Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Check HIPAA compliance for healthcare data security requirements. Use when auditing healthcare applications. Trigger with 'check HIPAA compliance', 'validate health data security', or 'audit PHI protection'.
.claude/skills/jeremylongshore-checking-hipaa-compliance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 293% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 281% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 586% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -69% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -60% | 0% |
Audit healthcare applications for HIPAA compliance, checking PHI data handling, access controls, encryption requirements, audit logging, and Business Associate Agreement adherence.
This skill automates the process of identifying potential HIPAA compliance issues within a software project. By using the hipaa-compliance-checker plugin, it helps developers and security professionals proactively address vulnerabilities and ensure adherence to HIPAA guidelines.
This skill activates when you need to:
User request: "Check HIPAA compliance of the patient data API codebase."
The skill will:
User request: "Assess the HIPAA readiness of our AWS infrastructure configuration."
The skill will:
This skill can be integrated with other security and compliance tools to provide a comprehensive view of a system's security posture. The generated reports can be used as input for vulnerability management systems and security information and event management (SIEM) platforms.
If security scanning fails:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 26,034 | 3,864 | -85% | 1 | 1 | 0% | 4,674 | 1,000 | -79% | 0 | 0 | — |
case-07 | pass→fail | 20,144 | 7,130 | -65% | 1 | 1 | 0% | 3,505 | 1,086 | -69% | 0 | 0 | — |
case-02 | fail→fail | 11,483 | 11,253 | -2% | 1 | 1 | 0% | 1,206 | 1,963 | +63% | 0 | 0 | — |
case-03 | fail→fail | 7,686 | 14,847 | +93% | 1 | 1 | 0% | 1,200 | 3,199 | +167% | 0 | 0 | — |
case-04 | pass→pass | 16,382 | 15,600 | -5% | 1 | 1 | 0% | 2,715 | 3,430 | +26% | 0 | 0 | — |
case-05 | pass→pass | 11,180 | 12,250 | +10% | 1 | 1 | 0% | 1,738 | 2,689 | +55% | 0 | 0 | — |
case-06 | pass→fail | 15,314 | 13,744 | -10% | 1 | 1 | 0% | 2,575 | 1,018 | -60% | 0 | 0 | — |
case-08 | pass→pass | 12,799 | 29,206 | +128% | 1 | 1 | 0% | 2,357 | 5,883 | +150% | 0 | 0 | — |
case-09 | pass→pass | 15,033 | 19,363 | +29% | 1 | 1 | 0% | 2,760 | 4,401 | +59% | 0 | 0 | — |
case-10 | fail→pass | 7,582 | 20,792 | +174% | 1 | 1 | 0% | 1,183 | 4,648 | +293% | 0 | 0 | — |
case-11 | fail→fail | 11,740 | 10,105 | -14% | 1 | 1 | 0% | 1,552 | 1,958 | +26% | 0 | 0 | — |
case-12 | fail→fail | 12,947 | 16,358 | +26% | 1 | 1 | 0% | 1,393 | 1,751 | +26% | 0 | 0 | — |
case-13 | fail→fail | 3,423 | 6,889 | +101% | 1 | 1 | 0% | 242 | 993 | +310% | 0 | 0 | — |
case-14 | pass→fail | 20,801 | 14,803 | -29% | 1 | 1 | 0% | 2,635 | 1,954 | -26% | 0 | 0 | — |
case-15 | pass→fail | 11,828 | 7,033 | -41% | 1 | 1 | 0% | 1,967 | 1,223 | -38% | 0 | 0 | — |
case-16 | pass→pass | 20,957 | 17,027 | -19% | 1 | 1 | 0% | 3,819 | 4,152 | +9% | 0 | 0 | — |
case-17 | pass→pass | 24,362 | 24,870 | +2% | 1 | 1 | 0% | 2,104 | 3,504 | +67% | 0 | 0 | — |
case-18 | fail→fail | 11,194 | 3,627 | -68% | 1 | 1 | 0% | 1,915 | 1,084 | -43% | 0 | 0 | — |
case-19 | fail→fail | 3,180 | 4,653 | +46% | 1 | 1 | 0% | 410 | 1,253 | +206% | 0 | 0 | — |
case-20 | fail→pass | 6,354 | 21,821 | +243% | 1 | 1 | 0% | 1,057 | 4,032 | +281% | 0 | 0 | — |
case-21 | fail→pass | 5,188 | 22,232 | +329% | 1 | 1 | 0% | 696 | 4,773 | +586% | 0 | 0 | — |
case-22 | fail→fail | 8,727 | 7,928 | -9% | 1 | 1 | 0% | 907 | 1,619 | +79% | 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, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -25 percentage points is the difference between those two pass rates over the 20 comparable cases. 7 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.