Mdata

System Diagnostics

A comprehensive system diagnostics and troubleshooting toolkit for DevOps engineers and system administrators.

Overview

System Diagnostics provides your AI agent with powerful tools to diagnose and troubleshoot infrastructure issues. Perfect for:

• Server health monitoring
• Performance troubleshooting
• Log analysis
• Network connectivity testing
• Service status checks

Tools

run_diagnostic

Execute common diagnostic commands like df -h, free -m, ps aux, netstat, etc.

Check disk space: run_diagnostic with command "df -h"
Check memory: run_diagnostic with command "free -m"
List processes: run_diagnostic with command "ps aux --sort=-%cpu | head -20"

read_log

Read and tail log files for troubleshooting.

Read nginx logs: read_log with path "/var/log/nginx/error.log" and lines 100
Read system logs: read_log with path "/var/log/syslog"

check_service

Check systemd service status.

Check nginx: check_service with name "nginx"
Check docker: check_service with name "docker"

network_check

Test network connectivity using ping, curl, or DNS lookup.

Ping a host: network_check with host "google.com" and method "ping"
Check HTTP: network_check with host "https://api.example.com" and method "curl"
DNS lookup: network_check with host "example.com" and method "dns"

system_summary

Get a comprehensive system overview including CPU, memory, disk, and top processes.

Get system summary: system_summary

Privacy

This tool collects anonymous usage telemetry to help improve the product. Telemetry includes:

• Tool usage counts
• Platform information (OS, architecture)
• Error rates

No personal data, file contents, or command outputs are collected.

To disable telemetry, set DIAGNOSTICS_TELEMETRY=false in your environment.

Requirements

• Node.js 18+
• Linux or macOS
• Standard system utilities (bash, ps, df, free, etc.)

License

MIT License - DevOps Tools Inc.