Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Validate network security architecture with firewall rule analysis, segmentation verification, and defense-in-depth assessment.
.claude/skills/williamzujkowski-network-security-architecture-validator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 38% | 0% |
Trigger conditions:
Not for:
Time normalization:
NOW_ET using NIST/time.gov semantics (America/New_York, ISO-8601): 2025-10-26T01:33:55-04:00NOW_ET for all citation access datesInput validation:
network_identifier must be non-empty stringarchitecture_scope must be one of: perimeter, internal, vpn, all]segmentation_model must be one of: dmz, zero-trust, hybrid] or omittedSource freshness:
Perimeter Security:
Internal Segmentation:
Encrypted Transit:
Monitoring and Detection:
For each finding, provide:
Token budgets:
Ambiguity thresholds:
Abort conditions:
Severity classification:
Required fields:
json{ "network_identifier": "string", "architecture_scope": "perimeter|internal|vpn|all", "segmentation_model": "dmz|zero-trust|hybrid or null", "timestamp": "ISO-8601 with timezone", "findings": [ { "id": "unique identifier", "category": "firewall|segmentation|vpn|monitoring", "severity": "critical|high|medium|low", "cvss_score": 0.0, "title": "brief description", "description": "detailed finding", "nist_control": "SC-7 or similar", "affected_zones": ["network zones or segments"], "remediation": "specific fix steps", "remediation_rule": "firewall rule or policy config" } ], "segmentation_analysis": { "zones_defined": ["list of network zones"], "isolation_score": "0-100 (100=perfect isolation)", "gaps": ["list of segmentation gaps"] }, "summary": { "total_findings": 0, "critical_count": 0, "high_count": 0, "overall_risk": "critical|high|medium|low" } }
Example: Firewall Rule Audit
yaml# Input network_identifier: "production-network" architecture_scope: "perimeter" segmentation_model: "dmz" # Output (abbreviated) { "network_identifier": "production-network", "findings": [ { "id": "NET-001", "category": "firewall", "severity": "high", "cvss_score": 7.5, "title": "Overly permissive SSH ingress rule", "nist_control": "SC-7(5)", "remediation_rule": "iptables -A INPUT -p tcp --dport 22 -s 10.0.0.0/8 -j ACCEPT" } ], "summary": {"high_count": 1, "overall_risk": "high"} }
Token budgets:
Safety:
Auditability:
Determinism:
NIST Standards:
CIS Controls:
Network Security Best Practices:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 16,687 | 12,904 | -23% | 1 | 1 | 0% | 3,643 | 3,978 | +9% | 0 | 0 | — |
case-02 | fail→pass | 15,195 | 14,030 | -8% | 1 | 1 | 0% | 3,291 | 4,790 | +46% | 0 | 0 | — |
case-03 | fail→pass | 13,762 | 10,182 | -26% | 1 | 1 | 0% | 3,072 | 4,071 | +33% | 0 | 0 | — |
case-04 | fail→fail | 10,649 | 13,638 | +28% | 1 | 1 | 0% | 1,870 | 4,823 | +158% | 0 | 0 | — |
case-05 | fail→pass | 15,754 | 12,472 | -21% | 1 | 1 | 0% | 2,879 | 3,156 | +10% | 0 | 0 | — |
case-06 | fail→fail | 12,325 | 15,065 | +22% | 1 | 1 | 0% | 1,882 | 4,169 | +122% | 0 | 0 | — |
case-07 | fail→pass | 12,954 | 7,682 | -41% | 1 | 1 | 0% | 2,706 | 3,725 | +38% | 0 | 0 | — |
case-08 | fail→pass | 12,525 | 4,990 | -60% | 1 | 1 | 0% | 2,247 | 2,902 | +29% | 0 | 0 | — |
case-09 | fail→fail | 11,299 | 8,989 | -20% | 1 | 1 | 0% | 2,569 | 3,848 | +50% | 0 | 0 | — |
case-10 | pass→pass | 5,863 | 5,489 | -6% | 1 | 1 | 0% | 1,556 | 3,126 | +101% | 0 | 0 | — |
case-11 | fail→pass | 13,015 | 7,592 | -42% | 1 | 1 | 0% | 2,963 | 3,495 | +18% | 0 | 0 | — |
case-12 | fail→fail | 11,263 | 9,624 | -15% | 1 | 1 | 0% | 2,177 | 4,084 | +88% | 0 | 0 | — |
case-13 | fail→fail | 10,784 | 12,014 | +11% | 1 | 1 | 0% | 2,212 | 4,258 | +92% | 0 | 0 | — |
case-14 | fail→pass | 4,230 | 6,832 | +62% | 1 | 1 | 0% | 956 | 3,445 | +260% | 0 | 0 | — |
case-15 | fail→fail | 12,612 | 16,153 | +28% | 1 | 1 | 0% | 1,520 | 3,743 | +146% | 0 | 0 | — |
case-16 | pass→pass | 9,657 | 8,260 | -14% | 1 | 1 | 0% | 2,230 | 3,812 | +71% | 0 | 0 | — |
case-17 | fail→pass | 13,391 | 6,961 | -48% | 1 | 1 | 0% | 2,925 | 3,523 | +20% | 0 | 0 | — |
case-18 | fail→pass | 12,408 | 6,220 | -50% | 1 | 1 | 0% | 2,414 | 3,345 | +39% | 0 | 0 | — |
case-19 | fail→pass | 9,948 | 8,236 | -17% | 1 | 1 | 0% | 1,043 | 3,688 | +254% | 0 | 0 | — |
case-20 | pass→pass | 9,729 | 5,921 | -39% | 1 | 1 | 0% | 1,964 | 3,202 | +63% | 0 | 0 | — |
case-21 | fail→pass | 13,134 | 10,475 | -20% | 1 | 1 | 0% | 2,539 | 3,987 | +57% | 0 | 0 | — |
case-22 | pass→pass | 13,598 | 12,147 | -11% | 1 | 1 | 0% | 1,747 | 3,063 | +75% | 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 +55 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.
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.