{
  "schema": "varexis-public-evidence-record/0.3",
  "generated_at_utc": "2026-10-18T15:12:44.585Z",
  "record_id": "VXI-EXT-PROCUREMENT-BEDROCK-GUARDRAIL-TAGGING-ENFORCEMENT-SCOPE-01-v1",
  "record_kind": "external_reference",
  "title": "Bedrock: Guardrail coverage depends on tagging and enforced controls",
  "system": "Bedrock",
  "evidence_family": "vendor_evidence_review",
  "index_status": "PUBLIC_SAFE_SOURCE_REVIEW",
  "evidence_state": "PRIMARY_DOCUMENTATION_REVIEW_ONLY",
  "canonical_question": "Does attaching a Bedrock guardrail establish that every part of a prompt was evaluated?",
  "near_questions": [
    "Which Bedrock prompt sections were evaluated by the guardrail?",
    "Can enforced comprehensive guarding override caller tags?",
    "Do XML guardrail tags apply to Converse requests?",
    "What configuration establishes whole-input evaluation?",
    "Does a guardrail ID prove every system message was checked?",
    "What evidence distinguishes selective from comprehensive guarding?"
  ],
  "supported_claims": [
    {
      "claim": "Bedrock supports selective input evaluation, including tagged content for InvokeModel APIs. Separately enforced organisation or account controls can require comprehensive evaluation regardless of caller tags. Coverage depends on the API, content selection and effective enforcement configuration; a guardrail identifier alone is insufficient.",
      "reliance": "DOCUMENTED_CAPABILITY_ONLY"
    }
  ],
  "not_demonstrated_claims": [
    "Runtime enforcement by INDEX, safety certification or successful detection of harmful content",
    "Buyer deployment testing, live runtime enforcement, certification or permission to act",
    "Expanded capabilities or licences for the existing Tool Admission and Audit Coverage executables"
  ],
  "public_summary": "Bedrock supports selective input evaluation, including tagged content for InvokeModel APIs. Separately enforced organisation or account controls can require comprehensive evaluation regardless of caller tags. Coverage depends on the API, content selection and effective enforcement configuration; a guardrail identifier alone is insufficient.",
  "related_record_ids": [],
  "source_review": {
    "source_type": "official_external_documentation",
    "source_title": "Bedrock selective guardrail evaluation",
    "source_url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-tagging.html",
    "checked_on": "2026-10-18",
    "candidate_record_id": "VXI-EXT-PROCUREMENT-BEDROCK-GUARDRAIL-TAGGING-ENFORCEMENT-SCOPE-01-v1",
    "candidate_record_sha256": "15b2b77c82141baf914d5dffc6cb48c564d8b1da49c49977acbe05a4e61f70f4",
    "method": "Primary documentation reviewed for the published bounded statement.",
    "deployment_tested_by_varexis": false,
    "revalidation": "The delivered snapshot preserves the review date shown. Recheck current primary documentation and deployment context before relying on present applicability; payment does not trigger fresh research."
  },
  "answer_policy": {
    "classification": "DOCUMENTED_WITH_DEPLOYMENT_BOUNDARY",
    "bounded_answer": "Bedrock supports selective input evaluation, including tagged content for InvokeModel APIs. Separately enforced organisation or account controls can require comprehensive evaluation regardless of caller tags. Coverage depends on the API, content selection and effective enforcement configuration; a guardrail identifier alone is insufficient.",
    "required_buyer_context": "The API and request content-selection mechanism. The guardrail version and enabled filters. Effective account or organisation selective/comprehensive controls. Buyer deployment not inspected.",
    "missing_evidence": "A broader claim needs evidence matching the intended use. Payment does not upgrade the evidence state."
  },
  "private_boundary": "Raw logs, thresholds, private paths, source code, protected mechanisms and reconstruction-level details are excluded from public and paid delivery.",
  "decision_profile": {
    "schema": "varexis-vendor-evidence-review/1",
    "review_status": "documentation_review_only",
    "checked_on": "2026-10-18",
    "claim_to_check": "Does attaching a Bedrock guardrail establish that every part of a prompt was evaluated?",
    "documented_statement": {
      "status": "documented",
      "statement": "Bedrock supports selective input evaluation, including tagged content for InvokeModel APIs. Separately enforced organisation or account controls can require comprehensive evaluation regardless of caller tags. Coverage depends on the API, content selection and effective enforcement configuration; a guardrail identifier alone is insufficient.",
      "source_ids": [
        "guardrail-tags",
        "guardrail-enforcement"
      ]
    },
    "buyer_evidence_required": [
      "The API and request content-selection mechanism.",
      "The guardrail version and enabled filters.",
      "Effective account or organisation selective/comprehensive controls."
    ],
    "review_condition": {
      "status": "varexis_review_policy",
      "statement": "Do not infer whole-prompt coverage or a tagging bypass without reviewing the applicable enforced configuration."
    },
    "unknowns": "The buyer configuration, contract, effective entitlement and operational evidence have not been inspected. Missing evidence does not establish absence or failure of the feature.",
    "evidence_sources": [
      {
        "source_id": "guardrail-tags",
        "url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-tagging.html",
        "final_url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-tagging.html",
        "title": "Bedrock selective guardrail evaluation",
        "checked_on": "2026-10-18",
        "captured_at_utc": "2026-10-18T15:01:59.939Z",
        "version_context": "Rolling official documentation captured on 2026-10-18; feature and deployment limits apply",
        "sha256": "317b8fe9796c65f317ded4d8cc339b0e69c3e30d9a807ff493e46468aa200f90",
        "bytes": 18671,
        "hash_basis": "Decoded HTTP response body bytes; identifies captured content, not truth or certification",
        "reviewed_sections": [
          "Does attaching a Bedrock guardrail establish that every part of a prompt was evaluated?"
        ]
      },
      {
        "source_id": "guardrail-enforcement",
        "url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-enforcements.html",
        "final_url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-enforcements.html",
        "title": "Bedrock guardrail enforcement",
        "checked_on": "2026-10-18",
        "captured_at_utc": "2026-10-18T15:02:00.622Z",
        "version_context": "Rolling official documentation captured on 2026-10-18; feature and deployment limits apply",
        "sha256": "6ac663b5d854513fe9efe4160a024770e9e329c0c2fbeeb925e4b4c8c66a610f",
        "bytes": 46130,
        "hash_basis": "Decoded HTTP response body bytes; identifies captured content, not truth or certification",
        "reviewed_sections": [
          "Does attaching a Bedrock guardrail establish that every part of a prompt was evaluated?"
        ]
      }
    ],
    "observed_deployment": false,
    "commercial_reliance_boundary": "Procurement evidence support only; no supplier clearance, legal conclusion, configuration audit or verification of the buyer's claim is supplied.",
    "paid_value": {
      "decision_context": "An AI gateway owner must establish the actual evaluated content before relying on a prompt-screening result.",
      "free": "Complete review, practical questions, sources and limitations; may be saved locally.",
      "single_request": "Existing server-saved version-bound route snapshot for later authenticated retrieval.",
      "evidence_receipt": "The same snapshot plus the existing unsigned delivery receipt.",
      "excluded": [
        "Fresh vendor research",
        "Continuous monitoring or change alerts",
        "Testing the buyer configuration",
        "Private evidence or vendor certification"
      ],
      "integrity": "Hashes identify content versions; they do not prove truth or provide a signature."
    },
    "revalidation": {
      "trigger": "A saved purchase does not recheck the source or buyer configuration.",
      "specific_watch": "The API and request content-selection mechanism. The guardrail version and enabled filters. Effective account or organisation selective/comprehensive controls."
    },
    "buyer_relevance": "An AI gateway owner must establish the actual evaluated content before relying on a prompt-screening result.",
    "refusal_conditions": [
      "Do not infer whole-prompt coverage or a tagging bypass without reviewing the applicable enforced configuration."
    ]
  },
  "niche": {
    "niche_id": "agent-authority-action-evidence",
    "label": "Agent Authority & Action Evidence",
    "classification_id": "governance",
    "classification_label": "Agent Governance & Policy Boundaries",
    "topic_id": "governance",
    "buyer_intent": "An AI gateway owner must establish the actual evaluated content before relying on a prompt-screening result.",
    "comparison_role": "supplementary_discovery"
  },
  "paid_access_note": "The full published answer and boundaries are free. Paid access saves the selected route and source versions, with an unsigned receipt when purchased. Preview coverage before paying. Fresh research, bespoke assessment and protected evidence are excluded.",
  "service_route": {
    "request_guide": "/varexis-index/request.html",
    "request_workflow": "/varexis-index/request-workflow.json",
    "service_id": "single_request",
    "service_routes_url": "/varexis-index/service-routes.json",
    "payment_required": true,
    "availability_endpoint": "/api/index/capabilities",
    "operational_status": "AUTOMATIC_SERVICE_CAPABILITIES_GATED",
    "checkout_url": "/index-access"
  },
  "authority_boundary": "VAREXIS-INDEX informs evidence reliance. It does not provide advice, certification, deployment authority, security assurance, production approval, or guarantee.",
  "human_page": "/varexis-index/catalog.html#bedrock-guardrail-tagging-enforcement-scope",
  "machine_readable_record": "/varexis-index/records/bedrock-guardrail-tagging-enforcement-scope.json"
}
