{
  "schemaVersion": "b18.sft-context-resolver.r1",
  "generatedAt": "2026-08-21T03:30:00.000Z",
  "implementation": {
    "name": "Glyphd SFT Context Resolver",
    "version": "0.2.0",
    "benchmarkVersion": "B18-R1",
    "benchmarkClass": "deterministic operation-directed context comparison",
    "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
    "latencySamplesPerCaseStrategy": 25
  },
  "cases": [
    {
      "id": "architecture-fit",
      "scenarioId": "architecture",
      "query": "Can the lobby sofa fit through the front door?",
      "expectedSolverStatus": "PASS",
      "requiredEntityIds": [
        "sofa",
        "front-door"
      ],
      "requiredRelationIds": [],
      "requiredMetricFactIds": [
        "entity:sofa:bounds",
        "entity:front-door:bounds"
      ],
      "typedCommandId": "sha256:28f67bf8ae2199a8d9fc904e5592e8aab948cbf9f2c62f039fe0e2c971c412f6",
      "sceneRevisionId": "sha256:71880593b846615ca053fbdbfa272d37768a7f1ba74e9187f4b4dabc8b9a8248",
      "sceneContentId": "sha256:58a75ea1d736e5c3922fd123a48b5559106b3c98b950e3550c9862273fb8f6c2",
      "fullSceneStatus": "PASS",
      "sftOperationalStatus": "PASS",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 13,
          "contextRelationCount": 17,
          "serializedBytes": 7870,
          "serializedTokens": 2157,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 10,
          "irrelevantRelationCount": 17,
          "irrelevantEntityRate": 0.8333333333333334,
          "irrelevantRelationRate": 1,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.466375,
            "p50": 0.560417,
            "p95": 0.765583,
            "max": 0.955,
            "mean": 0.579163
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 8,
          "contextRelationCount": 8,
          "serializedBytes": 4280,
          "serializedTokens": 1192,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 5,
          "irrelevantRelationCount": 8,
          "irrelevantEntityRate": 0.7142857142857143,
          "irrelevantRelationRate": 1,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.484208,
            "p50": 0.515834,
            "p95": 0.726959,
            "max": 0.753708,
            "mean": 0.55817
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 8,
          "contextRelationCount": 10,
          "serializedBytes": 4762,
          "serializedTokens": 1319,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 5,
          "irrelevantRelationCount": 10,
          "irrelevantEntityRate": 0.7142857142857143,
          "irrelevantRelationRate": 1,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.489541,
            "p50": 0.524166,
            "p95": 0.701666,
            "max": 0.754666,
            "mean": 0.544972
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 6,
          "contextRelationCount": 5,
          "serializedBytes": 3047,
          "serializedTokens": 853,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 3,
          "irrelevantRelationCount": 5,
          "irrelevantEntityRate": 0.6,
          "irrelevantRelationRate": 1,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.374666,
            "p50": 0.408167,
            "p95": 0.562959,
            "max": 0.597458,
            "mean": 0.423927
          }
        }
      ]
    },
    {
      "id": "architecture-service-access",
      "scenarioId": "architecture",
      "query": "Audit service access for air handler",
      "expectedSolverStatus": "FAIL",
      "requiredEntityIds": [
        "air-handler",
        "air-handler-access",
        "stored-crate"
      ],
      "requiredRelationIds": [
        "services:air-handler:zone"
      ],
      "requiredMetricFactIds": [
        "entity:air-handler-access:bounds",
        "entity:stored-crate:bounds"
      ],
      "typedCommandId": "sha256:5f53ec0738df83644545d96337668845b1efa04a5f0bee9ac8f7290e4589e875",
      "sceneRevisionId": "sha256:71880593b846615ca053fbdbfa272d37768a7f1ba74e9187f4b4dabc8b9a8248",
      "sceneContentId": "sha256:58a75ea1d736e5c3922fd123a48b5559106b3c98b950e3550c9862273fb8f6c2",
      "fullSceneStatus": "FAIL",
      "sftOperationalStatus": "FAIL",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 13,
          "contextRelationCount": 17,
          "serializedBytes": 7870,
          "serializedTokens": 2157,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 9,
          "irrelevantRelationCount": 16,
          "irrelevantEntityRate": 0.75,
          "irrelevantRelationRate": 0.9411764705882353,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.455,
            "p50": 0.508834,
            "p95": 0.76925,
            "max": 0.850292,
            "mean": 0.544944
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 6,
          "contextRelationCount": 7,
          "serializedBytes": 3627,
          "serializedTokens": 981,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.6666666666666666,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 0.5,
          "irrelevantEntityCount": 3,
          "irrelevantRelationCount": 6,
          "irrelevantEntityRate": 0.6,
          "irrelevantRelationRate": 0.8571428571428571,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": false,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.422,
            "p50": 0.445583,
            "p95": 0.545083,
            "max": 0.698708,
            "mean": 0.460248
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 11,
          "contextRelationCount": 14,
          "serializedBytes": 6608,
          "serializedTokens": 1809,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 7,
          "irrelevantRelationCount": 13,
          "irrelevantEntityRate": 0.7,
          "irrelevantRelationRate": 0.9285714285714286,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.612209,
            "p50": 0.662041,
            "p95": 0.85375,
            "max": 0.882209,
            "mean": 0.681545
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 7,
          "contextRelationCount": 7,
          "serializedBytes": 3948,
          "serializedTokens": 1076,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 3,
          "irrelevantRelationCount": 6,
          "irrelevantEntityRate": 0.5,
          "irrelevantRelationRate": 0.8571428571428571,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.441875,
            "p50": 0.495125,
            "p95": 0.871125,
            "max": 0.979458,
            "mean": 0.568067
          }
        }
      ]
    },
    {
      "id": "architecture-clearance-policy",
      "scenarioId": "architecture",
      "query": "Check architectural clearance in ground floor",
      "expectedSolverStatus": "FAIL",
      "requiredEntityIds": [
        "ground-floor",
        "front-door",
        "mechanical-door",
        "lobby",
        "corridor",
        "mechanical-room"
      ],
      "requiredRelationIds": [
        "lobby-front-door-corridor",
        "corridor-mechanical-door-room"
      ],
      "requiredMetricFactIds": [
        "entity:front-door:bounds",
        "entity:mechanical-door:bounds",
        "entity:lobby:attributes.routeWidthMm",
        "entity:corridor:attributes.routeWidthMm",
        "entity:mechanical-room:attributes.routeWidthMm",
        "policy:fixture-project-policy:minimumOpeningWidth",
        "policy:fixture-project-policy:minimumHeadroom",
        "policy:fixture-project-policy:minimumRouteWidth"
      ],
      "typedCommandId": "sha256:f1629ece7acfc0232865cf2931a0cbc1ab918047aa8957adcc6d26050cc56f27",
      "sceneRevisionId": "sha256:71880593b846615ca053fbdbfa272d37768a7f1ba74e9187f4b4dabc8b9a8248",
      "sceneContentId": "sha256:58a75ea1d736e5c3922fd123a48b5559106b3c98b950e3550c9862273fb8f6c2",
      "fullSceneStatus": "FAIL",
      "sftOperationalStatus": "FAIL",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 13,
          "contextRelationCount": 17,
          "serializedBytes": 7870,
          "serializedTokens": 2157,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 6,
          "irrelevantRelationCount": 15,
          "irrelevantEntityRate": 0.5,
          "irrelevantRelationRate": 0.8823529411764706,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.521167,
            "p50": 0.561708,
            "p95": 0.837083,
            "max": 0.844375,
            "mean": 0.60607
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 6,
          "contextRelationCount": 6,
          "serializedBytes": 3444,
          "serializedTokens": 937,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.5,
          "requiredRelationRecall": 0.5,
          "requiredMetricFactRecall": 0.625,
          "irrelevantEntityCount": 2,
          "irrelevantRelationCount": 5,
          "irrelevantEntityRate": 0.4,
          "irrelevantRelationRate": 0.8333333333333334,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": false,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.487084,
            "p50": 0.50475,
            "p95": 0.547958,
            "max": 0.8035,
            "mean": 0.519442
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 13,
          "contextRelationCount": 17,
          "serializedBytes": 7834,
          "serializedTokens": 2151,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 6,
          "irrelevantRelationCount": 15,
          "irrelevantEntityRate": 0.5,
          "irrelevantRelationRate": 0.8823529411764706,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.732,
            "p50": 0.791083,
            "p95": 0.961458,
            "max": 1.03275,
            "mean": 0.809053
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 13,
          "contextRelationCount": 14,
          "serializedBytes": 7230,
          "serializedTokens": 1997,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 6,
          "irrelevantRelationCount": 12,
          "irrelevantEntityRate": 0.5,
          "irrelevantRelationRate": 0.8571428571428571,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.694542,
            "p50": 0.728959,
            "p95": 0.923208,
            "max": 0.937959,
            "mean": 0.755148
          }
        }
      ]
    },
    {
      "id": "assembly-order",
      "scenarioId": "assembly",
      "query": "Audit assembly order for cabinet assembly",
      "expectedSolverStatus": "PASS",
      "requiredEntityIds": [
        "cabinet",
        "cabinet-base",
        "left-side",
        "right-side",
        "fixed-shelf",
        "cabinet-top",
        "cabinet-door"
      ],
      "requiredRelationIds": [
        "order:base:left",
        "order:base:right",
        "order:left:shelf",
        "order:right:shelf",
        "order:shelf:top",
        "order:top:door"
      ],
      "requiredMetricFactIds": [],
      "typedCommandId": "sha256:fb375e31bac6a9afd0769700e6495c09f402b01b963255108f04080b0398f2ed",
      "sceneRevisionId": "sha256:e6fee0ee7a7ad9484b7113da27d2f68a16f4aa79258217ff85e8643992099c2a",
      "sceneContentId": "sha256:e7dfa157b713fcffee9c9f039fc46d2ab565fd1d03f214d63c56d808a353cce6",
      "fullSceneStatus": "PASS",
      "sftOperationalStatus": "PASS",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 8,
          "contextRelationCount": 14,
          "serializedBytes": 5125,
          "serializedTokens": 1417,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 8,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.316209,
            "p50": 0.343625,
            "p95": 0.52225,
            "max": 0.602125,
            "mean": 0.370413
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 5,
          "contextRelationCount": 5,
          "serializedBytes": 2575,
          "serializedTokens": 722,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.5714285714285714,
          "requiredRelationRecall": 0.16666666666666666,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 4,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.8,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": false,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.302833,
            "p50": 0.320625,
            "p95": 0.403791,
            "max": 0.57075,
            "mean": 0.34113
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 8,
          "contextRelationCount": 14,
          "serializedBytes": 5121,
          "serializedTokens": 1419,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 8,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.46025,
            "p50": 0.48425,
            "p95": 0.714292,
            "max": 0.762667,
            "mean": 0.506807
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 8,
          "contextRelationCount": 14,
          "serializedBytes": 5108,
          "serializedTokens": 1417,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 8,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.455125,
            "p50": 0.48275,
            "p95": 1.214084,
            "max": 5.152083,
            "mean": 0.746848
          }
        }
      ]
    },
    {
      "id": "robot-alternate-route",
      "scenarioId": "robot",
      "query": "Find a path from mobile robot to charging dock",
      "expectedSolverStatus": "PASS",
      "requiredEntityIds": [
        "mobile-robot",
        "robot-start",
        "junction",
        "wide-aisle",
        "charging-dock"
      ],
      "requiredRelationIds": [
        "robot-at-start",
        "start-junction",
        "junction-wide",
        "wide-dock"
      ],
      "requiredMetricFactIds": [],
      "typedCommandId": "sha256:9acc2bd0387e70385dc72161e842ec2bcd57adfe18646902810f4971dda74c4b",
      "sceneRevisionId": "sha256:38b507c89cd9c02af209c426fa54660816bbb7ffabcbbd254e491e3f62f688ee",
      "sceneContentId": "sha256:3f8a676e211c1237db593d8d3f72963ad59dc2e365b472c0547f5ae8378c2aaa",
      "fullSceneStatus": "PASS",
      "sftOperationalStatus": "PASS",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 8,
          "contextRelationCount": 13,
          "serializedBytes": 5420,
          "serializedTokens": 1506,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 2,
          "irrelevantRelationCount": 9,
          "irrelevantEntityRate": 0.2857142857142857,
          "irrelevantRelationRate": 0.6923076923076923,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.320084,
            "p50": 0.343083,
            "p95": 0.416958,
            "max": 0.574458,
            "mean": 0.356648
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 6,
          "contextRelationCount": 9,
          "serializedBytes": 3983,
          "serializedTokens": 1103,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.8,
          "requiredRelationRecall": 0.5,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 1,
          "irrelevantRelationCount": 7,
          "irrelevantEntityRate": 0.2,
          "irrelevantRelationRate": 0.7777777777777778,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": false,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.396666,
            "p50": 0.427958,
            "p95": 0.517042,
            "max": 0.581208,
            "mean": 0.437808
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 8,
          "contextRelationCount": 13,
          "serializedBytes": 5398,
          "serializedTokens": 1501,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 2,
          "irrelevantRelationCount": 9,
          "irrelevantEntityRate": 0.2857142857142857,
          "irrelevantRelationRate": 0.6923076923076923,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.477958,
            "p50": 0.515667,
            "p95": 0.750542,
            "max": 0.789208,
            "mean": 0.542288
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 7,
          "contextRelationCount": 12,
          "serializedBytes": 4894,
          "serializedTokens": 1360,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 1,
          "irrelevantRelationCount": 8,
          "irrelevantEntityRate": 0.16666666666666666,
          "irrelevantRelationRate": 0.6666666666666666,
          "solverStatus": "PASS",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.442875,
            "p50": 0.461167,
            "p95": 0.572958,
            "max": 0.719792,
            "mean": 0.476233
          }
        }
      ]
    },
    {
      "id": "game-locked-exit",
      "scenarioId": "game",
      "query": "Find a path from player to exit",
      "expectedSolverStatus": "FAIL",
      "requiredEntityIds": [
        "player",
        "start-room",
        "hall",
        "exit"
      ],
      "requiredRelationIds": [
        "player-start",
        "start-hall",
        "hall-exit-door"
      ],
      "requiredMetricFactIds": [],
      "typedCommandId": "sha256:0ee34fa75c70bf8e8ea2ac56f685593785a8672e50de3fb14156d8e52b7009e6",
      "sceneRevisionId": "sha256:a6c6bc1bc03c60a131762be0445272382b681acea77c40f0624c8331d7378a36",
      "sceneContentId": "sha256:e41f727a1e80eede766d997f11c580181e4d42f170ae40efcb3f22bd71f1d7c6",
      "fullSceneStatus": "FAIL",
      "sftOperationalStatus": "FAIL",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 7,
          "contextRelationCount": 8,
          "serializedBytes": 3161,
          "serializedTokens": 848,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 2,
          "irrelevantRelationCount": 5,
          "irrelevantEntityRate": 0.3333333333333333,
          "irrelevantRelationRate": 0.625,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.202916,
            "p50": 0.222666,
            "p95": 0.269,
            "max": 0.420708,
            "mean": 0.230583
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 4,
          "contextRelationCount": 4,
          "serializedBytes": 1791,
          "serializedTokens": 479,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.75,
          "requiredRelationRecall": 0.3333333333333333,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 3,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.75,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": false,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.216125,
            "p50": 0.221875,
            "p95": 0.263459,
            "max": 0.318125,
            "mean": 0.229202
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 5,
          "contextRelationCount": 7,
          "serializedBytes": 2497,
          "serializedTokens": 665,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 4,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.254583,
            "p50": 0.265792,
            "p95": 0.303917,
            "max": 0.369667,
            "mean": 0.274192
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 5,
          "contextRelationCount": 7,
          "serializedBytes": 2484,
          "serializedTokens": 663,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 4,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "FAIL",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.248667,
            "p50": 0.255958,
            "p95": 0.3795,
            "max": 0.445583,
            "mean": 0.27294
          }
        }
      ]
    },
    {
      "id": "game-clearance-missing-evidence-control",
      "scenarioId": "game",
      "query": "[typed control] Find a 900 mm-clear path from player to exit",
      "command": {
        "kind": "QUERY_REACHABILITY",
        "fromId": "player",
        "toId": "exit",
        "minimumClearance": {
          "min": 900,
          "max": 900,
          "unit": "mm"
        }
      },
      "expectedSolverStatus": "UNKNOWN",
      "expectsMissingEvidence": true,
      "requiredEntityIds": [
        "player",
        "start-room",
        "hall",
        "exit"
      ],
      "requiredRelationIds": [
        "player-start",
        "start-hall",
        "hall-exit-door"
      ],
      "requiredMetricFactIds": [],
      "missingMetricFactIds": [
        "relation:player-start:attributes.minimumClearanceMm",
        "relation:start-hall:attributes.minimumClearanceMm",
        "relation:hall-exit-door:attributes.minimumClearanceMm"
      ],
      "typedCommandId": "sha256:694d0bacff5d4e420db17368795dc0a8e42117ab21c3aeb91fe5725735d62b45",
      "sceneRevisionId": "sha256:a6c6bc1bc03c60a131762be0445272382b681acea77c40f0624c8331d7378a36",
      "sceneContentId": "sha256:e41f727a1e80eede766d997f11c580181e4d42f170ae40efcb3f22bd71f1d7c6",
      "fullSceneStatus": "UNKNOWN",
      "sftOperationalStatus": "UNKNOWN",
      "sftIncludesIndependentOracle": true,
      "sftDetectsMissingEvidence": true,
      "rows": [
        {
          "strategy": "FULL_SCENE",
          "contextEntityCount": 7,
          "contextRelationCount": 8,
          "serializedBytes": 3161,
          "serializedTokens": 848,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 2,
          "irrelevantRelationCount": 5,
          "irrelevantEntityRate": 0.3333333333333333,
          "irrelevantRelationRate": 0.625,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.224167,
            "p50": 0.235792,
            "p95": 0.377416,
            "max": 0.48775,
            "mean": 0.256413
          }
        },
        {
          "strategy": "LEXICAL_TOP_K",
          "contextEntityCount": 4,
          "contextRelationCount": 4,
          "serializedBytes": 1791,
          "serializedTokens": 479,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 0.75,
          "requiredRelationRecall": 0.3333333333333333,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 3,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.75,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": false,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.240791,
            "p50": 0.250709,
            "p95": 0.312042,
            "max": 0.396625,
            "mean": 0.262695
          }
        },
        {
          "strategy": "GRAPH_NEIGHBORHOOD",
          "contextEntityCount": 5,
          "contextRelationCount": 7,
          "serializedBytes": 2497,
          "serializedTokens": 665,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 4,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.275167,
            "p50": 0.292667,
            "p95": 0.431541,
            "max": 0.459708,
            "mean": 0.304557
          }
        },
        {
          "strategy": "SFT_OPERATIONAL",
          "contextEntityCount": 5,
          "contextRelationCount": 7,
          "serializedBytes": 2484,
          "serializedTokens": 663,
          "tokenizer": "o200k_base@gpt-tokenizer-4.0.0",
          "requiredEntityRecall": 1,
          "requiredRelationRecall": 1,
          "requiredMetricFactRecall": 1,
          "irrelevantEntityCount": 0,
          "irrelevantRelationCount": 4,
          "irrelevantEntityRate": 0,
          "irrelevantRelationRate": 0.5714285714285714,
          "solverStatus": "UNKNOWN",
          "solverStatusMatchesFullScene": true,
          "solverResultMatchesFullScene": true,
          "falseAuthoritativeExecution": false,
          "sourceRevisionComplete": true,
          "constructionLatencyMs": {
            "samples": 25,
            "min": 0.274375,
            "p50": 0.297792,
            "p95": 0.378292,
            "max": 0.475292,
            "mean": 0.30539
          }
        }
      ]
    }
  ],
  "aggregates": [
    {
      "strategy": "FULL_SCENE",
      "caseCount": 7,
      "meanRequiredEntityRecall": 1,
      "meanRequiredRelationRecall": 1,
      "meanRequiredMetricFactRecall": 1,
      "meanIrrelevantEntityCount": 4.428571,
      "meanIrrelevantRelationCount": 10.714286,
      "meanSerializedBytes": 5782.428571,
      "meanSerializedTokens": 1584.285714,
      "medianSerializedBytesRatioToFull": 1,
      "meanConstructionLatencyMs": 0.420605,
      "solverStatusMatchRate": 1,
      "solverResultParityRate": 1,
      "falseAuthoritativeExecutionRate": 0,
      "falseUnknownOrUnsupportedRate": 0,
      "sourceRevisionCompletenessRate": 1
    },
    {
      "strategy": "LEXICAL_TOP_K",
      "caseCount": 7,
      "meanRequiredEntityRecall": 0.719728,
      "meanRequiredRelationRecall": 0.547619,
      "meanRequiredMetricFactRecall": 0.875,
      "meanIrrelevantEntityCount": 1.571429,
      "meanIrrelevantRelationCount": 5.142857,
      "meanSerializedBytes": 3070.142857,
      "meanSerializedTokens": 841.857143,
      "medianSerializedBytesRatioToFull": 0.543837,
      "meanConstructionLatencyMs": 0.401242,
      "solverStatusMatchRate": 0.285714,
      "solverResultParityRate": 0.142857,
      "falseAuthoritativeExecutionRate": 0,
      "falseUnknownOrUnsupportedRate": 0.714286,
      "sourceRevisionCompletenessRate": 1
    },
    {
      "strategy": "GRAPH_NEIGHBORHOOD",
      "caseCount": 7,
      "meanRequiredEntityRecall": 1,
      "meanRequiredRelationRecall": 1,
      "meanRequiredMetricFactRecall": 1,
      "meanIrrelevantEntityCount": 2.857143,
      "meanIrrelevantRelationCount": 9,
      "meanSerializedBytes": 4959.571429,
      "meanSerializedTokens": 1361.285714,
      "medianSerializedBytesRatioToFull": 0.839644,
      "meanConstructionLatencyMs": 0.523345,
      "solverStatusMatchRate": 1,
      "solverResultParityRate": 1,
      "falseAuthoritativeExecutionRate": 0,
      "falseUnknownOrUnsupportedRate": 0,
      "sourceRevisionCompletenessRate": 1
    },
    {
      "strategy": "SFT_OPERATIONAL",
      "caseCount": 7,
      "meanRequiredEntityRecall": 1,
      "meanRequiredRelationRecall": 1,
      "meanRequiredMetricFactRecall": 1,
      "meanIrrelevantEntityCount": 1.857143,
      "meanIrrelevantRelationCount": 6.714286,
      "meanSerializedBytes": 4170.714286,
      "meanSerializedTokens": 1147,
      "medianSerializedBytesRatioToFull": 0.785827,
      "meanConstructionLatencyMs": 0.506936,
      "solverStatusMatchRate": 1,
      "solverResultParityRate": 1,
      "falseAuthoritativeExecutionRate": 0,
      "falseUnknownOrUnsupportedRate": 0,
      "sourceRevisionCompletenessRate": 1
    }
  ],
  "gates": {
    "hardSafety": {
      "sftRequiredEntityRecallIsComplete": true,
      "sftRequiredRelationRecallIsComplete": true,
      "sftRequiredMetricFactRecallIsComplete": true,
      "noFalseAuthoritativeExecution": true,
      "missingEvidenceControlsRefused": true,
      "sftSolverParityIsComplete": true,
      "sourceRevisionTraceabilityIsComplete": true
    },
    "efficiency": {
      "targetMedianBytesRatioToFull": 0.6,
      "observedMedianBytesRatioToFull": 0.785827,
      "passed": false
    },
    "modelComparisonEligible": false
  },
  "resultState": "VERIFIED_MECHANISM",
  "conclusion": "FAILED",
  "modelExperimentDisposition": "NOT_RUN_EFFICIENCY_GATE_FAILED",
  "measuredMetrics": true,
  "performanceResult": false,
  "limitations": [
    "The benchmark compares deterministic context construction on seven authored fixture cases, including one missing-evidence control; it is not a frontier-model answer-quality benchmark.",
    "Recall and irrelevant-context metrics use an independently authored entity, relation, and metric-fact oracle per case.",
    "LEXICAL_TOP_K is a transparent lexical baseline, not an embedding database or production Vector RAG implementation.",
    "GRAPH_NEIGHBORHOOD is a one-hop structural baseline, not Microsoft GraphRAG or another complete graph-retrieval system.",
    "Latency rows measure this local Node.js process and are descriptive, not a production performance claim.",
    "The predeclared median serialized-byte target was not met, so the frozen pinned-model holdout was not executed."
  ],
  "deterministicMembershipDigest": "sha256:35b64669d8f733ff700aa844a8701332de1492a28b46bc8727561417edb5713b",
  "id": "sha256:f81e74a1f17de90665b17015f457a215d2e20cbc62f80e3e696df2bf3ab43373"
}
