MCPcat

v0.1.16β€’Analyticsβ€’stable

MCPcat is an analytics platform for MCP server owners 🐱.

agent-runtimeagentsaiai-agentsai-platform
Share:
99
Stars
0
Downloads
0
Weekly
0/5

What is MCPcat?

MCPcat is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcpcat is an analytics platform for mcp server owners 🐱.

MCPcat is an analytics platform for MCP server owners 🐱.

This server falls under the Analytics category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • MCPcat is an analytics platform for MCP server owners 🐱.

Use Cases

MCP server analytics and monitoring
Agent runtime insights
Debugging and performance tracking
MCPCat

Maintainer

LicenseMIT
Languagetypescript
Versionv0.1.16
UpdatedMay 20, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y mcpcat

Manual Installation

npx -y mcpcat

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use MCPcat

MCPcat is an analytics SDK for MCP server developers that wraps any existing MCP server to capture tool call telemetry, user behavior patterns, session replays, and trace-level debugging data. Rather than being a standalone MCP server itself, it is a TypeScript library you integrate into your own server with a single function call β€” mcpcat.track(server, projectId) β€” giving you immediate visibility into how agents and users are interacting with your tools. It supports exporting telemetry to Datadog, Sentry, PostHog, and OTLP-compatible backends.

Prerequisites

  • Node.js 18+ and npm installed
  • An existing TypeScript MCP server project to instrument
  • A MCPcat account and project ID (obtain at mcpcat.dev)
  • Optional: API keys for Datadog, Sentry, PostHog, or an OTLP endpoint for telemetry export
1

Add the mcpcat package to your MCP server project

Install the MCPcat TypeScript SDK as a dependency in your existing MCP server project.

npm install -S mcpcat
2

Import MCPcat and call track() on your server instance

In your MCP server entry point, import mcpcat and call track() with your server instance and your MCPcat project ID. This single call enables all telemetry collection.

import * as mcpcat from "mcpcat";

const mcpServer = new Server({ name: "my-mcp-server", version: "1.0.0" });
mcpcat.track(mcpServer, "proj_0000000");
3

Optionally add user identification

Enrich telemetry with user context by providing an identify callback. This lets you correlate tool calls with specific users in the MCPcat dashboard.

mcpcat.track(mcpServer, "proj_0000000", {
  identify: async (request, extra) => {
    const user = await myapi.getUser(request.params.arguments.token);
    return {
      userId: user.id,
      userName: user.name,
    };
  },
});
4

Configure sensitive data redaction

If your tool calls handle PII or credentials, add a redactSensitiveInformation callback to scrub data before it is sent to MCPcat.

mcpcat.track(mcpServer, "proj_0000000", {
  redactSensitiveInformation: async (text) => await redact(text),
});
5

Deploy your instrumented server and view analytics

Deploy your MCP server as normal. MCPcat automatically collects tool call data, session flows, and performance metrics. Visit the MCPcat dashboard to view real-time analytics, session replays, and debug traces.

MCPcat Examples

Client configuration

Example Claude Desktop config for running an mcpcat-instrumented server built from source.

{
  "mcpServers": {
    "my-instrumented-server": {
      "command": "node",
      "args": ["/path/to/my-mcp-server/dist/index.js"]
    }
  }
}

Minimal integration code

The minimal TypeScript code needed to add MCPcat tracking to any MCP server.

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import * as mcpcat from "mcpcat";

const server = new Server(
  { name: "echo-mcp", version: "0.1.0" },
  { capabilities: { tools: {} } }
);

// One line to enable full analytics
mcpcat.track(server, "proj_0000000");

// ... rest of your tool definitions

Troubleshooting MCPcat

track() call does not appear to send any data to MCPcat

Verify your project ID is correct (format: proj_XXXXXXX) and that your server has outbound internet access to reach the MCPcat collection endpoint. Check that mcpcat.track() is called before any tool handlers are registered.

TypeScript compilation errors after importing mcpcat

Ensure your tsconfig.json targets ES2020 or later and that moduleResolution is set to 'node16' or 'bundler'. Run 'npm install -S mcpcat' again to confirm the package is correctly installed with its type definitions.

Sensitive data appearing in MCPcat dashboard

Add a redactSensitiveInformation callback to the track() options that scrubs credentials, tokens, and PII from tool call arguments before telemetry is exported. Test the redaction function independently before deploying.

Frequently Asked Questions about MCPcat

What is MCPcat?

MCPcat is a Model Context Protocol (MCP) server that mcpcat is an analytics platform for mcp server owners 🐱. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install MCPcat?

Install via npm with the command: npx -y mcpcat. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with MCPcat?

MCPcat works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is MCPcat free to use?

Yes, MCPcat is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Analytics MCP Servers

Explore all analytics servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "mcpcat-typescript-sdk": { "command": "npx", "args": ["-y", "mcpcat"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide β†’

Ready to use MCPcat?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides