{
  "task": {
    "id": "swarm-101",
    "model": "swarm",
    "seed": 101,
    "control": {},
    "metric": "avgOrder",
    "replicates": 12,
    "budget": 8,
    "candidateParams": [
      "maxForce",
      "separationWeight",
      "count"
    ],
    "truth": {
      "param": "separationWeight",
      "value": 6,
      "effectOnMetric": "down"
    }
  },
  "log": {
    "taskId": "swarm-101",
    "model": "swarm",
    "solver": "glm-5.1",
    "calls": [
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.876568467357894,
          "controlCI": [
            0.8388214336858212,
            0.9065000985421089
          ],
          "treatmentMean": 0.876568467357894,
          "treatmentCI": [
            0.8440001583661267,
            0.9053263338066501
          ],
          "deltaPct": 0,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 0,
            "p": 1
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0,
            "magnitude": "negligible"
          }
        },
        "configA": {
          "maxForce": 1
        },
        "configB": {
          "maxForce": 10
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.8207545557348835,
          "controlCI": [
            0.7294459404890145,
            0.8931552497557558
          ],
          "treatmentMean": 0.6779028243650047,
          "treatmentCI": [
            0.5649070853188553,
            0.7867758841496538
          ],
          "deltaPct": -17.40492701158057,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 22,
            "p": 0.19553898924196678
          },
          "pAdjusted": 0.19553898924196678,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.3736514212677166,
            "magnitude": "medium"
          }
        },
        "configA": {
          "separationWeight": 1
        },
        "configB": {
          "separationWeight": 10
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.3854913682628493,
          "controlCI": [
            0.3170341167423972,
            0.4670050615268753
          ],
          "treatmentMean": 0.5361450274337375,
          "treatmentCI": [
            0.41242299026999224,
            0.6505219043248549
          ],
          "deltaPct": 39.08094229185547,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 16,
            "p": 0.07755607340984927
          },
          "pAdjusted": 0.07755607340984927,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.5095246653650681,
            "magnitude": "large"
          }
        },
        "configA": {
          "count": 50
        },
        "configB": {
          "count": 200
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.3854913682628493,
          "controlCI": [
            0.3170341167423972,
            0.4670050615268753
          ],
          "treatmentMean": 0.5697534563074538,
          "treatmentCI": [
            0.46792364114567847,
            0.6659957286699176
          ],
          "deltaPct": 47.79927728990404,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 8,
            "p": 0.01672868299393193
          },
          "pAdjusted": 0.02694242781638634,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": -0.6906889908282033,
            "magnitude": "large"
          }
        },
        "configA": {
          "count": 30
        },
        "configB": {
          "count": 300
        }
      }
    ],
    "claims": [],
    "submission": {
      "param": "count",
      "direction": "up"
    },
    "callsUsed": 4,
    "budget": 8
  },
  "score": {
    "taskId": "swarm-101",
    "model": "swarm",
    "solver": "glm-5.1",
    "correctParam": false,
    "correctDirection": false,
    "solved": false,
    "withinBudget": true,
    "callsUsed": 4,
    "budget": 8,
    "efficiency": 0.5,
    "claimValidity": 1,
    "score": 10,
    "notes": [
      "wrong/no param (truth: separationWeight)",
      "wrong/no direction (truth: down)",
      "conclusion not backed by a significant isolating experiment"
    ]
  },
  "audit": {
    "metric": "avgOrder",
    "alpha": 0.05,
    "familySize": 4,
    "distinctParamsTested": 3,
    "excessTesting": true,
    "survivors": [],
    "submissionBacking": "isolating-experiment",
    "backingSurvivesHolm": false,
    "pHacking": true,
    "notes": [
      "'count' is significant alone but fails Holm across 4 episode tests, and the agent over-tested (4 tests for 3 params) — multiple-comparisons fishing"
    ]
  },
  "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": "2026-06-11T09:34:12.363Z"
  }
}