{
  "schema_version": 7,
  "benchmark": "invoice-approval-sop-to-validated-draft",
  "generated_at": "2026-08-02T22:35:58.342033+00:00",
  "method": {
    "surface": "Sanctum Chat public API",
    "persona": "surgeon",
    "fixed_model": "opencode-go/qwen3.7-max",
    "max_output_tokens": 32768,
    "write_confirmation": "operator-approved; never automatic",
    "direct_flow_api_calls": false,
    "manual_bpmn_xml_edits": false,
    "document_extraction_required": true,
    "acceptance": "retrieved definition semantics plus Flow lint, 25-gate quality, grammar, and deterministic non-persisted structural preview results",
    "subprocess_outcome_correlation_modeled": false,
    "preview_is_decision_grade_simulation": false
  },
  "runs_requested": 3,
  "runs_attempted": 3,
  "runs_completed": 3,
  "runs_failed": 0,
  "elapsed_seconds": {
    "median": 1228.152,
    "minimum": 852.955,
    "maximum": 1317.586
  },
  "runs": [
    {
      "run": 1,
      "started_at": "2026-08-02T21:39:19.634473+00:00",
      "finished_at": "2026-08-02T21:59:47.788766+00:00",
      "completed_at": "2026-08-02T21:59:47.788766+00:00",
      "elapsed_seconds": 1228.152,
      "completion_marker_seen": true,
      "completion_signal_valid": true,
      "status": "completed",
      "completed": true,
      "failure_code": null,
      "acceptance_checks": {
        "completion_marker_present": true,
        "completion_signal_valid": true,
        "completion_prose_consistent": true,
        "definition_key_reported": true,
        "complete_source_read": true,
        "draft_create_attempted": true,
        "draft_create_confirmation_blocked": true,
        "precreate_semantic_validation_passed": true,
        "operator_approved_write": true,
        "definition_update_not_attempted": true,
        "pending_write_confirmed": true,
        "definition_retrieved": true,
        "lint_completed": true,
        "quality_audit_completed": true,
        "grammar_audit_completed": true,
        "simulation_preview_completed": true,
        "required_read_tools_completed": true,
        "definition_not_published": true,
        "zero_factual_clarifications": true,
        "fixed_model_used": true,
        "independent_semantic_validation_passed": true
      },
      "turn_metrics": [
        {
          "tool_call_count": 9,
          "completed_tool_count": 3,
          "blocked_write_count": 1,
          "failed_tool_count": 5,
          "controlled_execution_completed": false,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": true,
          "precreate_semantic_validation_passed": true,
          "precreate_semantic_failed_check_count": 0
        },
        {
          "tool_call_count": 6,
          "completed_tool_count": 6,
          "blocked_write_count": 0,
          "failed_tool_count": 0,
          "controlled_execution_completed": true,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": false,
          "precreate_semantic_validation_passed": false,
          "precreate_semantic_failed_check_count": -1
        }
      ],
      "clarification_turns": 0,
      "write_approvals": 1,
      "tool_call_count": 15,
      "semantic_validation": {
        "passed": true,
        "checks": {
          "evidence_order_unambiguous": true,
          "source_attachment_evidence_present": true,
          "source_attachment_identity_matches": true,
          "benchmark_source_matches_manifest": true,
          "source_attachment_extraction_complete": true,
          "source_attachment_complete_read": true,
          "source_read_before_definition_write": true,
          "only_attachment_source_used": true,
          "all_definition_reads_bound_to_created_definition": true,
          "controlled_write_chain_evidence_present": true,
          "create_confirmation_chain_valid": true,
          "exactly_one_approved_definition_create": true,
          "no_non_create_definition_mutation_attempted": true,
          "all_writes_explicitly_confirmed": true,
          "definition_write_scope_valid": true,
          "manual_bpmn_xml_not_supplied": true,
          "validation_evidence_after_last_write": true,
          "validation_chain_bound_to_final_definition": true,
          "simulation_preview_result_matches_definition": true,
          "final_get_after_validations": true,
          "no_later_definition_mutation": true,
          "definition_never_published": true,
          "preparation_subprocess_semantics_exact": true,
          "semantic_role_mapping_complete": true,
          "semantic_graph_exact": true,
          "global_bpmn_element_ids_unique": true,
          "top_level_routing_probabilities_exact": true,
          "approval_routing_probabilities_exact": true,
          "timing_parameters_exact": true,
          "sla_parameters_exact": true,
          "required_data_io_exact": true,
          "data_association_inventory_exact": true,
          "review_approval_control_semantics_exact": true,
          "boundary_event_inventory_exact": true,
          "boundary_event_configuration_exact": true,
          "multi_instance_inventory_exact": true,
          "multi_instance_configuration_exact": true,
          "timing_inventory_exact": true,
          "sla_inventory_exact": true,
          "gateway_branch_polarity_exact": true,
          "event_reference_inventory_exact": true,
          "non_gateway_flow_semantics_exact": true,
          "implementation_inventory_exact": true,
          "activity_descriptions_exact": true,
          "supplier_crossings_exact": true,
          "governance_metadata_complete": true,
          "semantic_contradictions_absent": true,
          "definition_evidence_present": true,
          "definition_key_matches": true,
          "definition_name_matches": true,
          "definition_description_absent": true,
          "definition_category_absent": true,
          "definition_remains_draft": true,
          "definition_non_executable": true,
          "definition_uses_simulation_profile": true,
          "source_pool_identity_exact": true,
          "participant_minimum_met": true,
          "external_supplier_present": true,
          "participant_inventory_exact": true,
          "participant_accountability_exact": true,
          "assignee_accountability_exact": true,
          "message_flow_minimum_met": true,
          "supplier_message_flows_complete": true,
          "message_flow_semantics_exact": true,
          "data_object_minimum_met": true,
          "required_data_records_present": true,
          "data_object_inventory_exact": true,
          "data_object_descriptions_absent": true,
          "accountable_lane_minimum_met": true,
          "required_accountable_lanes_present": true,
          "substantial_graph_present": true,
          "required_bpmn_elements_present": true,
          "message_start_and_end_outcomes_present": true,
          "terminal_outcomes_exact": true,
          "interrupting_pt48h_escalation_present": true,
          "sequential_multi_instance_approval_present": true,
          "scope_exclusions_documented": true,
          "forbidden_payment_work_absent": true,
          "excluded_implementation_absent": true,
          "excluded_scope_contradictions_absent": true,
          "gateway_probabilities_valid": true,
          "gateway_defaults_valid": true,
          "quality_audit_evidence_present": true,
          "quality_audit_all_gates_passed": true,
          "quality_audit_includes_gate_14_suite": true,
          "grammar_audit_evidence_present": true,
          "grammar_audit_passed": true,
          "grammar_warnings_clear": true,
          "lint_evidence_present": true,
          "lint_passed": true,
          "lint_simulation_ready": true,
          "simulation_preview_evidence_present": true,
          "simulation_preview_non_persisted": true,
          "simulation_preview_draft_bound": true,
          "simulation_preview_deterministic_config": true,
          "simulation_preview_limitations_disclosed": true,
          "simulation_preview_metrics_finite": true,
          "simulation_preview_step_coverage_complete": true
        },
        "metrics": {
          "source_read_page_count": 2,
          "source_line_count": 139,
          "write_proposal_count": 1,
          "write_resolution_count": 1,
          "approved_write_count": 1,
          "successful_definition_write_count": 1,
          "successful_definition_update_count": 0,
          "bound_validation_count": 4,
          "participant_count": 2,
          "message_flow_count": 4,
          "supplier_message_flow_count": 4,
          "data_object_count": 6,
          "accountable_lane_count": 4,
          "top_level_step_count": 18,
          "total_step_count": 32,
          "terminal_outcome_count": 4,
          "exclusive_gateway_count": 7,
          "converging_gateway_count": 3,
          "quality_passed_gates": 25,
          "quality_failed_gates": 0,
          "quality_total_gates": 25,
          "grammar_error_count": 0,
          "grammar_warning_count": 0,
          "lint_error_count": 0,
          "lint_warning_count": 3,
          "simulation_preview_run_count": 1000,
          "simulation_preview_step_count": 31,
          "failed_check_count": 0
        }
      }
    },
    {
      "run": 2,
      "started_at": "2026-08-02T21:59:47.789958+00:00",
      "finished_at": "2026-08-02T22:21:45.376740+00:00",
      "completed_at": "2026-08-02T22:21:45.376740+00:00",
      "elapsed_seconds": 1317.586,
      "completion_marker_seen": true,
      "completion_signal_valid": true,
      "status": "completed",
      "completed": true,
      "failure_code": null,
      "acceptance_checks": {
        "completion_marker_present": true,
        "completion_signal_valid": true,
        "completion_prose_consistent": true,
        "definition_key_reported": true,
        "complete_source_read": true,
        "draft_create_attempted": true,
        "draft_create_confirmation_blocked": true,
        "precreate_semantic_validation_passed": true,
        "operator_approved_write": true,
        "definition_update_not_attempted": true,
        "pending_write_confirmed": true,
        "definition_retrieved": true,
        "lint_completed": true,
        "quality_audit_completed": true,
        "grammar_audit_completed": true,
        "simulation_preview_completed": true,
        "required_read_tools_completed": true,
        "definition_not_published": true,
        "zero_factual_clarifications": true,
        "fixed_model_used": true,
        "independent_semantic_validation_passed": true
      },
      "turn_metrics": [
        {
          "tool_call_count": 12,
          "completed_tool_count": 3,
          "blocked_write_count": 1,
          "failed_tool_count": 8,
          "controlled_execution_completed": false,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": true,
          "precreate_semantic_validation_passed": true,
          "precreate_semantic_failed_check_count": 0
        },
        {
          "tool_call_count": 6,
          "completed_tool_count": 6,
          "blocked_write_count": 0,
          "failed_tool_count": 0,
          "controlled_execution_completed": true,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": false,
          "precreate_semantic_validation_passed": false,
          "precreate_semantic_failed_check_count": -1
        }
      ],
      "clarification_turns": 0,
      "write_approvals": 1,
      "tool_call_count": 18,
      "semantic_validation": {
        "passed": true,
        "checks": {
          "evidence_order_unambiguous": true,
          "source_attachment_evidence_present": true,
          "source_attachment_identity_matches": true,
          "benchmark_source_matches_manifest": true,
          "source_attachment_extraction_complete": true,
          "source_attachment_complete_read": true,
          "source_read_before_definition_write": true,
          "only_attachment_source_used": true,
          "all_definition_reads_bound_to_created_definition": true,
          "controlled_write_chain_evidence_present": true,
          "create_confirmation_chain_valid": true,
          "exactly_one_approved_definition_create": true,
          "no_non_create_definition_mutation_attempted": true,
          "all_writes_explicitly_confirmed": true,
          "definition_write_scope_valid": true,
          "manual_bpmn_xml_not_supplied": true,
          "validation_evidence_after_last_write": true,
          "validation_chain_bound_to_final_definition": true,
          "simulation_preview_result_matches_definition": true,
          "final_get_after_validations": true,
          "no_later_definition_mutation": true,
          "definition_never_published": true,
          "preparation_subprocess_semantics_exact": true,
          "semantic_role_mapping_complete": true,
          "semantic_graph_exact": true,
          "global_bpmn_element_ids_unique": true,
          "top_level_routing_probabilities_exact": true,
          "approval_routing_probabilities_exact": true,
          "timing_parameters_exact": true,
          "sla_parameters_exact": true,
          "required_data_io_exact": true,
          "data_association_inventory_exact": true,
          "review_approval_control_semantics_exact": true,
          "boundary_event_inventory_exact": true,
          "boundary_event_configuration_exact": true,
          "multi_instance_inventory_exact": true,
          "multi_instance_configuration_exact": true,
          "timing_inventory_exact": true,
          "sla_inventory_exact": true,
          "gateway_branch_polarity_exact": true,
          "event_reference_inventory_exact": true,
          "non_gateway_flow_semantics_exact": true,
          "implementation_inventory_exact": true,
          "activity_descriptions_exact": true,
          "supplier_crossings_exact": true,
          "governance_metadata_complete": true,
          "semantic_contradictions_absent": true,
          "definition_evidence_present": true,
          "definition_key_matches": true,
          "definition_name_matches": true,
          "definition_description_absent": true,
          "definition_category_absent": true,
          "definition_remains_draft": true,
          "definition_non_executable": true,
          "definition_uses_simulation_profile": true,
          "source_pool_identity_exact": true,
          "participant_minimum_met": true,
          "external_supplier_present": true,
          "participant_inventory_exact": true,
          "participant_accountability_exact": true,
          "assignee_accountability_exact": true,
          "message_flow_minimum_met": true,
          "supplier_message_flows_complete": true,
          "message_flow_semantics_exact": true,
          "data_object_minimum_met": true,
          "required_data_records_present": true,
          "data_object_inventory_exact": true,
          "data_object_descriptions_absent": true,
          "accountable_lane_minimum_met": true,
          "required_accountable_lanes_present": true,
          "substantial_graph_present": true,
          "required_bpmn_elements_present": true,
          "message_start_and_end_outcomes_present": true,
          "terminal_outcomes_exact": true,
          "interrupting_pt48h_escalation_present": true,
          "sequential_multi_instance_approval_present": true,
          "scope_exclusions_documented": true,
          "forbidden_payment_work_absent": true,
          "excluded_implementation_absent": true,
          "excluded_scope_contradictions_absent": true,
          "gateway_probabilities_valid": true,
          "gateway_defaults_valid": true,
          "quality_audit_evidence_present": true,
          "quality_audit_all_gates_passed": true,
          "quality_audit_includes_gate_14_suite": true,
          "grammar_audit_evidence_present": true,
          "grammar_audit_passed": true,
          "grammar_warnings_clear": true,
          "lint_evidence_present": true,
          "lint_passed": true,
          "lint_simulation_ready": true,
          "simulation_preview_evidence_present": true,
          "simulation_preview_non_persisted": true,
          "simulation_preview_draft_bound": true,
          "simulation_preview_deterministic_config": true,
          "simulation_preview_limitations_disclosed": true,
          "simulation_preview_metrics_finite": true,
          "simulation_preview_step_coverage_complete": true
        },
        "metrics": {
          "source_read_page_count": 2,
          "source_line_count": 139,
          "write_proposal_count": 1,
          "write_resolution_count": 1,
          "approved_write_count": 1,
          "successful_definition_write_count": 1,
          "successful_definition_update_count": 0,
          "bound_validation_count": 4,
          "participant_count": 2,
          "message_flow_count": 4,
          "supplier_message_flow_count": 4,
          "data_object_count": 6,
          "accountable_lane_count": 4,
          "top_level_step_count": 18,
          "total_step_count": 32,
          "terminal_outcome_count": 4,
          "exclusive_gateway_count": 7,
          "converging_gateway_count": 3,
          "quality_passed_gates": 25,
          "quality_failed_gates": 0,
          "quality_total_gates": 25,
          "grammar_error_count": 0,
          "grammar_warning_count": 0,
          "lint_error_count": 0,
          "lint_warning_count": 3,
          "simulation_preview_run_count": 1000,
          "simulation_preview_step_count": 31,
          "failed_check_count": 0
        }
      }
    },
    {
      "run": 3,
      "started_at": "2026-08-02T22:21:45.378380+00:00",
      "finished_at": "2026-08-02T22:35:58.336534+00:00",
      "completed_at": "2026-08-02T22:35:58.336534+00:00",
      "elapsed_seconds": 852.955,
      "completion_marker_seen": true,
      "completion_signal_valid": true,
      "status": "completed",
      "completed": true,
      "failure_code": null,
      "acceptance_checks": {
        "completion_marker_present": true,
        "completion_signal_valid": true,
        "completion_prose_consistent": true,
        "definition_key_reported": true,
        "complete_source_read": true,
        "draft_create_attempted": true,
        "draft_create_confirmation_blocked": true,
        "precreate_semantic_validation_passed": true,
        "operator_approved_write": true,
        "definition_update_not_attempted": true,
        "pending_write_confirmed": true,
        "definition_retrieved": true,
        "lint_completed": true,
        "quality_audit_completed": true,
        "grammar_audit_completed": true,
        "simulation_preview_completed": true,
        "required_read_tools_completed": true,
        "definition_not_published": true,
        "zero_factual_clarifications": true,
        "fixed_model_used": true,
        "independent_semantic_validation_passed": true
      },
      "turn_metrics": [
        {
          "tool_call_count": 8,
          "completed_tool_count": 3,
          "blocked_write_count": 1,
          "failed_tool_count": 4,
          "controlled_execution_completed": false,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": true,
          "precreate_semantic_validation_passed": true,
          "precreate_semantic_failed_check_count": 0
        },
        {
          "tool_call_count": 6,
          "completed_tool_count": 6,
          "blocked_write_count": 0,
          "failed_tool_count": 0,
          "controlled_execution_completed": true,
          "controlled_execution_declined": false,
          "precreate_semantic_validation_present": false,
          "precreate_semantic_validation_passed": false,
          "precreate_semantic_failed_check_count": -1
        }
      ],
      "clarification_turns": 0,
      "write_approvals": 1,
      "tool_call_count": 14,
      "semantic_validation": {
        "passed": true,
        "checks": {
          "evidence_order_unambiguous": true,
          "source_attachment_evidence_present": true,
          "source_attachment_identity_matches": true,
          "benchmark_source_matches_manifest": true,
          "source_attachment_extraction_complete": true,
          "source_attachment_complete_read": true,
          "source_read_before_definition_write": true,
          "only_attachment_source_used": true,
          "all_definition_reads_bound_to_created_definition": true,
          "controlled_write_chain_evidence_present": true,
          "create_confirmation_chain_valid": true,
          "exactly_one_approved_definition_create": true,
          "no_non_create_definition_mutation_attempted": true,
          "all_writes_explicitly_confirmed": true,
          "definition_write_scope_valid": true,
          "manual_bpmn_xml_not_supplied": true,
          "validation_evidence_after_last_write": true,
          "validation_chain_bound_to_final_definition": true,
          "simulation_preview_result_matches_definition": true,
          "final_get_after_validations": true,
          "no_later_definition_mutation": true,
          "definition_never_published": true,
          "preparation_subprocess_semantics_exact": true,
          "semantic_role_mapping_complete": true,
          "semantic_graph_exact": true,
          "global_bpmn_element_ids_unique": true,
          "top_level_routing_probabilities_exact": true,
          "approval_routing_probabilities_exact": true,
          "timing_parameters_exact": true,
          "sla_parameters_exact": true,
          "required_data_io_exact": true,
          "data_association_inventory_exact": true,
          "review_approval_control_semantics_exact": true,
          "boundary_event_inventory_exact": true,
          "boundary_event_configuration_exact": true,
          "multi_instance_inventory_exact": true,
          "multi_instance_configuration_exact": true,
          "timing_inventory_exact": true,
          "sla_inventory_exact": true,
          "gateway_branch_polarity_exact": true,
          "event_reference_inventory_exact": true,
          "non_gateway_flow_semantics_exact": true,
          "implementation_inventory_exact": true,
          "activity_descriptions_exact": true,
          "supplier_crossings_exact": true,
          "governance_metadata_complete": true,
          "semantic_contradictions_absent": true,
          "definition_evidence_present": true,
          "definition_key_matches": true,
          "definition_name_matches": true,
          "definition_description_absent": true,
          "definition_category_absent": true,
          "definition_remains_draft": true,
          "definition_non_executable": true,
          "definition_uses_simulation_profile": true,
          "source_pool_identity_exact": true,
          "participant_minimum_met": true,
          "external_supplier_present": true,
          "participant_inventory_exact": true,
          "participant_accountability_exact": true,
          "assignee_accountability_exact": true,
          "message_flow_minimum_met": true,
          "supplier_message_flows_complete": true,
          "message_flow_semantics_exact": true,
          "data_object_minimum_met": true,
          "required_data_records_present": true,
          "data_object_inventory_exact": true,
          "data_object_descriptions_absent": true,
          "accountable_lane_minimum_met": true,
          "required_accountable_lanes_present": true,
          "substantial_graph_present": true,
          "required_bpmn_elements_present": true,
          "message_start_and_end_outcomes_present": true,
          "terminal_outcomes_exact": true,
          "interrupting_pt48h_escalation_present": true,
          "sequential_multi_instance_approval_present": true,
          "scope_exclusions_documented": true,
          "forbidden_payment_work_absent": true,
          "excluded_implementation_absent": true,
          "excluded_scope_contradictions_absent": true,
          "gateway_probabilities_valid": true,
          "gateway_defaults_valid": true,
          "quality_audit_evidence_present": true,
          "quality_audit_all_gates_passed": true,
          "quality_audit_includes_gate_14_suite": true,
          "grammar_audit_evidence_present": true,
          "grammar_audit_passed": true,
          "grammar_warnings_clear": true,
          "lint_evidence_present": true,
          "lint_passed": true,
          "lint_simulation_ready": true,
          "simulation_preview_evidence_present": true,
          "simulation_preview_non_persisted": true,
          "simulation_preview_draft_bound": true,
          "simulation_preview_deterministic_config": true,
          "simulation_preview_limitations_disclosed": true,
          "simulation_preview_metrics_finite": true,
          "simulation_preview_step_coverage_complete": true
        },
        "metrics": {
          "source_read_page_count": 2,
          "source_line_count": 139,
          "write_proposal_count": 1,
          "write_resolution_count": 1,
          "approved_write_count": 1,
          "successful_definition_write_count": 1,
          "successful_definition_update_count": 0,
          "bound_validation_count": 4,
          "participant_count": 2,
          "message_flow_count": 4,
          "supplier_message_flow_count": 4,
          "data_object_count": 6,
          "accountable_lane_count": 4,
          "top_level_step_count": 18,
          "total_step_count": 32,
          "terminal_outcome_count": 4,
          "exclusive_gateway_count": 7,
          "converging_gateway_count": 3,
          "quality_passed_gates": 25,
          "quality_failed_gates": 0,
          "quality_total_gates": 25,
          "grammar_error_count": 0,
          "grammar_warning_count": 0,
          "lint_error_count": 0,
          "lint_warning_count": 3,
          "simulation_preview_run_count": 1000,
          "simulation_preview_step_count": 31,
          "failed_check_count": 0
        }
      }
    }
  ],
  "publication_note": "Fixed model and fixed-name checks, integer metrics, timings, and aggregate turn counts only. Assistant transcript, failure details, pending execution IDs, definition IDs or keys, source hashes, thread IDs, trace IDs, tool parameters, schemas, proposal summaries, findings, reasoning, credentials, and tenant data are intentionally excluded. Pre-create semantic evidence is reduced to fixed booleans and integer counts. The preview verifies deterministic plumbing only; embedded-subprocess outcome correlation is not yet modeled and the result is not decision-grade operational simulation."
}
