Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Derive security requirements from threat models and business context. Use when translating threats into actionable requirements, creating security user stories, or building security test cases.
.claude/skills/wshobson-security-requirement-extraction/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 44% | 0% |
Transform threat analysis into actionable security requirements.
Business Requirements → Security Requirements → Technical Controls
↓ ↓ ↓
"Protect customer "Encrypt PII at rest" "AES-256 encryption
data" with KMS key rotation"| Type | Focus | Example | | ------------------ | ----------------------- | ------------------------------------- | | Functional | What system must do | "System must authenticate users" | | Non-functional | How system must perform | "Authentication must complete in <2s" | | Constraint | Limitations imposed | "Must use approved crypto libraries" |
| Attribute | Description | | ---------------- | --------------------------- | | Traceability | Links to threats/compliance | | Testability | Can be verified | | Priority | Business importance | | Risk Level | Impact if not met |
Full template library lives in references/details.md. Read that file when you need concrete templates for this skill.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,756 | 17,631 | -15% | 1 | 1 | 0% | 3,754 | 3,863 | +3% | 0 | 0 | — |
case-02 | pass→pass | 16,570 | 15,980 | -4% | 1 | 1 | 0% | 3,511 | 3,813 | +9% | 0 | 0 | — |
case-03 | fail→fail | 9,764 | 6,662 | -32% | 1 | 1 | 0% | 1,030 | 896 | -13% | 0 | 0 | — |
case-04 | pass→fail | 11,289 | 9,150 | -19% | 1 | 1 | 0% | 2,263 | 2,360 | +4% | 0 | 0 | — |
case-05 | fail→pass | 16,530 | 12,789 | -23% | 1 | 1 | 0% | 2,965 | 2,865 | -3% | 0 | 0 | — |
case-06 | pass→pass | 18,750 | 19,481 | +4% | 1 | 1 | 0% | 3,550 | 3,925 | +11% | 0 | 0 | — |
case-07 | fail→fail | 14,738 | 17,425 | +18% | 1 | 1 | 0% | 2,765 | 3,768 | +36% | 0 | 0 | — |
case-08 | pass→pass | 9,715 | 11,637 | +20% | 1 | 1 | 0% | 1,888 | 2,575 | +36% | 0 | 0 | — |
case-09 | pass→pass | 15,607 | 13,086 | -16% | 1 | 1 | 0% | 2,935 | 3,061 | +4% | 0 | 0 | — |
case-10 | pass→pass | 15,796 | 12,190 | -23% | 1 | 1 | 0% | 3,203 | 2,931 | -8% | 0 | 0 | — |
case-11 | pass→pass | 13,017 | 11,157 | -14% | 1 | 1 | 0% | 2,446 | 2,365 | -3% | 0 | 0 | — |
case-12 | fail→fail | 11,482 | 13,263 | +16% | 1 | 1 | 0% | 2,147 | 2,940 | +37% | 0 | 0 | — |
case-13 | pass→pass | 11,776 | 10,639 | -10% | 1 | 1 | 0% | 2,135 | 2,569 | +20% | 0 | 0 | — |
case-14 | pass→pass | 15,299 | 14,091 | -8% | 1 | 1 | 0% | 2,895 | 3,296 | +14% | 0 | 0 | — |
case-15 | fail→pass | 18,970 | 17,611 | -7% | 1 | 1 | 0% | 3,624 | 4,051 | +12% | 0 | 0 | — |
case-16 | fail→pass | 16,458 | 19,608 | +19% | 1 | 1 | 0% | 3,483 | 4,506 | +29% | 0 | 0 | — |
case-17 | fail→pass | 12,028 | 14,964 | +24% | 1 | 1 | 0% | 2,332 | 3,368 | +44% | 0 | 0 | — |
case-18 | pass→pass | 9,270 | 8,177 | -12% | 1 | 1 | 0% | 1,833 | 2,195 | +20% | 0 | 0 | — |
case-19 | pass→pass | 7,767 | 8,120 | +5% | 1 | 1 | 0% | 1,417 | 2,056 | +45% | 0 | 0 | — |
case-20 | pass→pass | 12,559 | 9,828 | -22% | 1 | 1 | 0% | 2,474 | 2,463 | -0% | 0 | 0 | — |
case-21 | fail→fail | 14,469 | 73,739 | +410% | 1 | 1 | 0% | 2,924 | 3,249 | +11% | 0 | 0 | — |
case-22 | fail→fail | 7,082 | 10,149 | +43% | 1 | 1 | 0% | 1,260 | 2,281 | +81% | 0 | 0 | — |
case-23 | pass→pass | 18,202 | 17,826 | -2% | 1 | 1 | 0% | 3,533 | 3,680 | +4% | 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. 23 cases were attempted. The headline lift of +17 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is 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.