What is the Model Context Protocol, and what problem does it solve?

beginner

← All Model Context Protocol (MCP) questions

Re-explain

The Model Context Protocol (MCP) is an open standard that lets AI applications connect to outside tools and data through one common interface. Anthropic released it in November 2024.

The problem it solves is integration cost. Before MCP, every AI app wrote custom glue for every system: one connector for GitHub, another for Postgres, another for Slack. With M apps and N systems, you maintain M times N connectors.

MCP turns that into M plus N. You expose a system once as an MCP server, and any MCP-capable app can use it. The official analogy is a USB-C port for AI applications: one connector, many devices.

The real payoff is portability. Integrations move between hosts, and you add one without rewriting the app.

Rewriting in plainer words…

This answer doesn't lend itself to a diagram - it reads best . No credits were charged.

Why there's no diagram: “”

The interactive diagram is below the answer - jump to diagram ↓ · Below it, the related concept . Jump to it ↓

The diagram below the answer is the concept . Jump to it ↓

Tailored explanation · switch back to · ·

What should the new diagram focus on?

Guided practice for Model Context Protocol (MCP) One question at a time. Answer out loud, get graded, see what you missed.

How well did you know this?

AI:

Saved in this browser - sign in to keep your review list.

How should your speech become text?

Listening… your words appear above as you speak - tap Stop when you're done.

Recording · cr - tap Stop & transcribe when you're done.

Transcribing with AI…

Voice:

Keep going - a few more words and AI can grade it.

Interview lens How interviewers actually use this question

Likely follow-ups

Walk me through MCP's architecture. What are hosts, clients and servers? The host is the AI app. It runs one client per server, each on a 1:1 connection, and each server wraps an outside system.
What can an MCP server expose, and who decides when each kind gets used? Name tools, resources and prompts, and who controls each one: the model, the application or the user.
How is MCP different from just calling a REST API? MCP gives a model a shared format and runtime discovery. A REST API needs custom client code per service. MCP servers often wrap REST APIs anyway.
When would you skip MCP and use plain function calling in your own app? For one app with a few in-house tools, function calling is simpler. MCP pays off when tools get reused across hosts or shared with other teams.
What new security risks come with plugging third-party MCP servers into a host? Tool descriptions and outputs from each server enter the model's context untrusted. Mention prompt injection, tool poisoning and user consent before calls.

What you can say

MCP, the Model Context Protocol, is an open standard that lets AI apps connect to outside tools and data through one common interface.
Anthropic released it in November 2024, and the problem it goes after is integration cost.
Before MCP, every AI app wrote its own connector for every system, so M apps and N systems meant M times N connectors to maintain.
With MCP you expose a system once as an MCP server, any MCP-capable app can use it, and the work drops to M plus N.
The official analogy is a USB-C port for AI apps, and the payoff is portability, since you can add an integration or move it to another host without rewriting the app.
It is still one more protocol layer, though, so for a single app with a couple of in-house tools, I'd reach for plain function calling first.

Weak answers to avoid

Defines MCP but never says what it fixes The question asks what problem it solves. Name the M times N connector problem, then explain how one server per system turns it into M plus N.
Calls MCP a model feature MCP is a protocol between host apps and servers, and it sits outside the model. The model only sees whatever tools and data the host passes along.
Says MCP replaces APIs MCP servers usually wrap existing APIs. MCP sets a standard way for AI apps to find and call them, and the API underneath stays.
Treats MCP as Anthropic-only Anthropic released it, but it's an open standard that any MCP-capable app can use. The main payoff is portability across hosts, so leaving that out undersells MCP.

Interview lens

Likely follow-ups, what you can say, and the weak answers to avoid.

Want a quick review of the fundamentals? See the Model Context Protocol (MCP) cheatsheet.

← Back to all Model Context Protocol (MCP) questions

Pro · $10/mo

49 of 56 Model Context Protocol (MCP) answers are in Pro.

Full answers, code samples, and AI explanations that go simpler or deeper. Cancel anytime.

Full answers + code
AI explanations, simpler or deeper
1,000 AI credits / month
Cancel anytime

Get Pro → See plans & pricing →