Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →PCI DSS v4.0.1 compliance expert. Provides guidance on payment card industry security, ROC completion, SAQ selection, requirement interpretation, and the new March 2025 mandatory requirements.
.claude/skills/grcengclub-pci-dss-expert/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✓→✗ | ▼ Worse | 23% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 31% | 0% |
| case-11 | ✓→✓ | ＝ Same ✓ | 55% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 16% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 31% | 0% |
Deep expertise in Payment Card Industry Data Security Standard v4.0.1.
| Req | Title | Focus | |-----|-------|-------| | 1 | Network Security Controls | Firewalls, segmentation, NSCs | | 2 | Secure Configurations | Hardening, inventory, defaults | | 3 | Protect Stored Data | Encryption, PAN, SAD, retention | | 4 | Cryptography in Transit | TLS, secure channels | | 5 | Malware Protection | Anti-malware, phishing | | 6 | Secure Development | SDLC, patches, web apps | | 7 | Access Restriction | Need-to-know, RBAC | | 8 | User Authentication | MFA, passwords, accounts | | 9 | Physical Security | Facility, media, visitors | | 10 | Logging & Monitoring | Audit trails, SIEM, review | | 11 | Security Testing | Scans, pen tests, IDS/IPS | | 12 | Security Policies | Policies, training, IR |
ROC (Report on Compliance):
SAQ (Self-Assessment Questionnaire):
AOC (Attestation of Compliance):
Key concepts:
Critical new requirements:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 16,743 | 24,230 | +45% | 1 | 1 | 0% | 2,432 | 3,182 | +31% | 0 | 0 | — |
case-11 | pass→pass | 8,190 | 8,437 | +3% | 1 | 1 | 0% | 1,234 | 1,914 | +55% | 0 | 0 | — |
case-02 | pass→pass | 16,335 | 13,357 | -18% | 1 | 1 | 0% | 2,303 | 2,661 | +16% | 0 | 0 | — |
case-03 | pass→pass | 13,622 | 11,608 | -15% | 1 | 1 | 0% | 1,847 | 2,418 | +31% | 0 | 0 | — |
case-04 | pass→pass | 8,527 | 7,429 | -13% | 1 | 1 | 0% | 1,392 | 1,805 | +30% | 0 | 0 | — |
case-05 | pass→pass | 8,949 | 10,777 | +20% | 1 | 1 | 0% | 1,343 | 2,311 | +72% | 0 | 0 | — |
case-06 | pass→pass | 10,889 | 13,127 | +21% | 1 | 1 | 0% | 1,600 | 2,627 | +64% | 0 | 0 | — |
case-07 | pass→pass | 9,974 | 9,378 | -6% | 1 | 1 | 0% | 1,616 | 2,075 | +28% | 0 | 0 | — |
case-08 | pass→fail | 13,139 | 12,733 | -3% | 1 | 1 | 0% | 2,161 | 2,652 | +23% | 0 | 0 | — |
case-09 | pass→pass | 10,687 | 9,497 | -11% | 1 | 1 | 0% | 1,686 | 2,094 | +24% | 0 | 0 | — |
case-10 | pass→pass | 12,477 | 9,454 | -24% | 1 | 1 | 0% | 2,011 | 2,153 | +7% | 0 | 0 | — |
case-12 | fail→fail | 7,393 | 8,104 | +10% | 1 | 1 | 0% | 1,219 | 1,891 | +55% | 0 | 0 | — |
case-13 | pass→pass | 11,856 | 13,941 | +18% | 1 | 1 | 0% | 1,832 | 2,895 | +58% | 0 | 0 | — |
case-14 | pass→pass | 4,841 | 5,593 | +16% | 1 | 1 | 0% | 853 | 1,482 | +74% | 0 | 0 | — |
case-15 | pass→pass | 12,528 | 20,610 | +65% | 1 | 1 | 0% | 1,914 | 3,798 | +98% | 0 | 0 | — |
case-16 | pass→pass | 5,239 | 5,558 | +6% | 1 | 1 | 0% | 899 | 1,486 | +65% | 0 | 0 | — |
case-17 | pass→pass | 5,064 | 5,860 | +16% | 1 | 1 | 0% | 937 | 1,579 | +69% | 0 | 0 | — |
case-18 | pass→pass | 4,532 | 4,519 | -0% | 1 | 1 | 0% | 768 | 1,344 | +75% | 0 | 0 | — |
case-19 | pass→pass | 4,703 | 5,155 | +10% | 1 | 1 | 0% | 733 | 1,494 | +104% | 0 | 0 | — |
case-20 | pass→pass | 3,768 | 4,178 | +11% | 1 | 1 | 0% | 630 | 1,301 | +107% | 0 | 0 | — |
case-21 | pass→pass | 12,274 | 14,798 | +21% | 1 | 1 | 0% | 1,966 | 3,000 | +53% | 0 | 0 | — |
case-22 | pass→pass | 4,540 | 5,165 | +14% | 1 | 1 | 0% | 714 | 1,426 | +100% | 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 -100 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Other measured skills in the registry, with their headline benchmark lift.