Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Validate software supply chain security with SBOM generation, dependency scanning, provenance verification, and SLSA attestation.
.claude/skills/williamzujkowski-software-supply-chain-security-validator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 267% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 1403% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 163% | 0% |
Trigger conditions:
Not for:
Time normalization:
NOW_ET using NIST/time.gov semantics (America/New_York, ISO-8601): 2025-10-25T22:34:11-04:00NOW_ET for all citation access datesInput validation:
artifact_type must be one of: source, container, binary, packageartifact_location must be accessible path/URL or valid registry referencesbom_format must be: spdx, cyclonedx, or bothslsa_level if specified must be integer 1-4vulnerability_threshold must be: low, medium, high, or criticalverification_tier must be: T1 or T2Source freshness:
Fast path for 80% of supply chain validation:
<artifact>-sbom.<format>vulnerability_thresholdDecision point: If slsa_level specified or verification_tier=T2 → proceed to T2
T1 output schema:
json{ "sbom_location": "path/to/sbom.spdx.json", "format": "spdx", "component_count": 247, "vulnerabilities": [ { "cve_id": "CVE-2024-1234", "severity": "HIGH", "cvss_score": 8.2, "component": "lodash@4.17.20", "fixed_version": "4.17.21", "description": "Prototype pollution vulnerability" } ], "critical_count": 2, "high_count": 5, "compliance": "PASS_WITH_WARNINGS" }
Extended validation with provenance and attestation:
slsa_levelT2 output schema:
json{ "sbom": { /* T1 output */ }, "provenance": { "available": true, "slsa_level": "SLSA_BUILD_LEVEL_3", "builder": "https://github.com/slsa-framework/slsa-github-generator/.github/workflows/builder.yml@v1.2.0", "source_repo": "https://github.com/org/repo", "commit_sha": "abc123def456", "build_timestamp": "2025-10-24T14:32:00Z" }, "signature_verification": { "verified": true, "signer": "sigstore.dev", "certificate_identity": "https://github.com/org/repo/.github/workflows/release.yml@refs/heads/main", "transparency_log_entry": "https://rekor.sigstore.dev/api/v1/log/entries/24296fb24b8ad77a..." }, "dependency_provenance_coverage": "67%", "risk_score": 75, "eo_14028_compliance": { "sbom_present": true, "sbom_machine_readable": true, "cryptographic_signing": true, "provenance_attestation": true, "overall": "COMPLIANT" } }
SBOM format selection:
sbom_format=both and artifact is production-criticalVulnerability threshold enforcement:
critical: Block deployment if any CRITICAL severity CVEs foundhigh: Block if CRITICAL or HIGH severity CVEs foundmedium: Block if CRITICAL, HIGH, or MEDIUM severity CVEs foundlow: Report all vulnerabilities but allow deploymentSLSA level requirements:
Abort conditions:
Required fields (all tiers):
sbom_location (string): path to generated SBOM fileformat (string): spdx | cyclonedxcomponent_count (number): total dependencies foundvulnerabilities (array): CVE objects with id, severity, cvss_score, component, fixed_version, descriptioncritical_count (number): count of CRITICAL severity vulnerabilitieshigh_count (number): count of HIGH severity vulnerabilitiescompliance (string): PASS | PASS_WITH_WARNINGS | FAILAdditional fields (T2 only):
provenance (object): SLSA provenance detailssignature_verification (object): cryptographic verification resultsdependency_provenance_coverage (string): percentage of dependencies with provenancerisk_score (number): 0-100 supply chain risk scoreeo_14028_compliance (object): Executive Order compliance detailsOutput formats:
Example: Container image SBOM generation and vulnerability scan
bash# Input: artifact_type=container, location=ghcr.io/org/app:v1.2.3, tier=T1 # Generate SBOM and scan syft ghcr.io/org/app:v1.2.3 -o spdx-json > app-sbom.spdx.json grype sbom:app-sbom.spdx.json --fail-on high # Output { "sbom_location": "app-sbom.spdx.json", "component_count": 312, "vulnerabilities": [{ "cve_id": "CVE-2024-5678", "severity": "CRITICAL", "cvss_score": 9.8, "component": "openssl@1.1.1k", "fixed_version": "1.1.1w" }], "critical_count": 1, "high_count": 3, "compliance": "FAIL" } # See examples/sbom-example.txt for full example
Token budgets (mandatory enforcement):
Safety checks:
Auditability:
Determinism:
Performance targets:
SLSA Framework:
SBOM Standards:
Tools:
Compliance:
Vulnerability Databases:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,382 | 7,447 | -11% | 1 | 1 | 0% | 893 | 4,243 | +375% | 0 | 0 | — |
case-02 | fail→pass | 18,012 | 6,029 | -67% | 1 | 1 | 0% | 3,021 | 4,397 | +46% | 0 | 0 | — |
case-07 | fail→fail | 8,528 | 14,423 | +69% | 1 | 1 | 0% | 868 | 5,585 | +543% | 0 | 0 | — |
case-03 | fail→fail | 4,049 | 14,929 | +269% | 1 | 1 | 0% | 192 | 6,044 | +3048% | 0 | 0 | — |
case-04 | fail→pass | 6,527 | 5,391 | -17% | 1 | 1 | 0% | 1,189 | 4,367 | +267% | 0 | 0 | — |
case-05 | fail→pass | 8,492 | 5,543 | -35% | 1 | 1 | 0% | 304 | 4,568 | +1403% | 0 | 0 | — |
case-06 | fail→pass | 22,797 | 10,814 | -53% | 1 | 1 | 0% | 4,677 | 5,770 | +23% | 0 | 0 | — |
case-20 | fail→pass | 8,676 | 2,127 | -75% | 1 | 1 | 0% | 1,448 | 3,806 | +163% | 0 | 0 | — |
case-08 | fail→pass | 8,389 | 6,043 | -28% | 1 | 1 | 0% | 1,835 | 4,670 | +154% | 0 | 0 | — |
case-09 | pass→pass | 8,093 | 4,115 | -49% | 1 | 1 | 0% | 1,585 | 4,279 | +170% | 0 | 0 | — |
case-10 | fail→pass | 9,677 | 6,748 | -30% | 1 | 1 | 0% | 1,682 | 4,642 | +176% | 0 | 0 | — |
case-11 | pass→pass | 7,686 | 8,920 | +16% | 1 | 1 | 0% | 1,381 | 5,038 | +265% | 0 | 0 | — |
case-12 | pass→pass | 1,483 | 3,754 | +153% | 1 | 1 | 0% | 296 | 4,178 | +1311% | 0 | 0 | — |
case-13 | fail→pass | 13,586 | 3,285 | -76% | 1 | 1 | 0% | 2,960 | 4,128 | +39% | 0 | 0 | — |
case-14 | pass→fail | 6,688 | 4,853 | -27% | 1 | 1 | 0% | 1,367 | 4,441 | +225% | 0 | 0 | — |
case-15 | pass→pass | 9,899 | 7,351 | -26% | 1 | 1 | 0% | 1,858 | 4,939 | +166% | 0 | 0 | — |
case-16 | fail→pass | 7,010 | 2,408 | -66% | 1 | 1 | 0% | 1,152 | 3,852 | +234% | 0 | 0 | — |
case-17 | pass→pass | 5,000 | 4,893 | -2% | 1 | 1 | 0% | 866 | 4,322 | +399% | 0 | 0 | — |
case-18 | pass→pass | 7,582 | 6,045 | -20% | 1 | 1 | 0% | 1,420 | 4,623 | +226% | 0 | 0 | — |
case-19 | pass→pass | 2,924 | 1,962 | -33% | 1 | 1 | 0% | 596 | 3,793 | +536% | 0 | 0 | — |
case-21 | pass→pass | 9,082 | 7,888 | -13% | 1 | 1 | 0% | 1,610 | 4,839 | +201% | 0 | 0 | — |
case-22 | pass→pass | 7,108 | 4,869 | -31% | 1 | 1 | 0% | 1,389 | 4,412 | +218% | 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 +36 percentage points is the difference between those two pass rates over the 20 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.