Server Health Check

One command from the user → full picture of the server.

Config

Servers are listed in ~/.jarvis/servers.md (see the deploy skill). Never hardcode hosts or passwords in this file.

Checks

Run over SSH, all in one connection:

ssh $USER@$HOST '
  echo "=== uptime ===";  uptime
  echo "=== disk ===";    df -h / | tail -1
  echo "=== memory ===";  free -h | grep Mem
  echo "=== docker ===";  docker ps --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}"
'

If the user named a project, also tail its logs:

ssh $USER@$HOST "docker logs --tail 30 $PROJECT 2>&1"

Report Format

Short summary, problems first:

• 🔴 anything down, restarting, or erroring in logs
• 🟡 disk > 80%, memory > 90%, container restarted recently
• 🟢 one line: "all N containers up, disk X%, mem Y%"

Don't dump raw command output at the user unless they ask — summarize, flag anomalies, suggest the fix.

Escalation

If something is red: propose the fix (restart container, clear disk, check logs deeper) and ask "fix it?" — don't auto-restart production services without a go-ahead.