{
 "claude-sonnet-5-5": {
  "all": {
   "A": {
    "trials": 95,
    "passes": 29,
    "pass_rate": 0.3053,
    "ci95_cluster_bootstrap": [
     0.1667,
     0.4583
    ],
    "tasks_solved_any": 13,
    "mean_raw_reward": 0.3627,
    "tokens_mean": 140485,
    "tokens_median": 102789,
    "output_tokens_mean": 3854,
    "agent_wall_s_mean": 196,
    "agent_wall_s_median": 33,
    "timeouts": 4,
    "skill_read_rate": 0.0
   },
   "C": {
    "trials": 94,
    "passes": 51,
    "pass_rate": 0.5426,
    "ci95_cluster_bootstrap": [
     0.3789,
     0.6989
    ],
    "tasks_solved_any": 22,
    "mean_raw_reward": 0.6084,
    "tokens_mean": 152783,
    "tokens_median": 113786,
    "output_tokens_mean": 3397,
    "agent_wall_s_mean": 189,
    "agent_wall_s_median": 42,
    "timeouts": 3,
    "skill_read_rate": 0.8298
   },
   "paired": {
    "n_tasks": 32,
    "mean_delta": 0.2396,
    "delta_ci95_task_bootstrap": [
     0.1094,
     0.3802
    ],
    "tasks_C_better": [
     "energy-market-pricing",
     "fix-druid-loophole-cve",
     "manufacturing-codebook-normalization",
     "manufacturing-fjsp-optimization",
     "mario-coin-counting",
     "paratransit-routing",
     "radar-vital-signs",
     "sec-financial-report",
     "simpo-code-reproduction",
     "software-dependency-audit",
     "threejs-structure-parser",
     "threejs-to-obj",
     "weighted-gdp-calc"
    ],
    "tasks_C_worse": [
     "pddl-airport-planning",
     "python-scala-translation"
    ],
    "tasks_tied": [
     "dialogue-parser",
     "drone-planning-control",
     "exam-block-sequencing",
     "exoplanet-detection-period",
     "fix-erlang-ssh-cve",
     "flink-query",
     "invoice-fraud-detection",
     "jpg-ocr-stat",
     "mars-clouds-clustering",
     "multilingual-video-dubbing",
     "pddl-tpp-planning",
     "pdf-excel-diff",
     "powerlifting-coef-calc",
     "pptx-reference-formatting",
     "quantum-numerical-simulation",
     "setup-fuzzing-py",
     "xlsx-recover-data"
    ],
    "per_task": {
     "dialogue-parser": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "drone-planning-control": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "energy-market-pricing": {
      "A": [
       1,
       0,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0.3333
     },
     "exam-block-sequencing": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "exoplanet-detection-period": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "fix-druid-loophole-cve": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1
      ],
      "delta": 0.5
     },
     "fix-erlang-ssh-cve": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "flink-query": {
      "A": [
       0,
       1,
       1
      ],
      "C": [
       0,
       1,
       1
      ],
      "delta": 0.0
     },
     "invoice-fraud-detection": {
      "A": [
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "jpg-ocr-stat": {
      "A": [
       1,
       0,
       0
      ],
      "C": [
       1,
       0,
       0
      ],
      "delta": 0.0
     },
     "manufacturing-codebook-normalization": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1,
       0
      ],
      "delta": 0.3333
     },
     "manufacturing-fjsp-optimization": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       0
      ],
      "delta": 0.6667
     },
     "mario-coin-counting": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "mars-clouds-clustering": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "multilingual-video-dubbing": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "paratransit-routing": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       1
      ],
      "delta": 0.3333
     },
     "pddl-airport-planning": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": -0.3333
     },
     "pddl-tpp-planning": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "pdf-excel-diff": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "powerlifting-coef-calc": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "pptx-reference-formatting": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "python-scala-translation": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": -0.3333
     },
     "quantum-numerical-simulation": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "radar-vital-signs": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "sec-financial-report": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "setup-fuzzing-py": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "simpo-code-reproduction": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "software-dependency-audit": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1,
       0
      ],
      "delta": 0.3333
     },
     "threejs-structure-parser": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "threejs-to-obj": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       1,
       1
      ],
      "delta": 0.3333
     },
     "weighted-gdp-calc": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       0
      ],
      "delta": 0.5
     },
     "xlsx-recover-data": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     }
    },
    "sign_test_p": 0.0074
   },
   "per_domain": {
    "cybersecurity": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.4545,
     "delta_pp": 20.5
    },
    "finance-economics": {
     "tasks": 4,
     "A": 0.0,
     "C": 0.3636,
     "delta_pp": 36.4
    },
    "industrial-physical-systems": {
     "tasks": 4,
     "A": 0.1667,
     "C": 0.5,
     "delta_pp": 33.3
    },
    "mathematics-or-formal-reasoning": {
     "tasks": 4,
     "A": 0.3333,
     "C": 0.3333,
     "delta_pp": 0.0
    },
    "media-content-production": {
     "tasks": 4,
     "A": 0.3333,
     "C": 0.9167,
     "delta_pp": 58.3
    },
    "natural-science": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.5,
     "delta_pp": 25.0
    },
    "office-white-collar": {
     "tasks": 4,
     "A": 0.8333,
     "C": 0.8333,
     "delta_pp": 0.0
    },
    "software-engineering": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.4167,
     "delta_pp": 16.7
    }
   },
   "per_difficulty": {
    "easy": {
     "tasks": 2,
     "A": 0.5,
     "C": 0.5
    },
    "medium": {
     "tasks": 19,
     "A": 0.2632,
     "C": 0.5
    },
    "hard": {
     "tasks": 11,
     "A": 0.3438,
     "C": 0.625
    }
   },
   "C_by_skill_read": {
    "read_trials": 78,
    "read_pass_rate": 0.5513,
    "unread_trials": 16,
    "unread_pass_rate": 0.5
   },
   "A_skill_reads": 0,
   "infra_excluded": [
    "A/drone-planning-control/drone-planning-control__STRypKq:AgentSetupTimeoutError",
    "A/drone-planning-control/drone-planning-control__JrirP2w:ApiRateLimitError",
    "A/drone-planning-control/drone-planning-control__buSdw9j:ApiRateLimitError",
    "A/drone-planning-control/drone-planning-control__d8w2R6n:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__7tKS6ab:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__CXCVaUr:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__tLU9Dfz:ApiRateLimitError",
    "A/fix-druid-loophole-cve/fix-druid-loophole-cve__xe2C9V9:VerifierTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__CTcQBxD:AgentSetupTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__rirxUbY:AgentSetupTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__Q39kgYs:ApiRateLimitError",
    "A/invoice-fraud-detection/invoice-fraud-detection__rB7sCV3:ApiRateLimitError",
    "A/invoice-fraud-detection/invoice-fraud-detection__sTPCaqk:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__GcEHFau:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__NPj53HD:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__dZadEDh:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__THxWYzx:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__Xi72iw6:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__giGkLqv:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__3WcWfNG:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__9qyz2Fn:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__zcLujVr:ApiRateLimitError",
    "A/radar-vital-signs/radar-vital-signs__7K6oH9A:NonZeroAgentExitCodeError",
    "A/radar-vital-signs/radar-vital-signs__af3ddDg:NonZeroAgentExitCodeError",
    "A/radar-vital-signs/radar-vital-signs__uSQrqXH:NonZeroAgentExitCodeError",
    "A/weighted-gdp-calc/weighted-gdp-calc__qY9LzUg:AgentSetupTimeoutError",
    "A/weighted-gdp-calc/weighted-gdp-calc__AHUMkPG:ApiRateLimitError",
    "A/weighted-gdp-calc/weighted-gdp-calc__uUMKvjw:ApiRateLimitError",
    "A/weighted-gdp-calc/weighted-gdp-calc__vG5QSKQ:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__koabhX4:AgentSetupTimeoutError",
    "A/xlsx-recover-data/xlsx-recover-data__gRyjxpD:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__sqD34ov:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__tHTvSN9:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__EWWL53g:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__bRKfvTu:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__mibnaES:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__KUuX8k6:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__n9EGqG5:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__ounrT7d:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__ewgt4z2:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__8nFFHXP:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__jhufHPk:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__s37e8VE:ApiRateLimitError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__DaXR8qR:AgentSetupTimeoutError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__H4okFn5:VerifierTimeoutError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__xZRJz8U:VerifierTimeoutError",
    "C/invoice-fraud-detection/invoice-fraud-detection__Dkk27yw:ApiRateLimitError",
    "C/invoice-fraud-detection/invoice-fraud-detection__XZt7uhK:ApiRateLimitError",
    "C/invoice-fraud-detection/invoice-fraud-detection__nThxRrA:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__5ZXdUgW:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__GZKdJXY:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__WyTE3F7:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__4PMDJRK:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__dr9Erou:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__tFb5hUp:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__9kDEJqT:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__KpWDsQa:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__rey66qx:ApiRateLimitError",
    "C/radar-vital-signs/radar-vital-signs__3TJ4fo4:NonZeroAgentExitCodeError",
    "C/radar-vital-signs/radar-vital-signs__By5YaoT:NonZeroAgentExitCodeError",
    "C/radar-vital-signs/radar-vital-signs__UwGnfes:NonZeroAgentExitCodeError",
    "C/weighted-gdp-calc/weighted-gdp-calc__KXtmDHi:AgentSetupTimeoutError",
    "C/weighted-gdp-calc/weighted-gdp-calc__SqGfg7u:EnvironmentStartTimeoutError",
    "C/weighted-gdp-calc/weighted-gdp-calc__iR8kuai:ApiRateLimitError",
    "C/weighted-gdp-calc/weighted-gdp-calc__q5LEN2E:ApiRateLimitError",
    "C/weighted-gdp-calc/weighted-gdp-calc__zADeWcm:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__NWqKRGP:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__xHYh52q:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__zPL4NBc:ApiRateLimitError"
   ]
  },
  "excl_borderline_leak": {
   "A": {
    "trials": 89,
    "passes": 26,
    "pass_rate": 0.2921,
    "ci95_cluster_bootstrap": [
     0.1461,
     0.4494
    ],
    "tasks_solved_any": 12,
    "mean_raw_reward": 0.3535,
    "tokens_mean": 145063,
    "tokens_median": 106989,
    "output_tokens_mean": 4005,
    "agent_wall_s_mean": 208,
    "agent_wall_s_median": 54,
    "timeouts": 4,
    "skill_read_rate": 0.0
   },
   "C": {
    "trials": 88,
    "passes": 45,
    "pass_rate": 0.5114,
    "ci95_cluster_bootstrap": [
     0.3448,
     0.6742
    ],
    "tasks_solved_any": 20,
    "mean_raw_reward": 0.5817,
    "tokens_mean": 156244,
    "tokens_median": 118936,
    "output_tokens_mean": 3523,
    "agent_wall_s_mean": 200,
    "agent_wall_s_median": 51,
    "timeouts": 3,
    "skill_read_rate": 0.8182
   },
   "paired": {
    "n_tasks": 30,
    "mean_delta": 0.2222,
    "delta_ci95_task_bootstrap": [
     0.0944,
     0.3611
    ],
    "tasks_C_better": [
     "energy-market-pricing",
     "fix-druid-loophole-cve",
     "manufacturing-codebook-normalization",
     "manufacturing-fjsp-optimization",
     "paratransit-routing",
     "radar-vital-signs",
     "sec-financial-report",
     "simpo-code-reproduction",
     "software-dependency-audit",
     "threejs-structure-parser",
     "threejs-to-obj",
     "weighted-gdp-calc"
    ],
    "tasks_C_worse": [
     "pddl-airport-planning",
     "python-scala-translation"
    ],
    "tasks_tied": [
     "dialogue-parser",
     "drone-planning-control",
     "exam-block-sequencing",
     "exoplanet-detection-period",
     "fix-erlang-ssh-cve",
     "flink-query",
     "invoice-fraud-detection",
     "jpg-ocr-stat",
     "mars-clouds-clustering",
     "multilingual-video-dubbing",
     "pddl-tpp-planning",
     "pdf-excel-diff",
     "pptx-reference-formatting",
     "quantum-numerical-simulation",
     "setup-fuzzing-py",
     "xlsx-recover-data"
    ],
    "per_task": {
     "dialogue-parser": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "drone-planning-control": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "energy-market-pricing": {
      "A": [
       1,
       0,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0.3333
     },
     "exam-block-sequencing": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "exoplanet-detection-period": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "fix-druid-loophole-cve": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1
      ],
      "delta": 0.5
     },
     "fix-erlang-ssh-cve": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "flink-query": {
      "A": [
       0,
       1,
       1
      ],
      "C": [
       0,
       1,
       1
      ],
      "delta": 0.0
     },
     "invoice-fraud-detection": {
      "A": [
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "jpg-ocr-stat": {
      "A": [
       1,
       0,
       0
      ],
      "C": [
       1,
       0,
       0
      ],
      "delta": 0.0
     },
     "manufacturing-codebook-normalization": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1,
       0
      ],
      "delta": 0.3333
     },
     "manufacturing-fjsp-optimization": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       0
      ],
      "delta": 0.6667
     },
     "mars-clouds-clustering": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "multilingual-video-dubbing": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "paratransit-routing": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       1
      ],
      "delta": 0.3333
     },
     "pddl-airport-planning": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": -0.3333
     },
     "pddl-tpp-planning": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "pdf-excel-diff": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "pptx-reference-formatting": {
      "A": [
       1,
       1,
       1
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 0
     },
     "python-scala-translation": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": -0.3333
     },
     "quantum-numerical-simulation": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "radar-vital-signs": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "sec-financial-report": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "setup-fuzzing-py": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     },
     "simpo-code-reproduction": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "software-dependency-audit": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       1,
       0
      ],
      "delta": 0.3333
     },
     "threejs-structure-parser": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       1,
       1
      ],
      "delta": 1
     },
     "threejs-to-obj": {
      "A": [
       0,
       0,
       1
      ],
      "C": [
       0,
       1,
       1
      ],
      "delta": 0.3333
     },
     "weighted-gdp-calc": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       1,
       0
      ],
      "delta": 0.5
     },
     "xlsx-recover-data": {
      "A": [
       0,
       0,
       0
      ],
      "C": [
       0,
       0,
       0
      ],
      "delta": 0
     }
    },
    "sign_test_p": 0.0129
   },
   "per_domain": {
    "cybersecurity": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.4545,
     "delta_pp": 20.5
    },
    "finance-economics": {
     "tasks": 4,
     "A": 0.0,
     "C": 0.3636,
     "delta_pp": 36.4
    },
    "industrial-physical-systems": {
     "tasks": 4,
     "A": 0.1667,
     "C": 0.5,
     "delta_pp": 33.3
    },
    "mathematics-or-formal-reasoning": {
     "tasks": 4,
     "A": 0.3333,
     "C": 0.3333,
     "delta_pp": 0.0
    },
    "media-content-production": {
     "tasks": 3,
     "A": 0.4444,
     "C": 0.8889,
     "delta_pp": 44.4
    },
    "natural-science": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.5,
     "delta_pp": 25.0
    },
    "office-white-collar": {
     "tasks": 3,
     "A": 0.7778,
     "C": 0.7778,
     "delta_pp": 0.0
    },
    "software-engineering": {
     "tasks": 4,
     "A": 0.25,
     "C": 0.4167,
     "delta_pp": 16.7
    }
   },
   "per_difficulty": {
    "easy": {
     "tasks": 1,
     "A": 0.0,
     "C": 0.0
    },
    "medium": {
     "tasks": 18,
     "A": 0.2778,
     "C": 0.4717
    },
    "hard": {
     "tasks": 11,
     "A": 0.3438,
     "C": 0.625
    }
   },
   "C_by_skill_read": {
    "read_trials": 72,
    "read_pass_rate": 0.5139,
    "unread_trials": 16,
    "unread_pass_rate": 0.5
   },
   "A_skill_reads": 0,
   "infra_excluded": [
    "A/drone-planning-control/drone-planning-control__STRypKq:AgentSetupTimeoutError",
    "A/drone-planning-control/drone-planning-control__JrirP2w:ApiRateLimitError",
    "A/drone-planning-control/drone-planning-control__buSdw9j:ApiRateLimitError",
    "A/drone-planning-control/drone-planning-control__d8w2R6n:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__7tKS6ab:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__CXCVaUr:ApiRateLimitError",
    "A/energy-market-pricing/energy-market-pricing__tLU9Dfz:ApiRateLimitError",
    "A/fix-druid-loophole-cve/fix-druid-loophole-cve__xe2C9V9:VerifierTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__CTcQBxD:AgentSetupTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__rirxUbY:AgentSetupTimeoutError",
    "A/invoice-fraud-detection/invoice-fraud-detection__Q39kgYs:ApiRateLimitError",
    "A/invoice-fraud-detection/invoice-fraud-detection__rB7sCV3:ApiRateLimitError",
    "A/invoice-fraud-detection/invoice-fraud-detection__sTPCaqk:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__GcEHFau:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__NPj53HD:ApiRateLimitError",
    "A/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__dZadEDh:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__THxWYzx:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__Xi72iw6:ApiRateLimitError",
    "A/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__giGkLqv:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__3WcWfNG:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__9qyz2Fn:ApiRateLimitError",
    "A/paratransit-routing/paratransit-routing__zcLujVr:ApiRateLimitError",
    "A/radar-vital-signs/radar-vital-signs__7K6oH9A:NonZeroAgentExitCodeError",
    "A/radar-vital-signs/radar-vital-signs__af3ddDg:NonZeroAgentExitCodeError",
    "A/radar-vital-signs/radar-vital-signs__uSQrqXH:NonZeroAgentExitCodeError",
    "A/weighted-gdp-calc/weighted-gdp-calc__qY9LzUg:AgentSetupTimeoutError",
    "A/weighted-gdp-calc/weighted-gdp-calc__AHUMkPG:ApiRateLimitError",
    "A/weighted-gdp-calc/weighted-gdp-calc__uUMKvjw:ApiRateLimitError",
    "A/weighted-gdp-calc/weighted-gdp-calc__vG5QSKQ:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__koabhX4:AgentSetupTimeoutError",
    "A/xlsx-recover-data/xlsx-recover-data__gRyjxpD:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__sqD34ov:ApiRateLimitError",
    "A/xlsx-recover-data/xlsx-recover-data__tHTvSN9:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__EWWL53g:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__bRKfvTu:ApiRateLimitError",
    "C/drone-planning-control/drone-planning-control__mibnaES:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__KUuX8k6:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__n9EGqG5:ApiRateLimitError",
    "C/energy-market-pricing/energy-market-pricing__ounrT7d:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__ewgt4z2:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__8nFFHXP:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__jhufHPk:ApiRateLimitError",
    "C/exam-block-sequencing/exam-block-sequencing__s37e8VE:ApiRateLimitError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__DaXR8qR:AgentSetupTimeoutError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__H4okFn5:VerifierTimeoutError",
    "C/fix-druid-loophole-cve/fix-druid-loophole-cve__xZRJz8U:VerifierTimeoutError",
    "C/invoice-fraud-detection/invoice-fraud-detection__Dkk27yw:ApiRateLimitError",
    "C/invoice-fraud-detection/invoice-fraud-detection__XZt7uhK:ApiRateLimitError",
    "C/invoice-fraud-detection/invoice-fraud-detection__nThxRrA:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__5ZXdUgW:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__GZKdJXY:ApiRateLimitError",
    "C/manufacturing-codebook-normalization/manufacturing-codebook-normaliza__WyTE3F7:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__4PMDJRK:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__dr9Erou:ApiRateLimitError",
    "C/manufacturing-fjsp-optimization/manufacturing-fjsp-optimization__tFb5hUp:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__9kDEJqT:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__KpWDsQa:ApiRateLimitError",
    "C/paratransit-routing/paratransit-routing__rey66qx:ApiRateLimitError",
    "C/radar-vital-signs/radar-vital-signs__3TJ4fo4:NonZeroAgentExitCodeError",
    "C/radar-vital-signs/radar-vital-signs__By5YaoT:NonZeroAgentExitCodeError",
    "C/radar-vital-signs/radar-vital-signs__UwGnfes:NonZeroAgentExitCodeError",
    "C/weighted-gdp-calc/weighted-gdp-calc__KXtmDHi:AgentSetupTimeoutError",
    "C/weighted-gdp-calc/weighted-gdp-calc__SqGfg7u:EnvironmentStartTimeoutError",
    "C/weighted-gdp-calc/weighted-gdp-calc__iR8kuai:ApiRateLimitError",
    "C/weighted-gdp-calc/weighted-gdp-calc__q5LEN2E:ApiRateLimitError",
    "C/weighted-gdp-calc/weighted-gdp-calc__zADeWcm:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__NWqKRGP:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__xHYh52q:ApiRateLimitError",
    "C/xlsx-recover-data/xlsx-recover-data__zPL4NBc:ApiRateLimitError"
   ]
  }
 }
}