{
  "schema": "varexis-public-evidence-record/0.3",
  "generated_at_utc": "2026-10-18T19:34:16.969Z",
  "record_id": "VXI-EXT-PROCUREMENT-OPENAI-FUNCTION-CALL-APPLICATION-EXECUTION-EVIDENCE-01-v1",
  "record_kind": "external_reference",
  "title": "OpenAI function calling: A requested action needs an execution result",
  "system": "OpenAI",
  "evidence_family": "vendor_evidence_review",
  "index_status": "PUBLIC_SAFE_SOURCE_REVIEW",
  "evidence_state": "PRIMARY_DOCUMENTATION_REVIEW_ONLY",
  "canonical_question": "Does a model-generated OpenAI function call prove that the external action happened?",
  "near_questions": [
    "Is an OpenAI function call output proof of a completed business action?",
    "Who executes a custom function requested by an OpenAI model?",
    "What joins an OpenAI tool call to the application's result?",
    "Can a model's final reply replace an external system completion receipt?"
  ],
  "supported_claims": [
    {
      "claim": "OpenAI's function-calling workflow returns a tool call to the application, which executes its code and supplies the tool output in a subsequent request. The model can then produce a final response or further calls. The documented application-executed function flow separates the model's requested operation from the application-supplied result.",
      "reliance": "DOCUMENTED_CAPABILITY_ONLY"
    }
  ],
  "not_demonstrated_claims": [
    "Observed execution, authentic external results or coverage of hosted built-in tool execution",
    "Buyer deployment testing, live runtime enforcement, certification or permission to act",
    "Expanded capabilities or licences for existing INDEX executables"
  ],
  "public_summary": "OpenAI's function-calling workflow returns a tool call to the application, which executes its code and supplies the tool output in a subsequent request. The model can then produce a final response or further calls. The documented application-executed function flow separates the model's requested operation from the application-supplied result.",
  "related_record_ids": [],
  "source_review": {
    "source_type": "official_external_documentation",
    "source_title": "\n  Function calling | OpenAI API\n",
    "source_url": "https://developers.openai.com/api/docs/guides/function-calling",
    "checked_on": "2026-10-18",
    "candidate_record_id": "VXI-EXT-PROCUREMENT-OPENAI-FUNCTION-CALL-APPLICATION-EXECUTION-EVIDENCE-01-v1",
    "candidate_record_sha256": "a00a6b853f74e8a8ae5f0bf668006ecfe794f6c4d6481557ab13105fd730c49f",
    "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": "OpenAI's function-calling workflow returns a tool call to the application, which executes its code and supplies the tool output in a subsequent request. The model can then produce a final response or further calls. The documented application-executed function flow separates the model's requested operation from the application-supplied result.",
    "required_buyer_context": "Function name, validated arguments and tool-call identifier. Application execution record and authoritative external-system result. Linked tool output, failures and any subsequent model response. 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 a model-generated OpenAI function call prove that the external action happened?",
    "documented_statement": {
      "status": "documented",
      "statement": "OpenAI's function-calling workflow returns a tool call to the application, which executes its code and supplies the tool output in a subsequent request. The model can then produce a final response or further calls. The documented application-executed function flow separates the model's requested operation from the application-supplied result.",
      "source_ids": [
        "openai-function"
      ]
    },
    "review_inference": {
      "status": "review_inference",
      "statement": "Neither generated arguments nor a narrative reply establish external state without execution evidence.",
      "source_ids": [
        "openai-function"
      ]
    },
    "buyer_evidence_required": [
      "Function name, validated arguments and tool-call identifier.",
      "Application execution record and authoritative external-system result.",
      "Linked tool output, failures and any subsequent model response."
    ],
    "review_condition": {
      "status": "varexis_review_policy",
      "statement": "Refuse a completed-action claim when only model output or an unbound tool result is supplied."
    },
    "unknowns": "Application implementation, result authenticity and external side effects are unobserved.",
    "unknown_finding": {
      "status": "unknown",
      "statement": "Application implementation, result authenticity and external side effects are unobserved."
    },
    "evidence_sources": [
      {
        "source_id": "openai-function",
        "url": "https://developers.openai.com/api/docs/guides/function-calling",
        "final_url": "https://developers.openai.com/api/docs/guides/function-calling",
        "title": "\n  Function calling | OpenAI API\n",
        "checked_on": "2026-10-18",
        "captured_at_utc": "2026-10-18T19:20:53.029Z",
        "version_context": "Rolling official documentation captured 2026-10-18; exact response hash retained, no pinned release implied.",
        "sha256": "1a3f492cd442ee997e445d557babfb30e476f36ee32edf36e5e7d505b279343c",
        "bytes": 1478838,
        "hash_basis": "Decoded HTTP response body bytes; identifies captured content, not truth or certification",
        "reviewed_sections": [
          "Function calling flow",
          "Handling function calls"
        ],
        "capture_status": 200,
        "rights_review": "Factual paraphrase and source metadata only; response body retained privately and not redistributed."
      }
    ],
    "observed_deployment": false,
    "commercial_reliance_boundary": "Documentation review only; no supplier clearance, configuration audit, observed execution or certification of the buyer's claim.",
    "paid_value": {
      "decision_context": "Bind an action request to the outcome evidence needed for a business decision.",
      "free": "Complete review, sources, practical questions 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": "Recheck API flow and tool type; hosted built-in tools are outside this custom-function review."
    },
    "buyer_relevance": "Bind an action request to the outcome evidence needed for a business decision.",
    "refusal_conditions": [
      "Refuse a completed-action claim when only model output or an unbound tool result is supplied."
    ]
  },
  "niche": {
    "niche_id": "agent-authority-action-evidence",
    "label": "Agent Authority & Action Evidence",
    "classification_id": "audit",
    "classification_label": "Audit, Provenance & Evidence Accountability",
    "topic_id": "audit",
    "buyer_intent": "Bind an action request to the outcome evidence needed for a business decision.",
    "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#openai-function-call-application-execution-evidence",
  "machine_readable_record": "/varexis-index/records/openai-function-call-application-execution-evidence.json"
}
