GDPR (Data Controller)

Map GDPR controller duties for lawful processing, transparency, individual rights, impact assessments and breach handling.

Control framework · Regulated SaaS, Sovereign region

What it checks

Policies in the GDPR (Data Controller) template, in the order the gateway runs them
PolicyOnChecksAction
prompt-injectionRequestDetects jailbreak and prompt-injection signalsBlock
rbacRequestRequires the X-User-ID and X-Org-ID headersBlock
gdpr-complianceRequestRequires the X-GDPR-Consent headerBlock
pii-detectorRequest2 custom identifier patternsRedact
dlp-filterRequest3 custom patternsBlock
quality-scorerResponseAccuracy ≥ 0.80; relevancy ≥ 0.80; combined score ≥ 0.80Block
human-oversightResponseReturns an escalated result instead of the assistant contentEscalate
audit-loggerRequestAdds an allow-only audit marker to the decisionRecord
Policy YAML
pack:
  id: "fw_gdpr_baseline"
  name: "gdpr-baseline"
  version: "0.1.0"
  enabled: true
  author: "verdictan"
  description: "GDPR baseline starter controls for consent-gated processing, redaction, and review."
policies:
  chain:
    - "prompt-injection"
    - "rbac"
    - "gdpr-compliance"
    - "pii-detector"
    - "dlp-filter"
    - "quality-scorer"
    - "human-oversight"
    - "audit-logger"
policy:
  prompt-injection: {}
  rbac:
    deny_if_missing: ["X-User-ID", "X-Org-ID"]
  gdpr-compliance:
    consent_required: true
    consent_header: "X-GDPR-Consent"
    data_categories: ["personal", "sensitive", "biometric"]
  pii-detector:
    action: redact
    detect_patterns: ["(?i)(?:national\\s*id(?:entifier)?|national\\s+insurance\\s+number|ni\\s+number|nid)\\s*[:#-]?\\s*[A-Z0-9-]{5,20}", "(?i)\\b(?:iban)\\s*[:#-]?\\s*[A-Z]{2}\\d{2}[A-Z0-9]{11,30}\\b"]
  dlp-filter:
    detect_patterns: ["(?i)\\bgenetic\\s+(?:data|information|profile|test\\s+results)\\b", "(?i)\\bbiometric\\s+(?:data|template|identifier|record)\\b", "(?i)\\bhealth\\s+(?:data|information|record)\\b"]
    action: block
  quality-scorer:
    thresholds:
      min_aggregate: 0.80
      min_relevancy: 0.80
      min_accuracy: 0.80
  human-oversight:
    action: escalate
  audit-logger: {}