Skip to content
Agentmart
Submit a tool

What is MCP client?

An MCP client is a connector created by a host application that communicates with exactly one MCP server, attaching protocol version and capabilities to every request.

Also written: Model Context Protocol client

Explained

The MCP specification separates the host from the client, and the two words are easy to mix up. The host is the LLM application. It creates and manages one or more client instances, controls their permissions and lifecycle, enforces consent requirements, and coordinates the connection to the language model. A client is the per-server connector the host creates.

Each client talks to exactly one server, so a host that uses three servers runs three clients. According to the architecture page, a client attaches the protocol version and its capabilities to every request and maintains security boundaries between servers.

Clients declare which optional features they support. The features covered in this glossary are elicitation, sampling and roots, and a client lists each one it handles in the capabilities it sends. Servers must not rely on a feature the client has not declared. When a server needs input from the client side, the client gets an InputRequiredResult, forwards the work to the user or the model through the host, and retries the original request with the response.

Primary source

Architecture (modelcontextprotocol.io), read Oct 26, 2026

This entry paraphrases the page above. Sources change; check it for the current wording.

Listings that use this

The 5 top-ranked listings tagged for MCP client. Rank weighs adoption, freshness, documentation and publisher identity; how we rank and grade. Money never changes the order.

Cite this page

What is MCP client?. Agent Mart. https://agentmart.dev/glossary/mcp-client (last updated 2026-10-26)

Written Oct 26, 2026, updated Oct 26, 2026. All glossary terms