MCP Client
An MCP client is the component within an AI agent or agent framework that initiates connections to MCP servers, sends tools/list requests to discover available tools, and sends tools/call requests to invoke them.
Why this matters for AI agents
The MCP client is where the agent expresses its intent to invoke a tool. Policy enforcement at the client-server boundary is the natural enforcement point.
How AgntID relates
AgntID's embedded proxy intercepts MCP client calls before they reach the server, applying policy evaluation at the point of invocation without requiring changes to the client or server.
The static IAM gap
MCP clients authenticate to servers but do not enforce per-tool access policy. They send whatever tool calls the agent requests.
Related phrases
Related terms
Frequently asked questions
What is an MCP client?
An MCP client is the component within an AI agent or agent framework that initiates connections to MCP servers, sends tools/list requests to discover available tools, and sends tools/call requests to invoke them.
Why is the MCP client a natural enforcement point?
The MCP client is where the agent expresses its intent to invoke a tool. Policy enforcement at the client-server boundary intercepts calls before they reach the backend.
How does AgntID work with the MCP client?
AgntID's embedded proxy intercepts MCP client calls before they reach the server, applying policy evaluation at the point of invocation without requiring changes to the MCP client or server.
Secure every agent tool call at execution time.
AgntID gives infrastructure teams scoped, ephemeral access control for AI agents without replacing IAM, MCP servers, tools, or agent frameworks.