MCP Server
An MCP server is a service that exposes tools, resources, and capabilities to AI agents via the Model Context Protocol, responding to tools/list and tools/call requests.
Why this matters for AI agents
MCP servers expose real capabilities—read access to databases, write access to file systems, execution of scripts. Access to an MCP server does not mean every agent should invoke every tool the server exposes.
How AgntID relates
AgntID sits as a proxy in front of MCP servers, evaluating access policy before allowing tool calls to reach the underlying server—no server modification required.
The static IAM gap
Access to an MCP server is typically controlled at the authentication level. Once a client is authenticated, all tools on that server are available with no per-tool authorization.
Related phrases
Related terms
Frequently asked questions
What is an MCP server?
An MCP server is a service that exposes tools, resources, and capabilities to AI agents via the Model Context Protocol, responding to tools/list and tools/call requests.
Why is authenticating to an MCP server not the same as authorizing tool calls?
Authentication grants access to the server. It does not decide whether a specific agent should invoke a specific tool—every authenticated client can call every tool the server exposes.
How does AgntID protect MCP servers?
AgntID can sit as a proxy in front of MCP servers, evaluating access policy before allowing tool calls to reach the underlying server. The MCP server never needs to be modified.
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.