Quick Start

Bash
# System health check
ps aux | grep -v grep | head -10
df -h
free -m
netstat -tulpn | head -5

Tool Integration:

• search: when gathering information about errors, documentation, or solutions
• generate_text: when creating reports, documentation, or explanations
• code: when writing scripts, analyzing logs, or processing data
• cmd_execute: when running system commands, checking processes, or file operations

Workflow

Reasoning Workflow (8 steps):

Progress:

• Step 1: Problem identification (search tool)
• Step 2: System state analysis (cmd_execute tool)
• Step 3: Data collection (cmd_execute + code tool)
• Step 4: Pattern analysis (code tool)
• Step 5: Root cause analysis (generate_text tool)
• Step 6: Solution implementation (cmd_execute tool)

Error Correction Mechanism:

• Step 1 error → Re-search with different keywords
• Step 2 error → Use alternative commands (top instead of ps, lsof instead of netstat)
• Step 3 error → Check permissions, use sudo if needed
• Step 4 error → Simplify analysis, focus on key metrics
• Step 5 error → Gather more context data
• Step 6 error → Rollback and try alternative solution

Examples

Example 1: Input: "High CPU usage investigation" Output:

Bash
# Check top processes
ps aux --sort=-%cpu | head -10
# Check load average
uptime
# Identify specific process
top -b -n1 | head -20

Example 2: Input: "Network connectivity issue" Output:

Bash
# Check network interfaces
ip addr show
# Check routing
ip route show
# Test connectivity
ping -c 4 8.8.8.8
# Check listening ports
netstat -tulpn | grep LISTEN

CMD Commands Reference

System Monitoring:

• ps aux - Process list (when: checking running processes)
• top -b -n1 - Current system state (when: real-time monitoring needed)
• htop - Interactive process viewer (when: detailed process analysis)
• free -m - Memory usage (when: RAM issues suspected)
• df -h - Disk usage (when: storage problems)
• du -sh /path/* - Directory sizes (when: finding space usage)

Network Commands:

• netstat -tulpn - Network connections (when: port/service issues)
• ss -tulpn - Modern netstat (when: newer systems)
• lsof -i :port - Process using port (when: port conflicts)
• ping -c 4 host - Connectivity test (when: network issues)
• traceroute host - Route tracing (when: routing problems)

File Operations:

• ls -la - Detailed listing (when: checking permissions)
• find /path -name pattern - File search (when: locating files)
• grep -r "pattern" /path - Text search (when: log analysis)
• tail -f /var/log/file - Live log monitoring (when: real-time logs)

Computational Formulas

CPU Calculations:

• CPU utilization = (1 - idle_time/total_time) × 100
• Load average interpretation: 1.0 = 100% utilization per core
• CPU frequency scaling: performance = base_freq × multiplier

Memory Calculations:

• Memory usage % = (used_memory/total_memory) × 100
• Available memory = free + buffers + cache
• Swap usage = swap_used/swap_total × 100

Storage Calculations:

• Disk usage % = used_space/total_space × 100
• IOPS = operations/second
• Throughput = bytes_transferred/time_elapsed

Network Calculations:

• Bandwidth utilization = (bytes_sent + bytes_received)/interface_speed
• Latency = round_trip_time/2
• Packet loss % = (packets_lost/packets_sent) × 100

Unit Conversions:

• 1 KB = 1,024 bytes
• 1 MB = 1,024 KB = 1,048,576 bytes
• 1 GB = 1,024 MB = 1,073,741,824 bytes
• 1 Gbps = 125 MB/s (theoretical)

Best Practices

1. Always check system load before intensive operations
2. Use non-destructive commands first (read-only)
3. Document findings in structured format
4. Verify changes with before/after measurements
5. Keep command history for reproducibility
6. Use appropriate tools for system type (systemd vs init)

Common Pitfalls

1. Running commands without understanding impact - Always read man pages first
2. Ignoring permissions - Check with ls -la before file operations
3. Not checking system resources - Monitor CPU/memory during analysis
4. Mixing measurement units - Always convert to common units
5. Assuming root access - Many commands work without sudo
6. Not saving command output - Use tee to save results: command | tee output.txt

---