audit

Audits homelab server health across eight phases over SSH and remediates findings.

Updated Dec 22, 2024
One-click install
npx skills add https://github.com/jkrumm/homelab --skill audit-jkrumm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: audit
Source: https://github.com/jkrumm/homelab/tree/main/.claude/skills/audit
Command: npx skills add https://github.com/jkrumm/homelab --skill audit-jkrumm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manually checking a self-hosted homelab for failing containers, full disks, watchdog escalations, and monitoring gaps is slow and error-prone. This Skill runs a structured 8-phase health audit over SSH and then fixes or proposes fixes for every issue found. ## Core Features & Use Cases - 8-Phase Health Audit: Checks system resources, Docker container health across two compose stacks, watchdog escalation state, storage and mounts, Tailscale and Cloudflare Tunnel connectivity, pending updates, recent service errors, and UptimeKuma monitor coverage. - Guided Remediation: Auto-fixes reversible issues (container restarts, dangling image pruning, watchdog state resets) and asks for confirmation before irreversible or outward-facing changes like volume deletion or tunnel changes. - Structured Report: Produces a per-phase report with green/yellow/red status indicators and a prioritized recommendations list. - Use Case: Run a weekly audit of your HomeLab server to catch a crashed Immich container, a nearly full encrypted HDD, and a missing UptimeKuma monitor in one pass, then restart the container and sync the monitor config automatically. ## Quick Start Ask the assistant to run a full health audit of the homelab server and fix any issues it finds.

Frequently Asked Questions about audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit the health of a self-hosted homelab server?▼

Run the audit skill, which executes 8 sequential phases over SSH covering system resources, container health, watchdog state, storage, tunnel connectivity, updates, error logs, and monitor coverage. It outputs a structured report with green, yellow, and red status per phase.

How to check Docker container health across multiple compose stacks?▼

The audit checks both the main homelab stack and the homelab-private stack using compose project label filters and make health commands. Any container not running is flagged CRITICAL, and restart counts above 3 trigger a WARN with logs and a restart offer.

Does the audit automatically fix issues it finds?▼

Yes, reversible remediations like container restarts, dangling image pruning, and watchdog state resets are applied automatically and reported. Irreversible or outward-facing changes such as volume deletion, disk prune, or tunnel changes require explicit confirmation first.

Can the audit detect missing UptimeKuma monitors?▼

Yes, phase 8 compares running containers and Cloudflare Tunnel hostnames against the docker_container and HTTP entries in both monitors.yaml files. Missing monitors are flagged, and the repair flow edits the YAML, commits, pushes, and runs make uk-sync.

What access does the homelab audit require?▼

It requires SSH access to the homelab host and never runs local commands. Sudo operations like watchdog state resets retrieve the server password from the 1Password CLI item Private/homelab-server.

What actions will the audit never take on the server?▼

It never reboots the server, runs docker compose down, deletes volumes, or takes any action affecting all services simultaneously without explicit discussion. The LUKS-encrypted HDD mount failure is reported with a recovery hint but never auto-fixed.