Menü
Quick cluster health dashboard with live GPU, container, and disk status from all nodes
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
| name | cluster-status |
| description | Quick cluster health dashboard with live GPU, container, and disk status from all nodes |
| triggers | ["cluster status","cluster health","node status"] |
| disable-model-invocation | true |
!`ssh -o ConnectTimeout=3 -o BatchMode=yes foundry 'nvidia-smi --query-gpu=index,name,utilization.gpu,memory.used,memory.total,temperature.gpu --format=csv,noheader' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes foundry 'docker ps --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}" 2>/dev/null | head -15' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes foundry 'echo "DISK:" && df -h / /mnt/vault/models 2>/dev/null | tail -2 && echo "RAM:" && free -h | head -2' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes workshop 'nvidia-smi --query-gpu=index,name,utilization.gpu,memory.used,memory.total,temperature.gpu --format=csv,noheader' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes workshop 'docker ps --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}" 2>/dev/null | head -15' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes vault 'docker ps --format "table {{.Names}}\t{{.Status}}" 2>/dev/null | head -20' 2>/dev/null || echo "UNREACHABLE"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes vault 'df -h /mnt/user 2>/dev/null | tail -1' 2>/dev/null || echo "UNREACHABLE"`
!`systemctl is-active local-system-gateway local-system-mind local-system-memory local-system-perception local-system-ui 2>/dev/null || echo "systemctl not available (running from Windows)"`
!`ssh -o ConnectTimeout=3 -o BatchMode=yes dev 'nvidia-smi --query-gpu=index,name,memory.used,memory.total --format=csv,noheader' 2>/dev/null || echo "UNREACHABLE or no SSH alias"`
Auto-route tasks to the optimal tool, model, and subscription. Registry-backed provider and model truth only. Content governance enforced.
Manage overnight autonomous coding tasks
Cross-reference hardware inventory against node configs, detect discrepancies
Perpetual improvement loop. Find the weakest part of the system and make it better. Never stops. Use when continuing work, when idle, when asked to improve, or when nothing specific is requested.
Run drift detection to verify system matches architectural decisions
Structured process for making and documenting architecture decisions with ADR format