MCPcat
MCPcat is an analytics platform for MCP server owners π±.
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 π±.
MCPcat is an analytics platform for MCP server owners π±.
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 π±.
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mcpcat-python-sdkConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MCPcat
MCPcat is an analytics and observability SDK for MCP server developers that instruments any Python-based MCP server with session tracking, user identification, and telemetry export β without modifying the server's core logic. By calling a single `mcpcat.track()` function, server owners gain access to session replays, request traces, and integrations with platforms like Datadog and Sentry. It is designed to help teams understand how AI agents and users interact with their MCP tools in production.
Prerequisites
- Python 3.10 or higher installed
- An existing Python MCP server built with FastMCP or the standard MCP SDK
- A MCPcat project ID from an account at mcpcat.io
- Optional: Datadog, Sentry, or OpenTelemetry credentials for exporting telemetry
Install the MCPcat SDK
Install the core package via pip. Add the [community] extra if you use community MCP frameworks that require additional adapters.
pip install mcpcat
# or with community framework support:
pip install "mcpcat[community]"Create a project at mcpcat.io
Sign up or log in at mcpcat.io and create a new project. Copy the project ID (format: proj_xxxxxxxxx) β you will pass it to the track() call.
Instrument your MCP server
Import mcpcat and call mcpcat.track() with your FastMCP server instance and the project ID. This single call wraps all tool invocations with analytics middleware.
import mcpcat
from mcp.server import FastMCP
server = FastMCP(name="my-mcp-server", version="1.0.0")
mcpcat.track(server, "proj_0000000")Add user identification (optional)
Provide a callback to identify users per session so analytics are grouped by user. The callback receives the session context and returns a dict with user metadata.
from mcpcat import MCPCatOptions
def identify_user(session):
return {"user_id": session.get("user"), "plan": "pro"}
mcpcat.track(server, "proj_0000000", MCPCatOptions(identify=identify_user))Configure data redaction (optional)
Add a redaction callback to scrub sensitive data from telemetry before it is transmitted to MCPcat servers. The callback receives the raw payload and returns a sanitized version.
def redact_sync(data):
if "password" in data:
data["password"] = "[REDACTED]"
return data
mcpcat.track(server, "proj_0000000", redact_sensitive_information=redact_sync)Run your instrumented server
Start your MCP server as normal. MCPcat will automatically capture tool calls, session data, and errors and forward them to your project dashboard.
python your_mcp_server.pyMCPcat Examples
Minimal instrumented server
A complete minimal example of a FastMCP server with MCPcat tracking enabled.
{
"note": "MCPcat is a Python SDK added to your server source β it does not run as a standalone MCP server. Configure your actual MCP server in mcpServers and add mcpcat.track() in its code.",
"mcpServers": {
"my-server": {
"command": "python",
"args": ["/path/to/your_mcp_server.py"]
}
}
}Prompts to try
MCPcat is a server-side SDK, not an end-user MCP server with chat tools. Use the mcpcat.io dashboard to view analytics. These are developer-facing workflows.
- "Show me the session replay for user alice from yesterday"
- "Which of my MCP tools has the highest error rate this week?"
- "How many unique users called the search tool in the last 7 days?"
- "Export traces to Datadog for the last 100 tool invocations"Troubleshooting MCPcat
mcpcat.track() raises ImportError or AttributeError on the server object
Ensure you are passing a FastMCP or compatible MCP server instance β not a plain Python object. Upgrade to the latest mcpcat release with `pip install --upgrade mcpcat` and verify Python is 3.10+.
Analytics dashboard shows no data after instrumenting the server
Verify the project ID string exactly matches the one shown in your mcpcat.io dashboard (format: proj_xxxxxxxxx). Also confirm your server process has outbound internet access to send telemetry.
Sensitive data appears in MCPcat session replays
Implement and pass a `redact_sensitive_information` callback to mcpcat.track(). The callback receives each payload before transmission and should return a sanitized copy.
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 π±. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MCPcat?
Follow the installation instructions on the MCPcat GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
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.
MCPcat Alternatives β Similar Analytics Servers
Looking for alternatives to MCPcat? Here are other popular analytics servers you can use with Claude, Cursor, and VS Code.
OpenMetadata
β 14.0kOpenMetadata is a unified metadata platform for data discovery, data observability, and data governance powered by a central metadata repository, in-depth column level lineage, and seamless team collaboration.
Superset
β 10.9kAn MCP server that provides AI assistants with full access to Apache Superset instances, enabling interaction with dashboards, charts, datasets, databases, and SQL execution capabilities.
Horizon
β 4.4kπ‘ Your own AI-powered news radar. Generates daily briefings in English & Chinese. | η¨ AI ζε»Ίδ½ δΈε±ηζ°ι»ι·θΎΎ
MCP Server Chart
β 4.1kEnables generation of 25+ types of charts and data visualizations using AntV, including bar charts, line charts, maps, mind maps, and specialized diagrams like fishbone and sankey charts. Supports both statistical charts and geographic visualizations
Muapi CLI
β 997Official CLI for muapi.ai β generate images, videos & audio from the terminal. MCP server, 14 AI models, npm + pip installable.
Weather MCP Server
β 907Weather Data Fetcher MCP server built with Node.js, MCP SDK, and Zod. Provides weather details like temperature and forecast for cities such as Noida and Delhi via a registered tool. Simplifies API integration, enabling structured responses for clien
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.
Set Up MCPcat in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
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.