Content
<p align="center">
<img src="assets/logo.jpg" alt="MCP Server Monitor" width="200" />
</p>
<p align="center">
<a href="README.md"><b>English</b></a> | <a href="README_RU.md">Русский</a>
</p>
# mcp-server-monitor
> Real-time system monitoring through Claude Code. CPU, memory, disk, network, Docker, and process insights — right in your AI workflow.
<p align="center">
<img src="https://img.shields.io/badge/MCP-Server-blue" alt="MCP Server" />
<img src="https://img.shields.io/badge/TypeScript-5.7-blue" alt="TypeScript" />
<img src="https://img.shields.io/badge/Node.js-18%2B-green" alt="Node.js" />
<img src="https://img.shields.io/badge/License-MIT-yellow" alt="MIT License" />
</p>
<p align="center">
<img src="assets/banner.jpg" alt="MCP Server Monitor Banner" width="100%" />
</p>
## Why?
DevOps in your IDE. No more switching to terminals, dashboards, or Grafana just to check if your server is healthy. Ask Claude directly:
- *"Is the server overloaded?"*
- *"Which process is eating all the CPU?"*
- *"How much disk space is left?"*
- *"Are my Docker containers running?"*
## Quick Start
### Install globally
```bash
npm install -g mcp-server-monitor
```
### Or use with Claude Code
Add to your Claude Code MCP config (`~/.claude/claude_desktop_config.json`):
```json
{
"mcpServers": {
"server-monitor": {
"command": "npx",
"args": ["-y", "mcp-server-monitor"]
}
}
}
```
### Or build from source
```bash
git clone https://github.com/VisualLock/mcp-server-monitor.git
cd mcp-server-monitor
npm install
npm run build
```
## Tools (22 total)
### System
| Tool | Description |
|------|-------------|
| `system_overview` | Complete system overview: OS, CPU, memory, uptime |
| `system_uptime` | System uptime and boot time |
| `system_users` | Currently logged-in users |
### CPU
| Tool | Description |
|------|-------------|
| `cpu_info` | Detailed CPU hardware info: model, cores, cache |
| `cpu_load` | Current CPU load with per-core breakdown and visual bars |
| `cpu_temperature` | CPU temperature readings |
| `cpu_history` | Monitor CPU load over time (1-60 seconds) |
### Memory
| Tool | Description |
|------|-------------|
| `memory_info` | RAM and swap usage with visual bars |
| `memory_layout` | Physical memory module details |
### Disk
| Tool | Description |
|------|-------------|
| `disk_usage` | All mounted filesystems with usage bars |
| `disk_io` | Disk I/O statistics: read/write speeds |
| `disk_smart` | S.M.A.R.T. health data for physical disks |
### Network
| Tool | Description |
|------|-------------|
| `network_interfaces` | All network interfaces with IPs, MACs, status |
| `network_stats` | Traffic statistics per interface |
| `network_connections` | Active connections with state filtering |
| `network_latency` | Ping any host to measure latency |
### Processes
| Tool | Description |
|------|-------------|
| `process_list` | Top processes by CPU or memory usage |
| `process_find` | Find processes by name |
| `process_detail` | Detailed info for a specific PID |
### Docker
| Tool | Description |
|------|-------------|
| `docker_containers` | List all Docker containers with status |
| `docker_stats` | Container resource usage (CPU, mem, network) |
| `docker_images` | Locally available Docker images |
### Health & Alerts
| Tool | Description |
|------|-------------|
| `health_check` | Comprehensive health check with configurable thresholds |
| `performance_report` | Full performance report combining all metrics |
## Example Output
```
📊 CPU Load
═══════════════════════════════════
Total: 23.5%
User: 18.2%
System: 5.3%
Idle: 76.5%
Per-Core Load:
─────────────────────────────
Core 0: [████░░░░░░░░░░░░░░░░] 22.1%
Core 1: [█████░░░░░░░░░░░░░░░] 25.0%
Core 2: [████░░░░░░░░░░░░░░░░] 19.8%
Core 3: [██████░░░░░░░░░░░░░░] 27.3%
```
```
✅ Health Check — OK
═══════════════════════════════════════════════════════
✅ CPU Load: 23.5% — CPU load normal
✅ Memory: 62.3% (10.24 GB/16.00 GB) — Memory usage normal
✅ Swap: 5.1% (0.52 GB/10.00 GB) — Swap usage normal
✅ Disk (/): 45.2% (214.50 GB/475.00 GB) — Disk / OK
⚠️ Disk (/data): 82.1% (780.00 GB/950.00 GB) — Disk /data filling up
```
## Supported Platforms
- **Windows** 10/11/Server
- **macOS** (Intel & Apple Silicon)
- **Linux** (all major distros)
## Requirements
- Node.js 18+
- Docker (optional, for container monitoring)
## Contributing
PRs welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
## License
[MIT](LICENSE) — use it however you want.
MCP Config
Below is the configuration for this MCP Server. You can copy it directly to Cursor or other MCP clients.
mcp.json
Connection Info
You Might Also Like
markitdown
MarkItDown-MCP is a lightweight server for converting URIs to Markdown.
markitdown
Python tool for converting files and office documents to Markdown.
Filesystem
Node.js MCP Server for filesystem operations with dynamic access control.
TrendRadar
TrendRadar: Your hotspot assistant for real news in just 30 seconds.
mempalace
The highest-scoring AI memory system ever benchmarked. And it's free.
mempalace
The highest-scoring AI memory system ever benchmarked. And it's free.