCP and an API?

An API is how programs talk to one service. MCP is a standard layer on top that lets an AI app discover a server's tools at run time and call them when the model decides to. Many MCP servers simply wrap a service's API.

Checked 2026-11-14 against the official MCP docs (current spec 2026-07-28) and Anthropic's launch post and donation announcement.

Put your MCP tools to work in repeatable workflows

Tornic turns the Claude Code, Codex or Cursor CLI you already pay for into deterministic multi-step workflows, so your agents can use the MCP servers you connect the same way every run.

Explore Tornic

Related free tools