{
  "task": {
    "id": "origin-101-l2",
    "model": "origin",
    "seed": 101,
    "control": {},
    "metric": "finalPopulation",
    "replicates": 12,
    "budget": 8,
    "tier": "L2",
    "candidateParams": [
      "vision_range",
      "reproduce_threshold",
      "attack_damage"
    ],
    "truth": {
      "param": "reproduce_threshold",
      "value": 322,
      "effectOnMetric": "down",
      "magnitude": "medium"
    }
  },
  "log": {
    "taskId": "origin-101-l2",
    "model": "origin",
    "solver": "codex-gpt-5.5",
    "calls": [
      {
        "kind": "experiment",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 67.16666666666667,
          "controlCI": [
            55.416666666666664,
            78.75
          ],
          "treatmentMean": 62.166666666666664,
          "treatmentCI": [
            40.916666666666664,
            83.25
          ],
          "deltaPct": -7.444168734491326,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 34,
            "p": 0.7238779813411522
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.10198341204735074,
            "magnitude": "small"
          }
        },
        "configA": {},
        "configB": {
          "vision_range": 7.5
        }
      },
      {
        "kind": "experiment",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 67.16666666666667,
          "controlCI": [
            55.416666666666664,
            78.75
          ],
          "treatmentMean": 228.5,
          "treatmentCI": [
            169.25,
            280.8333333333333
          ],
          "deltaPct": 240.19851116625307,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 2,
            "p": 0.004192722492179168
          },
          "pAdjusted": 0.01642323032987303,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.8265622349255549,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "reproduce_threshold": 0.75
        }
      },
      {
        "kind": "probe",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 228.5,
          "controlCI": [
            168.5,
            281.5833333333333
          ],
          "treatmentMean": 34.083333333333336,
          "treatmentCI": [
            22.25,
            44.166666666666664
          ],
          "deltaPct": -85.08388037928519,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 2,
            "p": 0.004192722492179168
          },
          "pAdjusted": 0.012390760277253277,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8265622349255549,
            "magnitude": "large"
          }
        },
        "configA": {
          "reproduce_threshold": 0.75
        }
      },
      {
        "kind": "experiment",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 67.16666666666667,
          "controlCI": [
            55.416666666666664,
            78.75
          ],
          "treatmentMean": 222,
          "treatmentCI": [
            183.66666666666666,
            258.6666666666667
          ],
          "deltaPct": 230.52109181141435,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 0,
            "p": 0.0025263126954468973
          },
          "pAdjusted": 0.012631563477234486,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.8718533162913388,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "reproduce_threshold": 1.5
        }
      },
      {
        "kind": "experiment",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 67.16666666666667,
          "controlCI": [
            55.416666666666664,
            78.75
          ],
          "treatmentMean": 80.33333333333333,
          "treatmentCI": [
            69,
            92.25
          ],
          "deltaPct": 19.602977667493782,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 15,
            "p": 0.0646368510485622
          },
          "pAdjusted": 0.323184255242811,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.5334025586191666,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "attack_damage": 2
        }
      },
      {
        "kind": "probe",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 165.83333333333334,
          "controlCI": [
            134.75,
            197.41666666666666
          ],
          "treatmentMean": 34.083333333333336,
          "treatmentCI": [
            22.25,
            44.166666666666664
          ],
          "deltaPct": -79.44723618090453,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 0,
            "p": 0.0025166349849183867
          },
          "pAdjusted": 0.011964479582623477,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8721888380725838,
            "magnitude": "large"
          }
        },
        "configA": {
          "reproduce_threshold": 3
        }
      },
      {
        "kind": "experiment",
        "metric": "finalPopulation",
        "stats": {
          "metric": "finalPopulation",
          "label": "Final Population",
          "controlMean": 67.16666666666667,
          "controlCI": [
            55.416666666666664,
            78.75
          ],
          "treatmentMean": 66.33333333333333,
          "treatmentCI": [
            49.833333333333336,
            80.91666666666667
          ],
          "deltaPct": -1.2406947890818998,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 32,
            "p": 0.609845035502727
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.1473093729572844,
            "magnitude": "small"
          }
        },
        "configA": {},
        "configB": {
          "attack_damage": 0.5
        }
      }
    ],
    "claims": [],
    "submission": {
      "param": "vision_range",
      "direction": "down",
      "magnitude": "large"
    },
    "callsUsed": 7,
    "budget": 8
  },
  "score": {
    "taskId": "origin-101-l2",
    "model": "origin",
    "solver": "codex-gpt-5.5",
    "correctParam": false,
    "correctDirection": true,
    "correctMagnitude": false,
    "solved": false,
    "withinBudget": true,
    "callsUsed": 7,
    "budget": 8,
    "efficiency": 0.125,
    "claimValidity": 1,
    "score": 26.9,
    "notes": [
      "wrong/no param (truth: reproduce_threshold)",
      "magnitude off by one (said large, truth medium)",
      "conclusion not backed by a significant isolating experiment"
    ]
  },
  "audit": {
    "metric": "finalPopulation",
    "alpha": 0.05,
    "familySize": 5,
    "distinctParamsTested": 3,
    "excessTesting": true,
    "survivors": [
      "reproduce_threshold"
    ],
    "submissionBacking": "isolating-experiment",
    "backingSurvivesHolm": false,
    "pHacking": false,
    "notes": [
      "'vision_range' is significant alone but borderline: it does not survive family-wide Holm (low power, not over-testing)"
    ]
  },
  "provenance": {
    "bench": "0.1.0",
    "mcp": "0.1.1",
    "engines": {
      "swarm": "0.1.0",
      "market": "0.1.0",
      "social": "0.1.0",
      "morph": "0.1.0",
      "origin": "0.1.0"
    },
    "commit": "f305d9d54670c03a92b2c93b76040e15a5bc1840",
    "node": "v25.2.1",
    "createdAt": "2026-06-11T06:24:56.007Z"
  }
}