MCPgee
MCPgee
MCPgeeMCPgee

MCP Monitor

v1.0.0•Monitoring & Observability•stable

An observability tool for agentic AI pipelines that intercepts MCP and Python tool calls to provide real-time metrics, session replays, and alerts via a local dashboard. It enables centralized monitoring of multiple MCP servers through multiplexer an

computermcpmcp-servermonitormonitoring
Share:
82
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is MCP Monitor?

MCP Monitor is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to observability tool for agentic ai pipelines that intercepts mcp and python tool calls to provide real-time metrics, session replays, and alerts via a local dashboard. it enables centralized monitoring...

An observability tool for agentic AI pipelines that intercepts MCP and Python tool calls to provide real-time metrics, session replays, and alerts via a local dashboard. It enables centralized monitoring of multiple MCP servers through multiplexer an

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

Features

  • An observability tool for agentic AI pipelines that intercep

Use Cases

Agentic AI pipeline observability
Real-time metrics and session replay
Centralized MCP server monitoring
Partha-SUST16

Maintainer

LicenseApache-2.0
Languagego
Versionv1.0.0
UpdatedMay 16, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

Manual Installation

npx mcp-monitor

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 MCP Monitor

MCP Monitor is an observability platform for agentic AI pipelines that intercepts all tool calls flowing between MCP clients and MCP servers, recording them to a local SQLite database and surfacing real-time metrics, session replays, and cost estimates through a local web dashboard on port 4242. It also functions as an MCP multiplexer — acting as a single MCP server that proxies multiple downstream servers — and ships a Python SDK for monitoring non-MCP Python agent frameworks such as QwenAgent. Teams use it to debug flaky agentic workflows, track per-tool latency and error rates, and estimate API token spend without sending data to external services.

Prerequisites

  • Node.js 18 or higher
  • npm package manager
  • Python 3.9 or higher (required only if using the Python agent SDK)
  • An MCP client or agentic pipeline to monitor
  • Access to the GitHub Packages npm registry (for the npm install method)
1

Install mcp-monitor from GitHub Packages

Configure npm to use the GitHub Packages registry for the @partha-sust16 scope, then install the package globally.

echo "@partha-sust16:registry=https://npm.pkg.github.com" >> ~/.npmrc
npm install -g @partha-sust16/mcp-monitor
2

Alternative: build from source

Clone the repository, install dependencies, build the TypeScript source, and link the binary globally.

git clone https://github.com/Partha-SUST16/mcp_monitor.git
cd mcp_monitor
npm install
npm run build
npm link
3

Create a configuration file

Create mcp-monitor.config.json in your working directory. List the MCP servers you want to monitor, set the dashboard port, and configure alerting thresholds and pricing for cost estimation.

{
  "servers": [
    {
      "name": "filesystem",
      "transport": "stdio",
      "command": "npx @modelcontextprotocol/server-filesystem /tmp"
    }
  ],
  "dashboard": { "port": 4242 },
  "alerts": {
    "latencyP95Ms": 2000,
    "errorRatePercent": 10,
    "cooldownMinutes": 5
  },
  "pricing": {
    "charsPerToken": 4,
    "inputPerMillion": 3.0,
    "outputPerMillion": 15.0
  }
}
4

Start the dashboard and multiplexer

Run mcp-monitor start to launch both the web dashboard and the multiplexing MCP server. Open http://localhost:4242 in your browser to view live metrics.

mcp-monitor start -c mcp-monitor.config.json
5

Point your MCP client at the monitor

Update your MCP client configuration to use mcp-monitor as its single MCP server. The monitor will proxy all tool calls to the downstream servers listed in your config while recording metrics.

6

Review session data and statistics

Use the CLI commands to inspect recorded sessions, per-tool statistics, and cost breakdowns from any terminal.

mcp-monitor sessions --limit 20
mcp-monitor stats --since 24h
mcp-monitor cost --format json

MCP Monitor Examples

Client configuration

Configure Claude Desktop to route all MCP calls through the mcp-monitor multiplexer. Replace the config path with the absolute path to your mcp-monitor.config.json file.

{
  "mcpServers": {
    "mcp-monitor": {
      "command": "mcp-monitor",
      "args": ["serve", "-c", "/absolute/path/to/mcp-monitor.config.json"],
      "env": {}
    }
  }
}

Prompts to try

These are operational commands and dashboard interactions rather than AI prompts. Use them to explore the monitoring data.

- Open http://localhost:4242 and navigate to the Sessions tab to replay a recent agentic run.
- Run: mcp-monitor stats --since 1h  (view per-tool latency and error rates)
- Run: mcp-monitor cost --format csv  (export token cost breakdown)
- Run: mcp-monitor export --format json  (download all session data)
- Send a test event: curl -X POST http://localhost:4242/api/ingest -H 'Content-Type: application/json' -d '{"sessionId":"test","toolName":"read_file","status":"success","latencyMs":120}'

Troubleshooting MCP Monitor

npm install fails with 'E403 Forbidden' from GitHub Packages

GitHub Packages requires authentication even for public packages. Generate a GitHub personal access token with read:packages scope and add it to ~/.npmrc as: //npm.pkg.github.com/:_authToken=YOUR_TOKEN

Dashboard at localhost:4242 shows no data after starting

Confirm your MCP client is configured to route calls through mcp-monitor serve (not a direct connection to downstream servers). Tool calls only appear in the dashboard when they pass through the mcp-monitor proxy layer.

mcp-monitor command not found after npm install -g

Check that the global npm bin directory is in your PATH with 'npm bin -g'. On macOS with nvm this is sometimes ~/.nvm/versions/node/<version>/bin. Add it to your shell profile and re-source it.

Frequently Asked Questions about MCP Monitor

What is MCP Monitor?

MCP Monitor is a Model Context Protocol (MCP) server that observability tool for agentic ai pipelines that intercepts mcp and python tool calls to provide real-time metrics, session replays, and alerts via a local dashboard. it enables centralized monitoring of multiple mcp servers through multiplexer an It connects AI assistants to external tools and data sources through a standardized interface.

How do I install MCP Monitor?

Follow the installation instructions on the MCP Monitor GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with MCP Monitor?

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

Is MCP Monitor free to use?

Yes, MCP Monitor is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Learn More About MCP Servers

Getting Started with MCP

Set up your first MCP server in minutes

MCP Setup Guide

Configure MCP in Claude, Cursor & VS Code

All MCP Tutorials

18+ hands-on guides for developers

MCP FAQ

40+ answers about Model Context Protocol

MCP Monitor Alternatives — Similar Monitoring & Observability Servers

Looking for alternatives to MCP Monitor? Here are other popular monitoring & observability servers you can use with Claude, Cursor, and VS Code.

Netdata

★ 78.9k

Real-time infrastructure monitoring with metrics, logs, alerts, and ML-based anomaly detection.

Kubeshark

★ 11.9k

eBPF-powered network observability for Kubernetes. Indexes L4/L7 traffic with full K8s context, decrypts TLS without keys. Queryable by AI agents via MCP and humans via dashboard.

Mission Control

★ 4.9k

Self-hosted AI agent orchestration platform: dispatch tasks, run multi-agent workflows, monitor spend, and govern operations from one mission control dashboard.

Grafana

★ 3.0k

This MCP server enables natural-language querying of Grafana logs by automatically detecting log sources and service labels. It provides read-only access to log data with intelligent caching for efficient repeat queries.

Sentrux

★ 2.4k

Real-time architectural sensor that helps AI agents close the feedback loop, enabling recursive self-improvement of code quality. Pure Rust.

OpenInference

★ 986

OpenTelemetry Instrumentation for AI Observability

Browse More Monitoring & Observability MCP Servers

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

Monitoring & ObservabilityBrowse All Servers

Quick Config Preview

{ "mcpServers": { "mcp-monitor": { "command": "npx", "args": ["-y", "mcp-monitor"] } } }

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

Read the full setup guide →

Ready to use MCP Monitor?

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
Explore All ServersRead Our Guides