homelab-companion

A skill that catches the wrong AI default before it ships.

A Claude Code skill-pack that catches confidently-wrong homelab advice before it ships. Two modes share one catalog: preventive (flag the AI default trap when reviewing a Docker compose, ufw rule, systemd timer, or sync workflow) and retrospective (phase-by-phase post-mortem when something already broke). Each entry has a "Why LLMs miss this" section that names the plausible-but-wrong default a model reaches for without context — and how to redirect. In testing — 4 evals run 5 times each — the skill scored 22/22 vs the baseline's 4/22. The +83 percentage point gap is where AI defaults are confident but wrong.

Iteration-1 benchmark for homelab-companion: pass rate per eval, with-skill versus baseline, four evals showing 22/22 versus 4/22, an eval-3 zoom comparing tunnel-stall (wrong) versus parent-restart-orphans-child (correct mechanism).
22/22 with skill · 4/22 baseline · eval-3 zoom shows the mechanism gap
/plugin install ikkeseb/skills
specifications
HOST
cc skill-pack
MODES
preventive + retro
DELTA
+83 pp eval