Skip to main content

What is MCPcat?

MCPcat is an analytics platform that helps you understand how users interact with your MCP servers. With just one line of code, you can:
  • ๐ŸŽฌ Replay user sessions - See exactly how users interact with your tools
  • ๐Ÿ” Debug traces - Track down issues and understand LLM confusion points
  • ๐Ÿ“Š Analyze usage patterns - Discover what features users need most

Installation

1

Install the MCPcat Agent

Run the following command to install the open source MCPcat SDK:
npm i mcpcat
2

Track your MCP server

import * as mcpcat from "mcpcat";

/* 
Example MCP server
const mcpServer = new Server({ 
  name: "my-mcp-server", 
  version: "1.0.0" 
});
mcpServer.tool("my-tool", ...)
*/


// Call .track() AFTER tool call registering
mcpcat.track(mcpServer, "proj_YOUR_PROJECT_ID")
You can grab your project ID from MCPcatโ€™s dashboard.

Using Without an MCPcat Account

You can forward events directly to your observability platform:
mcpcat.track(server, null, {
  exporters: {
    otlp: {
      type: "otlp",
      endpoint: "http://localhost:4318/v1/traces"
    }
  }
})
Learn more about integration options โ†’