Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill should be used when performing maintenance or diagnostics on remote Linux systems. Triggers on "remote server", "Linux maintenance", "Ubuntu cleanup", "Debian", "disk space", "apt cleanup", "journal vacuum", "snap cleanup", "system diagnostics". Provides structured three-phase checklists with quantification.
.claude/skills/aiskillstore-remote-system-maintenance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -45% | 0% |
Structured guidance for diagnosing and maintaining remote Linux systems through SSH/tmux sessions, with emphasis on Ubuntu/Debian platforms.
Capture baseline system state:
Examine system health indicators:
Identify maintenance opportunities:
Execute these seven stages in order:
apt update to sync package listsapt upgrade for security and bug fixesapt autoremove to clean unused dependenciesapt clean to reclaim package cache spacejournalctl --vacuum-time=7d to limit log retention/tmp and /var/tmp for cleanup opportunitiesSnap keeps old revisions by default. To identify and remove:
bash# List all disabled snap revisions snap list --all | awk '/disabled/{print $1, $3}' # Remove specific revision snap remove <package-name> --revision=<revision-number>
Important: Requires explicit removal by revision number, not simple package uninstallation.
All maintenance sessions must generate structured logs recording:
Real-world recovery examples:
Typical maintenance session: 15-30 minutes including diagnostics, cleanup, and documentation.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 9,343 | 10,245 | +10% | 1 | 1 | 0% | 1,756 | 2,395 | +36% | 0 | 0 | — |
case-02 | fail→pass | 18,441 | 13,766 | -25% | 1 | 1 | 0% | 3,090 | 3,137 | +2% | 0 | 0 | — |
case-01 | fail→pass | 15,416 | 11,675 | -24% | 1 | 1 | 0% | 2,817 | 2,945 | +5% | 0 | 0 | — |
case-03 | fail→fail | 28,676 | 15,692 | -45% | 1 | 1 | 0% | 3,521 | 3,420 | -3% | 0 | 0 | — |
case-04 | fail→pass | 16,691 | 12,866 | -23% | 1 | 1 | 0% | 2,599 | 2,734 | +5% | 0 | 0 | — |
case-05 | fail→fail | 11,871 | 9,418 | -21% | 1 | 1 | 0% | 2,020 | 2,085 | +3% | 0 | 0 | — |
case-06 | pass→pass | 10,227 | 4,036 | -61% | 1 | 1 | 0% | 1,632 | 1,293 | -21% | 0 | 0 | — |
case-07 | pass→pass | 8,998 | 5,004 | -44% | 1 | 1 | 0% | 1,505 | 1,462 | -3% | 0 | 0 | — |
case-08 | pass→pass | 18,127 | 4,177 | -77% | 1 | 1 | 0% | 1,698 | 1,368 | -19% | 0 | 0 | — |
case-09 | fail→pass | 9,387 | 5,396 | -43% | 1 | 1 | 0% | 1,602 | 1,547 | -3% | 0 | 0 | — |
case-10 | pass→pass | 12,829 | 3,549 | -72% | 1 | 1 | 0% | 2,209 | 1,151 | -48% | 0 | 0 | — |
case-11 | pass→pass | 14,568 | 2,549 | -83% | 1 | 1 | 0% | 2,540 | 975 | -62% | 0 | 0 | — |
case-12 | pass→pass | 10,738 | 4,119 | -62% | 1 | 1 | 0% | 1,651 | 1,328 | -20% | 0 | 0 | — |
case-13 | pass→pass | 9,379 | 3,846 | -59% | 1 | 1 | 0% | 1,381 | 1,228 | -11% | 0 | 0 | — |
case-14 | pass→pass | 16,188 | 2,861 | -82% | 1 | 1 | 0% | 2,660 | 1,067 | -60% | 0 | 0 | — |
case-15 | fail→pass | 11,044 | 2,242 | -80% | 1 | 1 | 0% | 1,689 | 931 | -45% | 0 | 0 | — |
case-16 | fail→pass | 8,766 | 1,690 | -81% | 1 | 1 | 0% | 1,292 | 882 | -32% | 0 | 0 | — |
case-17 | fail→pass | 8,564 | 1,738 | -80% | 1 | 1 | 0% | 1,407 | 920 | -35% | 0 | 0 | — |
case-18 | fail→pass | 10,961 | 1,906 | -83% | 1 | 1 | 0% | 1,683 | 850 | -49% | 0 | 0 | — |
case-19 | pass→pass | 14,976 | 9,388 | -37% | 1 | 1 | 0% | 2,400 | 2,078 | -13% | 0 | 0 | — |
case-21 | pass→pass | 12,792 | 9,490 | -26% | 1 | 1 | 0% | 2,312 | 2,406 | +4% | 0 | 0 | — |
case-22 | pass→pass | 9,780 | 7,935 | -19% | 1 | 1 | 0% | 1,794 | 2,012 | +12% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +36 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.