| name | server-dashboard |
| description | Quick server health overview. Shows CPU, memory, disk, services, and recent errors in one view. Use when user asks for "サーバー状況" or "server status". |
Server Dashboard - Quick Health Check
PURPOSE
Provide instant overview of server health without digging through multiple commands.
WHEN TO USE
- TRIGGERS:
- "サーバー状況" / "サーバーの状態"
- "server status" / "server health"
- "システム確認"
- "何か問題ある?" (in server context)
WORKFLOW
- Gather system metrics (parallel when possible)
- Check critical services
- Scan recent logs for errors
- Format into dashboard view
- Highlight any issues
DATA COLLECTION
System Resources
uptime
cat /proc/loadavg
free -h
df -h | grep -E '^/dev/'
Services (auto-detect)
systemctl --user list-units --state=running --no-pager | head -20
systemctl list-units --state=failed --no-pager 2>/dev/null
pm2 list 2>/dev/null
docker ps --format "table {{.Names}}\t{{.Status}}" 2>/dev/null
Recent Errors
journalctl --user -p err -S "1 hour ago" --no-pager 2>/dev/null | tail -10
tail -5 /var/log/nginx/error.log 2>/dev/null
Network
ss -ltnp 2>/dev/null | head -15
OUTPUT FORMAT
🖥️ SERVER DASHBOARD - [hostname]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
📊 RESOURCES
CPU Load: [1m] [5m] [15m] [status emoji]
Memory: [used]/[total] ([%]) [status emoji]
Disk /: [used]/[total] ([%]) [status emoji]
🔧 SERVICES
✅ openclaw-gateway (running, 4h)
✅ nginx (running)
⚠️ some-service (failed)
🌐 PORTS
:18789 → openclaw-gateway
:80 → nginx
:443 → nginx
⚠️ RECENT ERRORS (last hour)
[timestamp] [service] [message]
...
📈 STATUS: [🟢 HEALTHY | 🟡 WARNING | 🔴 CRITICAL]
STATUS THRESHOLDS
- CPU Load > cores: 🟡, > 2x cores: 🔴
- Memory > 80%: 🟡, > 95%: 🔴
- Disk > 80%: 🟡, > 95%: 🔴
- Any failed service: 🟡
- Multiple errors: 🔴
SAFETY
- Read-only commands only
- No service restarts without explicit request
- Works with linux-service-triage for deeper diagnosis