swarm-zap/skills/homelab-service-health/SKILL.md

name, description

name	description
homelab-service-health	Run lightweight read-only health checks for OpenClaw homelab services (runtime, Docker workloads, LAN endpoints, optional namespace-scoped Kubernetes checks). Use for status snapshots, outage triage, drift detection, and safe next-step remediation planning.

Homelab Service Health

Priorities

• Detect breakage fast.
• Keep checks cheap for Raspberry Pi-class hardware.
• Recommend low-risk remediations before heavy actions.

Default scope

Unless user narrows scope, check in this order:

1. OpenClaw runtime health/status.
2. Docker services listed in TOOLS.md (for this workspace: searxng, whisper-server, brave-search MCP).
3. Reachability of documented LAN endpoints.
4. Optional Kubernetes checks scoped to intended namespace only.

Safety constraints

• Read-only first; do not restart/change config without approval.
• Avoid destructive or cluster-wide kubectl operations.
• Prefer summary signals over full log dumps.

Workflow

1. Load expected service inventory from TOOLS.md.
2. Run quick status/reachability checks.
3. Classify each component: healthy | degraded | down | unknown.
4. Provide likely cause and smallest safe next step for failures.
5. Offer opt-in remediation commands/actions.

Output contract

• Overall: healthy | degraded | incident
• Healthy now: short list
• Issues: service -> symptom -> likely cause -> safe next step
• Recommended actions: ordered low-risk first
• Escalate when: explicit trigger for deeper investigation