{
  "version": 1,
  "date": "2026-09-07",
  "engineChecks": {
    "passed": 16,
    "failed": 0
  },
  "historicalPrototype": {
    "model": "MiniMax-M2.7",
    "calls": 6,
    "formatPasses": 4,
    "candidateSelections": 6,
    "inputTokens": 18784,
    "outputTokens": 6273,
    "limitation": "Structurally accepted prose included unsupported claims."
  },
  "trials": [
    {
      "at": "2026-09-07T15:08:32.203Z",
      "model": "MiniMax-M2.7",
      "calls": 4,
      "observedTick": 137563,
      "results": [
        {
          "case": "live-traffic",
          "state": "rejected",
          "latencyMs": 9491,
          "verified": false,
          "candidate": null,
          "expected": "review-crew-cap",
          "expectedSelection": false,
          "claims": 4,
          "usage": {
            "inputTokens": 1537,
            "outputTokens": 681
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "live-resources",
          "state": "rejected",
          "latencyMs": 8997,
          "verified": false,
          "candidate": null,
          "expected": "review-deliveries",
          "expectedSelection": false,
          "claims": 5,
          "usage": {
            "inputTokens": 1507,
            "outputTokens": 664
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "missing-history",
          "state": "analyzed",
          "latencyMs": 6977,
          "verified": true,
          "candidate": null,
          "expected": null,
          "expectedSelection": true,
          "claims": 0,
          "usage": {
            "inputTokens": 1461,
            "outputTokens": 541
          },
          "error": null
        },
        {
          "case": "gym-with-measured-memory",
          "state": "rejected",
          "latencyMs": 9046,
          "verified": false,
          "candidate": null,
          "expected": "graded-road",
          "expectedSelection": false,
          "claims": 4,
          "usage": {
            "inputTokens": 3202,
            "outputTokens": 734
          },
          "error": "CLAIMS_REJECTED"
        }
      ],
      "label": "First typed protocol"
    },
    {
      "at": "2026-09-07T15:11:42.636Z",
      "model": "MiniMax-M2.7",
      "calls": 4,
      "observedTick": 137563,
      "results": [
        {
          "case": "live-traffic",
          "state": "analyzed",
          "latencyMs": 11103,
          "verified": true,
          "candidate": "review-crew-cap",
          "expected": "review-crew-cap",
          "expectedSelection": true,
          "claims": 5,
          "usage": {
            "inputTokens": 1709,
            "outputTokens": 806
          },
          "error": null
        },
        {
          "case": "live-resources",
          "state": "analyzed",
          "latencyMs": 8465,
          "verified": true,
          "candidate": "review-deliveries",
          "expected": "review-deliveries",
          "expectedSelection": true,
          "claims": 5,
          "usage": {
            "inputTokens": 1675,
            "outputTokens": 712
          },
          "error": null
        },
        {
          "case": "missing-history",
          "state": "analyzed",
          "latencyMs": 7418,
          "verified": true,
          "candidate": "none",
          "expected": "none",
          "expectedSelection": true,
          "claims": 0,
          "usage": {
            "inputTokens": 1609,
            "outputTokens": 546
          },
          "error": null
        },
        {
          "case": "gym-with-measured-memory",
          "state": "analyzed",
          "latencyMs": 14301,
          "verified": true,
          "candidate": "graded-road",
          "expected": "graded-road",
          "expectedSelection": true,
          "claims": 4,
          "usage": {
            "inputTokens": 3402,
            "outputTokens": 1011
          },
          "error": null,
          "simulatedOutcome": 0.32000000000000006
        }
      ],
      "label": "Explicit candidate IDs"
    },
    {
      "at": "2026-09-07T15:32:54.769Z",
      "model": "MiniMax-M3",
      "calls": 4,
      "observedTick": 138986,
      "results": [
        {
          "case": "live-traffic",
          "state": "rejected",
          "latencyMs": 8010,
          "verified": false,
          "candidate": "review-crew-cap",
          "expected": "review-crew-cap",
          "expectedSelection": false,
          "claims": 7,
          "usage": {
            "inputTokens": 1959,
            "outputTokens": 1442
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "live-resources",
          "state": "rejected",
          "latencyMs": 19222,
          "verified": false,
          "candidate": "review-deliveries",
          "expected": "review-deliveries",
          "expectedSelection": false,
          "claims": 5,
          "usage": {
            "inputTokens": 1925,
            "outputTokens": 1110
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "missing-history",
          "state": "analyzed",
          "latencyMs": 3077,
          "verified": true,
          "candidate": "none",
          "expected": "none",
          "expectedSelection": true,
          "claims": 1,
          "usage": {
            "inputTokens": 1859,
            "outputTokens": 460
          },
          "error": null
        },
        {
          "case": "gym-with-measured-memory",
          "state": "analyzed",
          "latencyMs": 2733,
          "verified": true,
          "candidate": "graded-road",
          "expected": "graded-road",
          "expectedSelection": true,
          "claims": 4,
          "usage": {
            "inputTokens": 3606,
            "outputTokens": 773
          },
          "error": null,
          "simulatedOutcome": 0.32000000000000006
        }
      ],
      "label": "M3 initial"
    },
    {
      "at": "2026-09-07T15:40:56.650Z",
      "model": "MiniMax-M3",
      "calls": 4,
      "observedTick": 139500,
      "results": [
        {
          "case": "live-traffic",
          "state": "rejected",
          "latencyMs": 9253,
          "verified": false,
          "candidate": "review-crew-cap",
          "expected": "review-crew-cap",
          "expectedSelection": false,
          "claims": 10,
          "usage": {
            "inputTokens": 4200,
            "outputTokens": 1448
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "live-resources",
          "state": "analyzed",
          "latencyMs": 7316,
          "verified": true,
          "candidate": "review-deliveries",
          "expected": "review-deliveries",
          "expectedSelection": true,
          "claims": 10,
          "usage": {
            "inputTokens": 4166,
            "outputTokens": 1315
          },
          "error": null
        },
        {
          "case": "missing-history",
          "state": "rejected",
          "latencyMs": 6478,
          "verified": false,
          "candidate": "none",
          "expected": "none",
          "expectedSelection": false,
          "claims": 1,
          "usage": {
            "inputTokens": 3690,
            "outputTokens": 405
          },
          "error": "CLAIMS_REJECTED"
        },
        {
          "case": "gym-with-measured-memory",
          "state": "analyzed",
          "latencyMs": 4479,
          "verified": true,
          "candidate": "graded-road",
          "expected": "graded-road",
          "expectedSelection": true,
          "claims": 5,
          "usage": {
            "inputTokens": 4116,
            "outputTokens": 723
          },
          "error": null,
          "simulatedOutcome": 0.32000000000000006
        }
      ],
      "label": "M3 fact-specific schema"
    },
    {
      "at": "2026-09-07T15:46:19.806Z",
      "model": "MiniMax-M3",
      "calls": 4,
      "observedTick": 139749,
      "results": [
        {
          "case": "live-traffic",
          "state": "analyzed",
          "latencyMs": 18198,
          "verified": true,
          "candidate": "review-crew-cap",
          "expected": "review-crew-cap",
          "expectedSelection": true,
          "claims": 12,
          "usage": {
            "inputTokens": 2454,
            "outputTokens": 1575
          },
          "error": null
        },
        {
          "case": "live-resources",
          "state": "analyzed",
          "latencyMs": 10122,
          "verified": true,
          "candidate": "review-deliveries",
          "expected": "review-deliveries",
          "expectedSelection": true,
          "claims": 10,
          "usage": {
            "inputTokens": 2420,
            "outputTokens": 1837
          },
          "error": null
        },
        {
          "case": "missing-history",
          "state": "analyzed",
          "latencyMs": 6041,
          "verified": true,
          "candidate": "none",
          "expected": "none",
          "expectedSelection": true,
          "claims": 1,
          "usage": {
            "inputTokens": 2264,
            "outputTokens": 374
          },
          "error": null
        },
        {
          "case": "gym-with-measured-memory",
          "state": "analyzed",
          "latencyMs": 4396,
          "verified": true,
          "candidate": "graded-road",
          "expected": "graded-road",
          "expectedSelection": true,
          "claims": 5,
          "usage": {
            "inputTokens": 3789,
            "outputTokens": 919
          },
          "error": null,
          "simulatedOutcome": 0.32000000000000006
        }
      ],
      "label": "M3 explicit JSON-text claim transport"
    }
  ],
  "currentM3Conclusion": "All four M3 cases accepted; all 28 claims matched their snapshots. Missing history produced abstention; measured memory selected the road. JSON-text transport is decoded once before exact typed verification. This small trial is not an accuracy guarantee.",
  "simulation": {
    "domain": "abstract queue gym, not live Moon physics",
    "seed": 7,
    "requests": 60,
    "baselineCompleted": 4,
    "extraCrewCompleted": 6,
    "roadCompleted": 8,
    "liftCompleted": 2,
    "gameActions": 0
  },
  "reportChecks": {
    "passed": true,
    "webglDisabled": true,
    "hardwareGraphicsDisabled": true,
    "interactionChecks": 10
  }
}