MCP client
An MCP client is the part of an AI application that holds one connection to one MCP server, negotiating capabilities and carrying requests, results and notifications between that server and the host app.
Also called: MCP host
By Divya Sharma · Checked Nov 15, 2026 · How we review
The Model Context Protocol has three roles. The host is the application the user runs, such as Claude Code, Cursor or a chat app. The server is the program that offers tools and data. The client sits between them: the host creates one client for each server it connects to.
What a client does
When it connects, the client opens a session with its server, the two declare which features they support, and from then on the client passes messages both ways. It lists and calls the server's tools, reads its resources, fetches its prompts and receives notifications when anything changes. It also keeps each server isolated, so one server cannot see another's traffic or the full conversation.
Client versus host
People often say "MCP client" for the whole app, and this page covers both names. Strictly, the host is the app that owns the model, the user interface and permission prompts, and it may run many clients at once, one per MCP server. When you add five servers to an app's config, the host starts five clients.
Why it matters
Which client you use decides what MCP features you get. Some clients support only tools, some also show resources and prompts, and support for remote servers and the transports they use varies. Check the client's documentation before choosing a server from the MCP tools directory.
Related terms
- Model Context Protocol
- The Model Context Protocol (MCP) is an open standard that defines how AI applications connect to external tools, data and prompts, using JSON-RPC messages between a host application, its clients and separate servers.
- MCP server
- An MCP server is a program that exposes tools, data resources or prompt templates to AI applications through the Model Context Protocol, so any MCP-compatible client can use them without custom integration code.
- MCP transport (stdio, Streamable HTTP)
- An MCP transport is the channel that carries MCP's JSON-RPC messages between client and server. The specification defines two: stdio, for a server run as a local subprocess, and Streamable HTTP, for a server reached over the network.
- MCP resources and prompts
- MCP resources and prompts are two things an MCP server can offer besides tools. Resources are pieces of data, each with a URI, that an app can add to context; prompts are reusable message templates a user chooses to run.
- Tool calling
- Tool calling, also called function calling, is a model API feature where you describe functions with a schema, the model replies with a structured request to call one, and your code runs it and returns the result.