Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill should be used when the user asks to "run pentest commands", "scan with nmap", "use metasploit exploits", "crack passwords with hydra or john", "scan web vulnerabilities with nikto", "enumerate networks", or needs essential penetration testing command references.
.claude/skills/dokhacgiakhoa-pentest-commands/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-21 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 59% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 51% | 0% |
Provide a comprehensive command reference for penetration testing tools including network scanning, exploitation, password cracking, and web application testing. Enable quick command lookup during security assessments.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 4,805 | 7,974 | +66% | 1 | 1 | 0% | 783 | 1,229 | +57% | 0 | 0 | — |
case-02 | fail→fail | 8,975 | 10,772 | +20% | 1 | 1 | 0% | 1,044 | 1,473 | +41% | 0 | 0 | — |
case-03 | pass→pass | 4,190 | 3,802 | -9% | 1 | 1 | 0% | 763 | 1,200 | +57% | 0 | 0 | — |
case-04 | pass→fail | 5,252 | 5,069 | -3% | 1 | 1 | 0% | 880 | 1,399 | +59% | 0 | 0 | — |
case-05 | fail→fail | 9,078 | 10,584 | +17% | 1 | 1 | 0% | 1,081 | 1,534 | +42% | 0 | 0 | — |
case-06 | pass→fail | 5,278 | 9,597 | +82% | 1 | 1 | 0% | 878 | 1,325 | +51% | 0 | 0 | — |
case-07 | fail→fail | 7,006 | 20,421 | +191% | 1 | 1 | 0% | 1,300 | 2,346 | +80% | 0 | 0 | — |
case-08 | pass→pass | 2,861 | 3,478 | +22% | 1 | 1 | 0% | 576 | 1,143 | +98% | 0 | 0 | — |
case-09 | pass→pass | 5,756 | 6,300 | +9% | 1 | 1 | 0% | 1,140 | 1,672 | +47% | 0 | 0 | — |
case-10 | pass→fail | 20,505 | 17,835 | -13% | 1 | 1 | 0% | 3,009 | 2,859 | -5% | 0 | 0 | — |
case-11 | fail→pass | 8,575 | 8,249 | -4% | 1 | 1 | 0% | 1,550 | 1,915 | +24% | 0 | 0 | — |
case-12 | pass→pass | 8,133 | 5,603 | -31% | 1 | 1 | 0% | 1,559 | 1,474 | -5% | 0 | 0 | — |
case-13 | pass→pass | 5,577 | 6,611 | +19% | 1 | 1 | 0% | 935 | 1,762 | +88% | 0 | 0 | — |
case-14 | fail→pass | 13,695 | 4,622 | -66% | 1 | 1 | 0% | 2,105 | 1,325 | -37% | 0 | 0 | — |
case-15 | pass→pass | 4,890 | 4,716 | -4% | 1 | 1 | 0% | 812 | 1,300 | +60% | 0 | 0 | — |
case-16 | fail→fail | 14,277 | 14,465 | +1% | 1 | 1 | 0% | 1,503 | 1,566 | +4% | 0 | 0 | — |
case-17 | fail→fail | 8,964 | 9,130 | +2% | 1 | 1 | 0% | 872 | 1,263 | +45% | 0 | 0 | — |
case-18 | fail→fail | 8,333 | 6,322 | -24% | 1 | 1 | 0% | 1,358 | 957 | -30% | 0 | 0 | — |
case-19 | pass→pass | 7,541 | 4,387 | -42% | 1 | 1 | 0% | 1,497 | 1,288 | -14% | 0 | 0 | — |
case-20 | pass→pass | 6,982 | 4,512 | -35% | 1 | 1 | 0% | 1,277 | 1,316 | +3% | 0 | 0 | — |
case-21 | fail→pass | 15,074 | 13,016 | -14% | 1 | 1 | 0% | 1,854 | 971 | -48% | 0 | 0 | — |
case-22 | pass→pass | 7,355 | 5,493 | -25% | 1 | 1 | 0% | 1,361 | 1,501 | +10% | 0 | 0 | — |
case-23 | pass→fail | 11,413 | 7,464 | -35% | 1 | 1 | 0% | 1,970 | 1,166 | -41% | 0 | 0 | — |
case-24 | pass→pass | 4,675 | 4,362 | -7% | 1 | 1 | 0% | 804 | 1,337 | +66% | 0 | 0 | — |
case-25 | pass→pass | 16,157 | 5,516 | -66% | 1 | 1 | 0% | 1,650 | 1,500 | -9% | 0 | 0 | — |
case-26 | pass→pass | 13,597 | 15,517 | +14% | 1 | 1 | 0% | 2,363 | 3,143 | +33% | 0 | 0 | — |
case-27 | pass→pass | 4,654 | 8,493 | +82% | 1 | 1 | 0% | 913 | 2,089 | +129% | 0 | 0 | — |
case-28 | pass→pass | 7,480 | 6,289 | -16% | 1 | 1 | 0% | 1,480 | 1,821 | +23% | 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. 28 cases were attempted. The headline lift of -25 percentage points is the difference between those two pass rates over the 28 comparable cases. 5 cases got worse with the skill loaded, and they are included in that figure.
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.