{
  "updated": "2026-11-02",
  "questions": 95,
  "trials": 3,
  "sessions": 4560,
  "sessions_run": 5616,
  "harnesses": {
    "claude-code": "Claude Code with the plugin loaded (native skill discovery and the startup hook), Skill/Read/Grep/Glob only, in a fresh minimal Clarinet project; sessions without skills load no plugin.",
    "codex-catalog": "Codex, tool-disabled: the skill catalog plus up to three rounds of bounded, recorded skill-file reads; no shell or web; sessions without skills get no catalog."
  },
  "notes": [
    "Every model answered each question three times without the skills and three times with them.",
    "5,616 sessions were run in all; the 4,560 counted here are each question's latest run, for both conditions.",
    "paired.unassessed counts pairs in which one session ended without an answer (provider errors).",
    "Accuracy is over answered sessions, and only the exact letter counts.",
    "paired matches each question and trial across the two conditions: a gain means only the answer with the skills was right, a loss only the answer without them. gain_points is (gains - losses) per 100 pairs.",
    "Claude models ran in Claude Code and OpenAI models in Codex, with different harnesses: compare the two conditions within a model, not models with each other.",
    "Token means are per session as each harness reports them and include the harness prompt; tokenizers differ between vendors. mean_cost_usd is what Claude Code reported at API prices; the Codex sessions ran on a subscription, so they have no dollar cost."
  ],
  "models": {
    "Claude Haiku 5.5": {
      "harness": "claude-code",
      "model_id": "claude-haiku-5-5",
      "without_skills": {
        "correct": 228,
        "answered": 285,
        "accuracy_pct": 80.0,
        "mean_input_tokens": 19729,
        "mean_output_tokens": 1510,
        "mean_cost_usd": 0.0019
      },
      "with_skills": {
        "correct": 267,
        "answered": 285,
        "accuracy_pct": 93.68,
        "sessions_reading_skills": 230,
        "mean_input_tokens": 40458,
        "mean_output_tokens": 1188,
        "mean_cost_usd": 0.0039
      },
      "paired": {
        "gain": 42,
        "loss": 3,
        "both_correct": 225,
        "both_wrong": 15
      },
      "gain_points": 13.68
    },
    "Claude Sonnet 5.5": {
      "harness": "claude-code",
      "model_id": "claude-sonnet-5-5",
      "without_skills": {
        "correct": 244,
        "answered": 285,
        "accuracy_pct": 85.61,
        "mean_input_tokens": 14773,
        "mean_output_tokens": 541,
        "mean_cost_usd": 0.021
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 265,
        "mean_input_tokens": 35264,
        "mean_output_tokens": 450,
        "mean_cost_usd": 0.0654
      },
      "paired": {
        "gain": 41,
        "loss": 0,
        "both_correct": 244,
        "both_wrong": 0
      },
      "gain_points": 14.39
    },
    "Claude Opus 5.5": {
      "harness": "claude-code",
      "model_id": "claude-opus-5-5",
      "without_skills": {
        "correct": 254,
        "answered": 285,
        "accuracy_pct": 89.12,
        "mean_input_tokens": 14678,
        "mean_output_tokens": 634,
        "mean_cost_usd": 0.0415
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 130,
        "mean_input_tokens": 24783,
        "mean_output_tokens": 527,
        "mean_cost_usd": 0.0872
      },
      "paired": {
        "gain": 31,
        "loss": 0,
        "both_correct": 254,
        "both_wrong": 0
      },
      "gain_points": 10.88
    },
    "Claude Fable 5.1": {
      "harness": "claude-code",
      "model_id": "claude-fable-5-1",
      "without_skills": {
        "correct": 242,
        "answered": 285,
        "accuracy_pct": 84.91,
        "mean_input_tokens": 40168,
        "mean_output_tokens": 1349,
        "mean_cost_usd": 0.1833
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 283,
        "mean_input_tokens": 55706,
        "mean_output_tokens": 659,
        "mean_cost_usd": 0.4018
      },
      "paired": {
        "gain": 43,
        "loss": 0,
        "both_correct": 242,
        "both_wrong": 0
      },
      "gain_points": 15.09
    },
    "gpt-5.6-luna": {
      "harness": "codex-catalog",
      "model_id": "gpt-5.6-luna",
      "effort": "medium",
      "without_skills": {
        "correct": 180,
        "answered": 283,
        "accuracy_pct": 63.6,
        "mean_input_tokens": 7834,
        "mean_output_tokens": 247
      },
      "with_skills": {
        "correct": 275,
        "answered": 283,
        "accuracy_pct": 97.17,
        "sessions_reading_skills": 283,
        "mean_input_tokens": 33976,
        "mean_output_tokens": 242
      },
      "paired": {
        "gain": 95,
        "loss": 0,
        "both_correct": 178,
        "both_wrong": 8,
        "unassessed": 4
      },
      "gain_points": 33.81
    },
    "gpt-6-sol": {
      "harness": "codex-catalog",
      "model_id": "gpt-6-sol",
      "effort": "medium",
      "without_skills": {
        "correct": 236,
        "answered": 285,
        "accuracy_pct": 82.81,
        "mean_input_tokens": 10138,
        "mean_output_tokens": 237
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 265,
        "mean_input_tokens": 44316,
        "mean_output_tokens": 215
      },
      "paired": {
        "gain": 49,
        "loss": 0,
        "both_correct": 236,
        "both_wrong": 0
      },
      "gain_points": 17.19
    },
    "gpt-6.1-sol": {
      "harness": "codex-catalog",
      "model_id": "gpt-6.1-sol",
      "effort": "medium",
      "without_skills": {
        "correct": 260,
        "answered": 285,
        "accuracy_pct": 91.23,
        "mean_input_tokens": 10651,
        "mean_output_tokens": 136
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 260,
        "mean_input_tokens": 44625,
        "mean_output_tokens": 158
      },
      "paired": {
        "gain": 25,
        "loss": 0,
        "both_correct": 260,
        "both_wrong": 0
      },
      "gain_points": 8.77
    },
    "gpt-6-astra": {
      "harness": "codex-catalog",
      "model_id": "gpt-6-astra",
      "effort": "medium",
      "without_skills": {
        "correct": 263,
        "answered": 285,
        "accuracy_pct": 92.28,
        "mean_input_tokens": 10587,
        "mean_output_tokens": 109
      },
      "with_skills": {
        "correct": 285,
        "answered": 285,
        "accuracy_pct": 100.0,
        "sessions_reading_skills": 239,
        "mean_input_tokens": 34790,
        "mean_output_tokens": 122
      },
      "paired": {
        "gain": 22,
        "loss": 0,
        "both_correct": 263,
        "both_wrong": 0
      },
      "gain_points": 7.72
    }
  },
  "topics": {
    "Claude Haiku 5.5": {
      "api": {
        "without_skills": {
          "correct": 9,
          "answered": 15,
          "accuracy_pct": 60.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 15,
          "accuracy_pct": 60.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 4,
          "answered": 9,
          "accuracy_pct": 44.44
        },
        "with_skills": {
          "correct": 4,
          "answered": 9,
          "accuracy_pct": 44.44
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 5,
          "answered": 9,
          "accuracy_pct": 55.56
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 4,
          "answered": 9,
          "accuracy_pct": 44.44
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 61,
          "answered": 81,
          "accuracy_pct": 75.31
        },
        "with_skills": {
          "correct": 78,
          "answered": 81,
          "accuracy_pct": 96.3
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 2,
          "answered": 3,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 38,
          "answered": 42,
          "accuracy_pct": 90.48
        },
        "with_skills": {
          "correct": 41,
          "answered": 42,
          "accuracy_pct": 97.62
        }
      },
      "security": {
        "without_skills": {
          "correct": 26,
          "answered": 27,
          "accuracy_pct": 96.3
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 3,
          "answered": 6,
          "accuracy_pct": 50.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 25,
          "answered": 33,
          "accuracy_pct": 75.76
        },
        "with_skills": {
          "correct": 30,
          "answered": 33,
          "accuracy_pct": 90.91
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "Claude Sonnet 5.5": {
      "api": {
        "without_skills": {
          "correct": 10,
          "answered": 15,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 6,
          "answered": 9,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 6,
          "answered": 9,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 67,
          "answered": 81,
          "accuracy_pct": 82.72
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 2,
          "answered": 3,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 33,
          "answered": 42,
          "accuracy_pct": 78.57
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 4,
          "answered": 6,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 29,
          "answered": 33,
          "accuracy_pct": 87.88
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "Claude Opus 5.5": {
      "api": {
        "without_skills": {
          "correct": 11,
          "answered": 15,
          "accuracy_pct": 73.33
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 7,
          "answered": 9,
          "accuracy_pct": 77.78
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 72,
          "answered": 81,
          "accuracy_pct": 88.89
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 36,
          "answered": 42,
          "accuracy_pct": 85.71
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 5,
          "answered": 6,
          "accuracy_pct": 83.33
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 26,
          "answered": 33,
          "accuracy_pct": 78.79
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 10,
          "answered": 12,
          "accuracy_pct": 83.33
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "Claude Fable 5.1": {
      "api": {
        "without_skills": {
          "correct": 11,
          "answered": 15,
          "accuracy_pct": 73.33
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 3,
          "answered": 9,
          "accuracy_pct": 33.33
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 65,
          "answered": 81,
          "accuracy_pct": 80.25
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 2,
          "answered": 3,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 37,
          "answered": 42,
          "accuracy_pct": 88.1
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 3,
          "answered": 6,
          "accuracy_pct": 50.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 28,
          "answered": 33,
          "accuracy_pct": 84.85
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 11,
          "answered": 12,
          "accuracy_pct": 91.67
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 4,
          "answered": 6,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "gpt-5.6-luna": {
      "api": {
        "without_skills": {
          "correct": 9,
          "answered": 14,
          "accuracy_pct": 64.29
        },
        "with_skills": {
          "correct": 13,
          "answered": 15,
          "accuracy_pct": 86.67
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 3,
          "answered": 6,
          "accuracy_pct": 50.0
        },
        "with_skills": {
          "correct": 5,
          "answered": 6,
          "accuracy_pct": 83.33
        }
      },
      "control": {
        "without_skills": {
          "correct": 23,
          "answered": 27,
          "accuracy_pct": 85.19
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 1,
          "answered": 9,
          "accuracy_pct": 11.11
        },
        "with_skills": {
          "correct": 8,
          "answered": 9,
          "accuracy_pct": 88.89
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 5,
          "answered": 9,
          "accuracy_pct": 55.56
        },
        "with_skills": {
          "correct": 8,
          "answered": 8,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 6,
          "answered": 9,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 46,
          "answered": 80,
          "accuracy_pct": 57.5
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 0,
          "answered": 3,
          "accuracy_pct": 0.0
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 25,
          "answered": 42,
          "accuracy_pct": 59.52
        },
        "with_skills": {
          "correct": 39,
          "answered": 42,
          "accuracy_pct": 92.86
        }
      },
      "security": {
        "without_skills": {
          "correct": 19,
          "answered": 27,
          "accuracy_pct": 70.37
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 4,
          "answered": 6,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 4,
          "answered": 5,
          "accuracy_pct": 80.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 22,
          "answered": 33,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 5,
          "answered": 6,
          "accuracy_pct": 83.33
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "gpt-6-sol": {
      "api": {
        "without_skills": {
          "correct": 9,
          "answered": 15,
          "accuracy_pct": 60.0
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 4,
          "answered": 6,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 7,
          "answered": 9,
          "accuracy_pct": 77.78
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 5,
          "answered": 9,
          "accuracy_pct": 55.56
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 6,
          "answered": 9,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 64,
          "answered": 81,
          "accuracy_pct": 79.01
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 0,
          "answered": 3,
          "accuracy_pct": 0.0
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 34,
          "answered": 42,
          "accuracy_pct": 80.95
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 25,
          "answered": 27,
          "accuracy_pct": 92.59
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 31,
          "answered": 33,
          "accuracy_pct": 93.94
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "gpt-6.1-sol": {
      "api": {
        "without_skills": {
          "correct": 11,
          "answered": 15,
          "accuracy_pct": 73.33
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 4,
          "answered": 6,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 6,
          "answered": 9,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 73,
          "answered": 81,
          "accuracy_pct": 90.12
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 2,
          "answered": 3,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 36,
          "answered": 42,
          "accuracy_pct": 85.71
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 32,
          "answered": 33,
          "accuracy_pct": 96.97
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    },
    "gpt-6-astra": {
      "api": {
        "without_skills": {
          "correct": 11,
          "answered": 15,
          "accuracy_pct": 73.33
        },
        "with_skills": {
          "correct": 15,
          "answered": 15,
          "accuracy_pct": 100.0
        }
      },
      "bitcoin": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "control": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "deployment": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "epochs": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "fuzzing": {
        "without_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 9,
          "answered": 9,
          "accuracy_pct": 100.0
        }
      },
      "language": {
        "without_skills": {
          "correct": 73,
          "answered": 81,
          "accuracy_pct": 90.12
        },
        "with_skills": {
          "correct": 81,
          "answered": 81,
          "accuracy_pct": 100.0
        }
      },
      "oracles": {
        "without_skills": {
          "correct": 2,
          "answered": 3,
          "accuracy_pct": 66.67
        },
        "with_skills": {
          "correct": 3,
          "answered": 3,
          "accuracy_pct": 100.0
        }
      },
      "protocols": {
        "without_skills": {
          "correct": 37,
          "answered": 42,
          "accuracy_pct": 88.1
        },
        "with_skills": {
          "correct": 42,
          "answered": 42,
          "accuracy_pct": 100.0
        }
      },
      "security": {
        "without_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 27,
          "answered": 27,
          "accuracy_pct": 100.0
        }
      },
      "staking": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      },
      "testing": {
        "without_skills": {
          "correct": 29,
          "answered": 33,
          "accuracy_pct": 87.88
        },
        "with_skills": {
          "correct": 33,
          "answered": 33,
          "accuracy_pct": 100.0
        }
      },
      "upgrades": {
        "without_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 12,
          "answered": 12,
          "accuracy_pct": 100.0
        }
      },
      "wallets": {
        "without_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        },
        "with_skills": {
          "correct": 6,
          "answered": 6,
          "accuracy_pct": 100.0
        }
      }
    }
  }
}
