Warehouse Automation

Protect warehouse and interlock data, screen configured terms, check output quality, require review and retain logs.

Industry starter · Regulated SaaS, Private cloud

What it checks

Policies in the Warehouse Automation template, in the order the gateway runs them
PolicyOnChecksAction
prompt-injectionRequestDetects jailbreak and prompt-injection signalsBlock
rbacRequestEnforces role, identity, and data-sensitivity access rulesBlock
pii-detectorRequest3 custom identifier patternsRedact
dlp-filterRequest3 blocked terms such as “facility blueprint” and “safety interlock”Block
safety-filterRequestBlocks or escalates configured dangerous termsBlock
quality-scorerResponseAccuracy ≥ 0.80; relevancy ≥ 0.78; combined score ≥ 0.87; at least 120 characters and 2 sentencesBlock
human-oversightResponseNo action is set, so it does not review responsesNone
audit-loggerRequestAdds an allow-only audit marker to the decisionRecord
Policy YAML
pack:
  id: "cfg_warehouse_automation"
  name: "warehouse-automation"
  version: "0.1.0"
  enabled: true
  author: "verdictan"
  description: "Warehouse controls for automation data, default safety keyword filtering, a configurable human-oversight point, and records."
history:
  capture:
    enabled: true
    mode: metadata_only
    include_blocked: true
    fail_open: false
policies:
  chain:
    - "prompt-injection"
    - "rbac"
    - "pii-detector"
    - "dlp-filter"
    - "safety-filter"
    - "quality-scorer"
    - "human-oversight"
    - "audit-logger"
policy:
  prompt-injection: {}
  rbac: {}
  pii-detector:
    action: redact
    detect_patterns:
      - "(?i)(?:operator|badge)\\s*(?:id|#|number)\\s*[:#-]?\\s*[A-Z0-9-]{4,16}"
      - "(?i)(?:maintenance|hotline|on[-\\s]?call)\\s*(?:phone|line|#)?\\s*[:#-]?\\s*(?:\\+?\\d[\\d\\s().-]{7,}\\d)"
      - "(?i)(?:shift|crew|duty)\\s+roster\\b"
  dlp-filter:
    blocked_terms:
      - "facility blueprint"
      - "safety interlock"
      - "control system password"
    action: block
  safety-filter: {}
  quality-scorer:
    min_output_chars: 120
    min_sentences: 2
    thresholds:
      min_aggregate: 0.87
      min_relevancy: 0.78
      min_accuracy: 0.80
    weights:
      relevancy: 0.5
      accuracy: 0.5
  human-oversight: {}
  audit-logger: {}