{
  "configs": {
    "F1/bare": {
      "task": "F1",
      "cfg": "bare",
      "label": "bare",
      "n": 3,
      "hidden_mean": 5,
      "hidden_median": 5,
      "hidden_per_run": [
        5,
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 3,
        "test_req2": 3,
        "test_req3": 3,
        "test_req4": 3,
        "test_req5": 3
      },
      "suite_green": 3,
      "typecheck_green": 3,
      "tests_added_mean": 4.667,
      "code_diff_mean": 141.667,
      "code_diff_median": 137,
      "all_diff_mean": 142.333,
      "non_code_files_mean": 0.333,
      "tokens_mean": 519113.667,
      "tokens_median": 361842,
      "out_tokens_mean": 12001,
      "usd_mean": 0.343,
      "usd_median": 0.286,
      "usd_po_mean": 0.021,
      "wall_s_mean": 84.5,
      "wall_s_median": 71.7,
      "driver_turns_mean": 1.333,
      "questions_mean": 0.333,
      "caps": {},
      "end_reasons": [
        "done",
        "done",
        "done"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        false,
        false,
        false
      ],
      "skills_per_run": {
        "bare-r1": [],
        "bare-r2": [],
        "bare-r3": []
      },
      "skills_failed": {}
    },
    "F1/sp": {
      "task": "F1",
      "cfg": "sp",
      "label": "sp (hook-routed)",
      "n": 3,
      "hidden_mean": 5,
      "hidden_median": 5,
      "hidden_per_run": [
        5,
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 3,
        "test_req2": 3,
        "test_req3": 3,
        "test_req4": 3,
        "test_req5": 3
      },
      "suite_green": 3,
      "typecheck_green": 3,
      "tests_added_mean": 12.333,
      "code_diff_mean": 235,
      "code_diff_median": 227,
      "all_diff_mean": 237,
      "non_code_files_mean": 0.667,
      "tokens_mean": 724600.667,
      "tokens_median": 765108,
      "out_tokens_mean": 16869.333,
      "usd_mean": 0.502,
      "usd_median": 0.525,
      "usd_po_mean": 0.041,
      "wall_s_mean": 118.1,
      "wall_s_median": 119.6,
      "driver_turns_mean": 2.667,
      "questions_mean": 1.667,
      "caps": {},
      "end_reasons": [
        "done",
        "done",
        "done"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        true,
        true,
        true
      ],
      "skills_per_run": {
        "sp-r1": [
          "brainstorming"
        ],
        "sp-r2": [
          "brainstorming",
          "test-driven-development"
        ],
        "sp-r3": [
          "brainstorming"
        ]
      },
      "skills_failed": {}
    },
    "F1/spd": {
      "task": "F1",
      "cfg": "spd",
      "label": "spd (Superpowers driven)",
      "n": 3,
      "hidden_mean": 5,
      "hidden_median": 5,
      "hidden_per_run": [
        5,
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 3,
        "test_req2": 3,
        "test_req3": 3,
        "test_req4": 3,
        "test_req5": 3
      },
      "suite_green": 3,
      "typecheck_green": 3,
      "tests_added_mean": 27.667,
      "code_diff_mean": 386,
      "code_diff_median": 371,
      "all_diff_mean": 582,
      "non_code_files_mean": 2,
      "tokens_mean": 4479876.333,
      "tokens_median": 3887015,
      "out_tokens_mean": 61684.667,
      "usd_mean": 3.738,
      "usd_median": 3.412,
      "usd_po_mean": 0.116,
      "wall_s_mean": 607.367,
      "wall_s_median": 600.8,
      "driver_turns_mean": 7,
      "questions_mean": 4,
      "caps": {},
      "end_reasons": [
        "done",
        "done",
        "done"
      ],
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        true,
        true,
        true
      ],
      "skills_per_run": {
        "spd-r1": [
          "test-driven-development"
        ],
        "spd-r2": [
          "test-driven-development"
        ],
        "spd-r3": [
          "test-driven-development",
          "finishing-a-development-branch"
        ]
      },
      "skills_failed": {}
    },
    "F1/mp": {
      "task": "F1",
      "cfg": "mp",
      "label": "mp (Main Flow)",
      "n": 3,
      "hidden_mean": 5,
      "hidden_median": 5,
      "hidden_per_run": [
        5,
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 3,
        "test_req2": 3,
        "test_req3": 3,
        "test_req4": 3,
        "test_req5": 3
      },
      "suite_green": 3,
      "typecheck_green": 3,
      "tests_added_mean": 27.667,
      "code_diff_mean": 372,
      "code_diff_median": 369,
      "all_diff_mean": 578,
      "non_code_files_mean": 7.667,
      "tokens_mean": 4476103.667,
      "tokens_median": 4286993,
      "out_tokens_mean": 56959,
      "usd_mean": 2.043,
      "usd_median": 2.102,
      "usd_po_mean": 0.205,
      "wall_s_mean": 439.033,
      "wall_s_median": 472.4,
      "driver_turns_mean": 12,
      "questions_mean": 7,
      "caps": {
        "mp-r1": [
          "driver_turns_12"
        ],
        "mp-r3": [
          "driver_turns_12"
        ]
      },
      "end_reasons": [
        "cap_driver_turns",
        "done",
        "cap_driver_turns"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        false,
        false,
        false
      ],
      "skills_per_run": {
        "mp-r1": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ],
        "mp-r2": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ],
        "mp-r3": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ]
      },
      "skills_failed": {}
    },
    "F1/both": {
      "task": "F1",
      "cfg": "both",
      "label": "both (sp+mp, mp flow)",
      "n": 1,
      "hidden_mean": 5,
      "hidden_median": 5,
      "hidden_per_run": [
        5
      ],
      "req_pass_rate": {
        "test_req1": 1,
        "test_req2": 1,
        "test_req3": 1,
        "test_req4": 1,
        "test_req5": 1
      },
      "suite_green": 1,
      "typecheck_green": 1,
      "tests_added_mean": 22,
      "code_diff_mean": 374,
      "code_diff_median": 374,
      "all_diff_mean": 561,
      "non_code_files_mean": 8,
      "tokens_mean": 5754354,
      "tokens_median": 5754354,
      "out_tokens_mean": 57632,
      "usd_mean": 2.326,
      "usd_median": 2.326,
      "usd_po_mean": 0.218,
      "wall_s_mean": 448.2,
      "wall_s_median": 448.2,
      "driver_turns_mean": 12,
      "questions_mean": 6,
      "caps": {},
      "end_reasons": [
        "done"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        true
      ],
      "skills_per_run": {
        "both-r1": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ]
      },
      "skills_failed": {}
    },
    "F2/bare": {
      "task": "F2",
      "cfg": "bare",
      "label": "bare",
      "n": 2,
      "hidden_mean": 4.5,
      "hidden_median": 4.5,
      "hidden_per_run": [
        4,
        5
      ],
      "req_pass_rate": {
        "test_req1": 2,
        "test_req2": 2,
        "test_req3": 1,
        "test_req4": 2,
        "test_req5": 2
      },
      "suite_green": 2,
      "typecheck_green": 2,
      "tests_added_mean": 5,
      "code_diff_mean": 206.5,
      "code_diff_median": 206.5,
      "all_diff_mean": 209.5,
      "non_code_files_mean": 1,
      "tokens_mean": 651322,
      "tokens_median": 651322.0,
      "out_tokens_mean": 15475,
      "usd_mean": 0.413,
      "usd_median": 0.413,
      "usd_po_mean": 0.031,
      "wall_s_mean": 103.15,
      "wall_s_median": 103.15,
      "driver_turns_mean": 2,
      "questions_mean": 0.5,
      "caps": {},
      "end_reasons": [
        "done",
        "done"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        false,
        false
      ],
      "skills_per_run": {
        "f2-bare-r1": [],
        "f2-bare-r2": []
      },
      "skills_failed": {}
    },
    "F2/spd": {
      "task": "F2",
      "cfg": "spd",
      "label": "spd (Superpowers driven)",
      "n": 2,
      "hidden_mean": 5,
      "hidden_median": 5.0,
      "hidden_per_run": [
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 2,
        "test_req2": 2,
        "test_req3": 2,
        "test_req4": 2,
        "test_req5": 2
      },
      "suite_green": 2,
      "typecheck_green": 2,
      "tests_added_mean": 33,
      "code_diff_mean": 488.5,
      "code_diff_median": 488.5,
      "all_diff_mean": 750.5,
      "non_code_files_mean": 2.5,
      "tokens_mean": 4490365,
      "tokens_median": 4490365.0,
      "out_tokens_mean": 56644.5,
      "usd_mean": 3.213,
      "usd_median": 3.213,
      "usd_po_mean": 0.101,
      "wall_s_mean": 511.7,
      "wall_s_median": 511.7,
      "driver_turns_mean": 6.5,
      "questions_mean": 3.5,
      "caps": {},
      "end_reasons": [
        "done",
        "done"
      ],
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        true,
        true
      ],
      "skills_per_run": {
        "f2-spd-r1": [
          "writing-plans",
          "test-driven-development"
        ],
        "f2-spd-r2": [
          "test-driven-development"
        ]
      },
      "skills_failed": {}
    },
    "F2/mp": {
      "task": "F2",
      "cfg": "mp",
      "label": "mp (Main Flow)",
      "n": 2,
      "hidden_mean": 5,
      "hidden_median": 5.0,
      "hidden_per_run": [
        5,
        5
      ],
      "req_pass_rate": {
        "test_req1": 2,
        "test_req2": 2,
        "test_req3": 2,
        "test_req4": 2,
        "test_req5": 2
      },
      "suite_green": 2,
      "typecheck_green": 2,
      "tests_added_mean": 31,
      "code_diff_mean": 469,
      "code_diff_median": 469.0,
      "all_diff_mean": 658,
      "non_code_files_mean": 6.5,
      "tokens_mean": 4520638.5,
      "tokens_median": 4520638.5,
      "out_tokens_mean": 57142,
      "usd_mean": 2.067,
      "usd_median": 2.067,
      "usd_po_mean": 0.203,
      "wall_s_mean": 421.2,
      "wall_s_median": 421.2,
      "driver_turns_mean": 12,
      "questions_mean": 6.5,
      "caps": {
        "f2-mp-r1": [
          "driver_turns_12"
        ]
      },
      "end_reasons": [
        "cap_driver_turns",
        "done"
      ],
      "models": [
        "claude-sonnet-5-5"
      ],
      "superpowers_injected": [
        false,
        false
      ],
      "skills_per_run": {
        "f2-mp-r1": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ],
        "f2-mp-r2": [
          "grilling",
          "domain-modeling",
          "tdd",
          "code-review"
        ]
      },
      "skills_failed": {}
    }
  },
  "stages": {
    "F1/sp_skill": {
      "(none)": {
        "runs_with_stage": 3,
        "tokens_mean": 19549,
        "tokens_median": 19538,
        "out_tokens_mean": 238,
        "usd_mean": 0.03,
        "usd_median": 0.03,
        "wall_s_mean": 5.333,
        "wall_s_median": 5.4
      },
      "brainstorming": {
        "runs_with_stage": 3,
        "tokens_mean": 249260,
        "tokens_median": 247900,
        "out_tokens_mean": 8037.333,
        "usd_mean": 0.219,
        "usd_median": 0.244,
        "wall_s_mean": 40.533,
        "wall_s_median": 40.1
      },
      "test-driven-development": {
        "runs_with_stage": 1,
        "tokens_mean": 595318,
        "tokens_median": 595318,
        "out_tokens_mean": 15622,
        "usd_mean": 0.405,
        "usd_median": 0.405,
        "wall_s_mean": 93.0,
        "wall_s_median": 93.0
      },
      "implementation (no skill)": {
        "runs_with_stage": 2,
        "tokens_mean": 386028.5,
        "tokens_median": 386028.5,
        "out_tokens_mean": 5080,
        "usd_mean": 0.176,
        "usd_median": 0.176,
        "wall_s_mean": 60.3,
        "wall_s_median": 60.3
      }
    },
    "F1/sp_phase": {
      "before any skill": {
        "runs_with_stage": 3,
        "tokens_mean": 19549,
        "tokens_median": 19538,
        "out_tokens_mean": 238,
        "usd_mean": 0.03,
        "usd_median": 0.03,
        "wall_s_mean": 5.333,
        "wall_s_median": 5.4
      },
      "brainstorming": {
        "runs_with_stage": 3,
        "tokens_mean": 249260,
        "tokens_median": 247900,
        "out_tokens_mean": 8037.333,
        "usd_mean": 0.219,
        "usd_median": 0.244,
        "wall_s_mean": 40.533,
        "wall_s_median": 40.1
      },
      "implementation (no skill)": {
        "runs_with_stage": 2,
        "tokens_mean": 386028.5,
        "tokens_median": 386028.5,
        "out_tokens_mean": 5080,
        "usd_mean": 0.176,
        "usd_median": 0.176,
        "wall_s_mean": 60.3,
        "wall_s_median": 60.3
      },
      "execution": {
        "runs_with_stage": 1,
        "tokens_mean": 595318,
        "tokens_median": 595318,
        "out_tokens_mean": 15622,
        "usd_mean": 0.405,
        "usd_median": 0.405,
        "wall_s_mean": 93.0,
        "wall_s_median": 93.0
      }
    },
    "F1/spd_skill": {
      "brainstorming": {
        "runs_with_stage": 3,
        "tokens_mean": 182021.667,
        "tokens_median": 160834,
        "out_tokens_mean": 4160.667,
        "usd_mean": 0.166,
        "usd_median": 0.156,
        "wall_s_mean": 38.8,
        "wall_s_median": 33.6
      },
      "writing-plans": {
        "runs_with_stage": 3,
        "tokens_mean": 321354.667,
        "tokens_median": 298267,
        "out_tokens_mean": 11827,
        "usd_mean": 0.266,
        "usd_median": 0.274,
        "wall_s_mean": 86.733,
        "wall_s_median": 95.1
      },
      "executing-plans": {
        "runs_with_stage": 3,
        "tokens_mean": 66938,
        "tokens_median": 68739,
        "out_tokens_mean": 505,
        "usd_mean": 0.047,
        "usd_median": 0.047,
        "wall_s_mean": 4.6,
        "wall_s_median": 4.6
      },
      "test-driven-development": {
        "runs_with_stage": 3,
        "tokens_mean": 2927098.333,
        "tokens_median": 2854967,
        "out_tokens_mean": 22508.333,
        "usd_mean": 0.974,
        "usd_median": 0.936,
        "wall_s_mean": 457.033,
        "wall_s_median": 448.9
      },
      "finishing-a-development-branch": {
        "runs_with_stage": 1,
        "tokens_mean": 976505,
        "tokens_median": 976505,
        "out_tokens_mean": 5297,
        "usd_mean": 0.282,
        "usd_median": 0.282,
        "wall_s_mean": 50.7,
        "wall_s_median": 50.7
      },
      "test-driven-development [subagent]": {
        "runs_with_stage": 3,
        "tokens_mean": 656962,
        "tokens_median": 565811,
        "out_tokens_mean": 20918,
        "usd_mean": 2.192,
        "usd_median": 2.047,
        "wall_s_mean": 0.0,
        "wall_s_median": 0.0
      }
    },
    "F1/spd_phase": {
      "brainstorming": {
        "runs_with_stage": 3,
        "tokens_mean": 182021.667,
        "tokens_median": 160834,
        "out_tokens_mean": 4160.667,
        "usd_mean": 0.166,
        "usd_median": 0.156,
        "wall_s_mean": 38.8,
        "wall_s_median": 33.6
      },
      "writing-plans": {
        "runs_with_stage": 3,
        "tokens_mean": 321354.667,
        "tokens_median": 298267,
        "out_tokens_mean": 11827,
        "usd_mean": 0.266,
        "usd_median": 0.274,
        "wall_s_mean": 86.733,
        "wall_s_median": 95.1
      },
      "execution": {
        "runs_with_stage": 3,
        "tokens_mean": 2994036.333,
        "tokens_median": 2916459,
        "out_tokens_mean": 23013.333,
        "usd_mean": 1.021,
        "usd_median": 0.981,
        "wall_s_mean": 461.633,
        "wall_s_median": 453.5
      },
      "finishing": {
        "runs_with_stage": 1,
        "tokens_mean": 976505,
        "tokens_median": 976505,
        "out_tokens_mean": 5297,
        "usd_mean": 0.282,
        "usd_median": 0.282,
        "wall_s_mean": 50.7,
        "wall_s_median": 50.7
      },
      "subagents (reviewers, explorers)": {
        "runs_with_stage": 3,
        "tokens_mean": 656962,
        "tokens_median": 565811,
        "out_tokens_mean": 20918,
        "usd_mean": 2.192,
        "usd_median": 2.047,
        "wall_s_mean": 0.0,
        "wall_s_median": 0.0
      }
    },
    "F1/spd_slash": {
      "/brainstorming": {
        "runs_with_stage": 3,
        "tokens_mean": 182021.667,
        "tokens_median": 160834,
        "out_tokens_mean": null,
        "usd_mean": 0.166,
        "usd_median": 0.156,
        "wall_s_mean": 40.283,
        "wall_s_median": 35.44
      },
      "/writing-plans": {
        "runs_with_stage": 3,
        "tokens_mean": 321354.667,
        "tokens_median": 298267,
        "out_tokens_mean": null,
        "usd_mean": 0.266,
        "usd_median": 0.274,
        "wall_s_mean": 88.133,
        "wall_s_median": 95.82
      },
      "/executing-plans": {
        "runs_with_stage": 3,
        "tokens_mean": 3976500,
        "tokens_median": 3427914,
        "out_tokens_mean": null,
        "usd_mean": 3.307,
        "usd_median": 3.028,
        "wall_s_mean": 478.95,
        "wall_s_median": 453.89
      }
    },
    "F1/mp": {
      "/grill-with-docs": {
        "runs_with_stage": 3,
        "tokens_mean": 316507.333,
        "tokens_median": 348862,
        "out_tokens_mean": 5999,
        "usd_mean": 0.204,
        "usd_median": 0.215,
        "wall_s_mean": 59.933,
        "wall_s_median": 59.9
      },
      "/to-spec": {
        "runs_with_stage": 3,
        "tokens_mean": 163344,
        "tokens_median": 167578,
        "out_tokens_mean": 5134.667,
        "usd_mean": 0.118,
        "usd_median": 0.113,
        "wall_s_mean": 42.1,
        "wall_s_median": 40.8
      },
      "/to-tickets": {
        "runs_with_stage": 3,
        "tokens_mean": 257955.333,
        "tokens_median": 268062,
        "out_tokens_mean": 5867.333,
        "usd_mean": 0.143,
        "usd_median": 0.151,
        "wall_s_mean": 44.4,
        "wall_s_median": 47.8
      },
      "/implement": {
        "runs_with_stage": 3,
        "tokens_mean": 2655564.667,
        "tokens_median": 2613062,
        "out_tokens_mean": 30214.667,
        "usd_mean": 1.125,
        "usd_median": 1.195,
        "wall_s_mean": 224.567,
        "wall_s_median": 245.4
      },
      "/code-review": {
        "runs_with_stage": 3,
        "tokens_mean": 1047585.667,
        "tokens_median": 1107555,
        "out_tokens_mean": 9293.667,
        "usd_mean": 0.439,
        "usd_median": 0.445,
        "wall_s_mean": 64.2,
        "wall_s_median": 65.8
      },
      "/pr": {
        "runs_with_stage": 1,
        "tokens_mean": 105440,
        "tokens_median": 105440,
        "out_tokens_mean": 1349,
        "usd_mean": 0.041,
        "usd_median": 0.041,
        "wall_s_mean": 11.6,
        "wall_s_median": 11.6
      }
    },
    "F1/both": {
      "/grill-with-docs": {
        "runs_with_stage": 1,
        "tokens_mean": 344391,
        "tokens_median": 344391,
        "out_tokens_mean": 6193,
        "usd_mean": 0.225,
        "usd_median": 0.225,
        "wall_s_mean": 61.8,
        "wall_s_median": 61.8
      },
      "/to-spec": {
        "runs_with_stage": 1,
        "tokens_mean": 217572,
        "tokens_median": 217572,
        "out_tokens_mean": 4835,
        "usd_mean": 0.123,
        "usd_median": 0.123,
        "wall_s_mean": 42.9,
        "wall_s_median": 42.9
      },
      "/to-tickets": {
        "runs_with_stage": 1,
        "tokens_mean": 382953,
        "tokens_median": 382953,
        "out_tokens_mean": 7115,
        "usd_mean": 0.196,
        "usd_median": 0.196,
        "wall_s_mean": 49.4,
        "wall_s_median": 49.4
      },
      "/implement": {
        "runs_with_stage": 1,
        "tokens_mean": 3070579,
        "tokens_median": 3070579,
        "out_tokens_mean": 24618,
        "usd_mean": 1.112,
        "usd_median": 1.112,
        "wall_s_mean": 194.2,
        "wall_s_median": 194.2
      },
      "/code-review": {
        "runs_with_stage": 1,
        "tokens_mean": 1615265,
        "tokens_median": 1615265,
        "out_tokens_mean": 13410,
        "usd_mean": 0.625,
        "usd_median": 0.625,
        "wall_s_mean": 86.8,
        "wall_s_median": 86.8
      },
      "/pr": {
        "runs_with_stage": 1,
        "tokens_mean": 123594,
        "tokens_median": 123594,
        "out_tokens_mean": 1461,
        "usd_mean": 0.045,
        "usd_median": 0.045,
        "wall_s_mean": 13.1,
        "wall_s_median": 13.1
      }
    },
    "F2/spd_skill": {
      "brainstorming": {
        "runs_with_stage": 2,
        "tokens_mean": 253890,
        "tokens_median": 253890.0,
        "out_tokens_mean": 6038,
        "usd_mean": 0.207,
        "usd_median": 0.207,
        "wall_s_mean": 53.05,
        "wall_s_median": 53.05
      },
      "writing-plans": {
        "runs_with_stage": 2,
        "tokens_mean": 397486.5,
        "tokens_median": 397486.5,
        "out_tokens_mean": 14059.5,
        "usd_mean": 0.307,
        "usd_median": 0.307,
        "wall_s_mean": 90.75,
        "wall_s_median": 90.75
      },
      "executing-plans": {
        "runs_with_stage": 2,
        "tokens_mean": 69973,
        "tokens_median": 69973.0,
        "out_tokens_mean": 286.5,
        "usd_mean": 0.046,
        "usd_median": 0.046,
        "wall_s_mean": 3.5,
        "wall_s_median": 3.5
      },
      "test-driven-development": {
        "runs_with_stage": 2,
        "tokens_mean": 3304676.5,
        "tokens_median": 3304676.5,
        "out_tokens_mean": 23243.5,
        "usd_mean": 1.052,
        "usd_median": 1.052,
        "wall_s_mean": 361.25,
        "wall_s_median": 361.25
      },
      "test-driven-development [subagent]": {
        "runs_with_stage": 2,
        "tokens_mean": 464339,
        "tokens_median": 464339.0,
        "out_tokens_mean": 13017,
        "usd_mean": 1.602,
        "usd_median": 1.602,
        "wall_s_mean": 0.0,
        "wall_s_median": 0.0
      }
    },
    "F2/spd_phase": {
      "brainstorming": {
        "runs_with_stage": 2,
        "tokens_mean": 253890,
        "tokens_median": 253890.0,
        "out_tokens_mean": 6038,
        "usd_mean": 0.207,
        "usd_median": 0.207,
        "wall_s_mean": 53.05,
        "wall_s_median": 53.05
      },
      "writing-plans": {
        "runs_with_stage": 2,
        "tokens_mean": 397486.5,
        "tokens_median": 397486.5,
        "out_tokens_mean": 14059.5,
        "usd_mean": 0.307,
        "usd_median": 0.307,
        "wall_s_mean": 90.75,
        "wall_s_median": 90.75
      },
      "execution": {
        "runs_with_stage": 2,
        "tokens_mean": 3374649.5,
        "tokens_median": 3374649.5,
        "out_tokens_mean": 23530,
        "usd_mean": 1.098,
        "usd_median": 1.098,
        "wall_s_mean": 364.75,
        "wall_s_median": 364.75
      },
      "subagents (reviewers, explorers)": {
        "runs_with_stage": 2,
        "tokens_mean": 464339,
        "tokens_median": 464339.0,
        "out_tokens_mean": 13017,
        "usd_mean": 1.602,
        "usd_median": 1.602,
        "wall_s_mean": 0.0,
        "wall_s_median": 0.0
      }
    },
    "F2/spd_slash": {
      "/brainstorming": {
        "runs_with_stage": 2,
        "tokens_mean": 485364,
        "tokens_median": 485364.0,
        "out_tokens_mean": null,
        "usd_mean": 0.354,
        "usd_median": 0.354,
        "wall_s_mean": 96.815,
        "wall_s_median": 96.815
      },
      "/writing-plans": {
        "runs_with_stage": 2,
        "tokens_mean": 166012.5,
        "tokens_median": 166012.5,
        "out_tokens_mean": null,
        "usd_mean": 0.159,
        "usd_median": 0.159,
        "wall_s_mean": 49.755,
        "wall_s_median": 49.755
      },
      "/executing-plans": {
        "runs_with_stage": 2,
        "tokens_mean": 3838988.5,
        "tokens_median": 3838988.5,
        "out_tokens_mean": null,
        "usd_mean": 2.7,
        "usd_median": 2.7,
        "wall_s_mean": 365.1,
        "wall_s_median": 365.1
      }
    },
    "F2/mp": {
      "/grill-with-docs": {
        "runs_with_stage": 2,
        "tokens_mean": 232965,
        "tokens_median": 232965.0,
        "out_tokens_mean": 5946,
        "usd_mean": 0.182,
        "usd_median": 0.182,
        "wall_s_mean": 56.7,
        "wall_s_median": 56.7
      },
      "/to-spec": {
        "runs_with_stage": 2,
        "tokens_mean": 180084.5,
        "tokens_median": 180084.5,
        "out_tokens_mean": 4452,
        "usd_mean": 0.115,
        "usd_median": 0.115,
        "wall_s_mean": 37.6,
        "wall_s_median": 37.6
      },
      "/to-tickets": {
        "runs_with_stage": 2,
        "tokens_mean": 253688.5,
        "tokens_median": 253688.5,
        "out_tokens_mean": 5152.5,
        "usd_mean": 0.131,
        "usd_median": 0.131,
        "wall_s_mean": 38.5,
        "wall_s_median": 38.5
      },
      "/implement": {
        "runs_with_stage": 2,
        "tokens_mean": 2689071,
        "tokens_median": 2689071.0,
        "out_tokens_mean": 31452,
        "usd_mean": 1.159,
        "usd_median": 1.159,
        "wall_s_mean": 218.35,
        "wall_s_median": 218.35
      },
      "/code-review": {
        "runs_with_stage": 2,
        "tokens_mean": 1107122.5,
        "tokens_median": 1107122.5,
        "out_tokens_mean": 9453.5,
        "usd_mean": 0.458,
        "usd_median": 0.458,
        "wall_s_mean": 63.85,
        "wall_s_median": 63.85
      },
      "/pr": {
        "runs_with_stage": 1,
        "tokens_mean": 115414,
        "tokens_median": 115414,
        "out_tokens_mean": 1372,
        "usd_mean": 0.043,
        "usd_median": 0.043,
        "wall_s_mean": 12.4,
        "wall_s_median": 12.4
      }
    }
  },
  "runs": [
    {
      "run": "bare-r1",
      "task": "F1",
      "cfg": "bare",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 149,
      "code_diff_removed": 11,
      "diff_added": 150,
      "diff_removed": 12,
      "tests_added_suite": 6,
      "tokens": {
        "in": 42,
        "out": 16265,
        "cr": 786548,
        "cw": 37896,
        "total": 840751
      },
      "usd": 0.4716,
      "usd_po": 0.0321,
      "wall_s": 118.8,
      "driver_turns": 2,
      "questions_routed_to_po": 1,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [],
      "superpowers_injected": false,
      "commits": 0,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "bare": {
          "turns": [
            1,
            2
          ],
          "in": 42,
          "out": 16265,
          "cr": 786548,
          "cw": 37896,
          "total": 840751,
          "usd": 0.4716,
          "wall_s": 118.8,
          "api_calls": 21
        }
      }
    },
    {
      "run": "bare-r2",
      "task": "F1",
      "cfg": "bare",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 118,
      "code_diff_removed": 10,
      "diff_added": 118,
      "diff_removed": 10,
      "tests_added_suite": 4,
      "tokens": {
        "in": 22,
        "out": 9513,
        "cr": 316959,
        "cw": 28254,
        "total": 354748
      },
      "usd": 0.2716,
      "usd_po": 0.016,
      "wall_s": 63.0,
      "driver_turns": 1,
      "questions_routed_to_po": 0,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [],
      "superpowers_injected": false,
      "commits": 0,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "bare": {
          "turns": [
            1
          ],
          "in": 22,
          "out": 9513,
          "cr": 316959,
          "cw": 28254,
          "total": 354748,
          "usd": 0.2716,
          "wall_s": 63.0,
          "api_calls": 11
        }
      }
    },
    {
      "run": "bare-r3",
      "task": "F1",
      "cfg": "bare",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 125,
      "code_diff_removed": 12,
      "diff_added": 125,
      "diff_removed": 12,
      "tests_added_suite": 4,
      "tokens": {
        "in": 22,
        "out": 10225,
        "cr": 321705,
        "cw": 29890,
        "total": 361842
      },
      "usd": 0.2862,
      "usd_po": 0.0148,
      "wall_s": 71.7,
      "driver_turns": 1,
      "questions_routed_to_po": 0,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [],
      "superpowers_injected": false,
      "commits": 0,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "bare": {
          "turns": [
            1
          ],
          "in": 22,
          "out": 10225,
          "cr": 321705,
          "cw": 29890,
          "total": 361842,
          "usd": 0.2862,
          "wall_s": 71.7,
          "api_calls": 11
        }
      }
    },
    {
      "run": "both-r1",
      "task": "F1",
      "cfg": "both",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 356,
      "code_diff_removed": 18,
      "diff_added": 543,
      "diff_removed": 18,
      "tests_added_suite": 22,
      "tokens": {
        "in": 192,
        "out": 57632,
        "cr": 5502518,
        "cw": 194012,
        "total": 5754354
      },
      "usd": 2.3259,
      "usd_po": 0.2185,
      "wall_s": 448.2,
      "driver_turns": 12,
      "questions_routed_to_po": 6,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": true,
      "commits": 7,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3,
            4
          ],
          "in": 22,
          "out": 6193,
          "cr": 313031,
          "cw": 25145,
          "total": 344391,
          "usd": 0.2252,
          "wall_s": 61.8,
          "api_calls": 11
        },
        "/to-spec": {
          "turns": [
            5,
            6
          ],
          "in": 12,
          "out": 4835,
          "cr": 204355,
          "cw": 8370,
          "total": 217572,
          "usd": 0.1227,
          "wall_s": 42.9,
          "api_calls": 5
        },
        "/to-tickets": {
          "turns": [
            7,
            8
          ],
          "in": 16,
          "out": 7115,
          "cr": 362815,
          "cw": 13007,
          "total": 382953,
          "usd": 0.1958,
          "wall_s": 49.4,
          "api_calls": 7
        },
        "/implement": {
          "turns": [
            9
          ],
          "in": 92,
          "out": 24618,
          "cr": 2960003,
          "cw": 85866,
          "total": 3070579,
          "usd": 1.1123,
          "wall_s": 194.2,
          "api_calls": 36
        },
        "/code-review": {
          "turns": [
            10,
            11
          ],
          "in": 46,
          "out": 13410,
          "cr": 1541663,
          "cw": 60146,
          "total": 1615265,
          "usd": 0.6253,
          "wall_s": 86.8,
          "api_calls": 13
        },
        "/pr": {
          "turns": [
            12
          ],
          "in": 4,
          "out": 1461,
          "cr": 120651,
          "cw": 1478,
          "total": 123594,
          "usd": 0.0447,
          "wall_s": 13.1,
          "api_calls": 1
        }
      }
    },
    {
      "run": "f2-bare-r1",
      "task": "F2",
      "cfg": "bare",
      "hidden_pass": 4,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": false,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 204,
      "code_diff_removed": 10,
      "diff_added": 208,
      "diff_removed": 10,
      "tests_added_suite": 5,
      "tokens": {
        "in": 28,
        "out": 15066,
        "cr": 444890,
        "cw": 34247,
        "total": 494231
      },
      "usd": 0.3767,
      "usd_po": 0.0167,
      "wall_s": 91.6,
      "driver_turns": 1,
      "questions_routed_to_po": 0,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [],
      "superpowers_injected": false,
      "commits": 0,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "bare": {
          "turns": [
            1
          ],
          "in": 28,
          "out": 15066,
          "cr": 444890,
          "cw": 34247,
          "total": 494231,
          "usd": 0.3767,
          "wall_s": 91.6,
          "api_calls": 14
        }
      }
    },
    {
      "run": "f2-bare-r2",
      "task": "F2",
      "cfg": "bare",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 189,
      "code_diff_removed": 10,
      "diff_added": 191,
      "diff_removed": 10,
      "tests_added_suite": 5,
      "tokens": {
        "in": 42,
        "out": 15884,
        "cr": 757517,
        "cw": 34970,
        "total": 808413
      },
      "usd": 0.4503,
      "usd_po": 0.0453,
      "wall_s": 114.7,
      "driver_turns": 3,
      "questions_routed_to_po": 1,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [],
      "superpowers_injected": false,
      "commits": 1,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "bare": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 42,
          "out": 15884,
          "cr": 757517,
          "cw": 34970,
          "total": 808413,
          "usd": 0.4503,
          "wall_s": 114.7,
          "api_calls": 21
        }
      }
    },
    {
      "run": "f2-mp-r1",
      "task": "F2",
      "cfg": "mp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 460,
      "code_diff_removed": 23,
      "diff_added": 666,
      "diff_removed": 25,
      "tests_added_suite": 31,
      "tokens": {
        "in": 172,
        "out": 60658,
        "cr": 4559790,
        "cw": 202896,
        "total": 4823516
      },
      "usd": 2.1824,
      "usd_po": 0.1989,
      "wall_s": 443.0,
      "driver_turns": 12,
      "questions_routed_to_po": 7,
      "end_reason": "cap_driver_turns",
      "caps": [
        "driver_turns_12"
      ],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": false,
      "commits": 7,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 18,
          "out": 6627,
          "cr": 229372,
          "cw": 22326,
          "total": 258343,
          "usd": 0.2015,
          "wall_s": 62.8,
          "api_calls": 9
        },
        "/to-spec": {
          "turns": [
            4,
            5
          ],
          "in": 12,
          "out": 5218,
          "cr": 195137,
          "cw": 10113,
          "total": 210480,
          "usd": 0.1317,
          "wall_s": 45.4,
          "api_calls": 5
        },
        "/to-tickets": {
          "turns": [
            6,
            7,
            8
          ],
          "in": 16,
          "out": 7288,
          "cr": 356825,
          "cw": 10505,
          "total": 374634,
          "usd": 0.1863,
          "wall_s": 53.0,
          "api_calls": 7
        },
        "/implement": {
          "turns": [
            9,
            10
          ],
          "in": 92,
          "out": 34442,
          "cr": 2978947,
          "cw": 104921,
          "total": 3118402,
          "usd": 1.2795,
          "wall_s": 238.8,
          "api_calls": 36
        },
        "/code-review": {
          "turns": [
            11,
            12
          ],
          "in": 34,
          "out": 7083,
          "cr": 799509,
          "cw": 55031,
          "total": 861657,
          "usd": 0.3835,
          "wall_s": 43.1,
          "api_calls": 6
        }
      }
    },
    {
      "run": "f2-mp-r2",
      "task": "F2",
      "cfg": "mp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 440,
      "code_diff_removed": 15,
      "diff_added": 607,
      "diff_removed": 18,
      "tests_added_suite": 31,
      "tokens": {
        "in": 152,
        "out": 53626,
        "cr": 3976936,
        "cw": 187047,
        "total": 4217761
      },
      "usd": 1.9511,
      "usd_po": 0.2073,
      "wall_s": 399.4,
      "driver_turns": 12,
      "questions_routed_to_po": 6,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": false,
      "commits": 2,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 16,
          "out": 5265,
          "cr": 184055,
          "cw": 18251,
          "total": 207587,
          "usd": 0.1625,
          "wall_s": 50.6,
          "api_calls": 8
        },
        "/to-spec": {
          "turns": [
            4,
            5
          ],
          "in": 10,
          "out": 3686,
          "cr": 137547,
          "cw": 8446,
          "total": 149689,
          "usd": 0.0982,
          "wall_s": 29.8,
          "api_calls": 4
        },
        "/to-tickets": {
          "turns": [
            6,
            7
          ],
          "in": 8,
          "out": 3017,
          "cr": 124377,
          "cw": 5341,
          "total": 132743,
          "usd": 0.0764,
          "wall_s": 24.0,
          "api_calls": 3
        },
        "/implement": {
          "turns": [
            8
          ],
          "in": 74,
          "out": 28462,
          "cr": 2129926,
          "cw": 101278,
          "total": 2259740,
          "usd": 1.0391,
          "wall_s": 197.9,
          "api_calls": 27
        },
        "/code-review": {
          "turns": [
            9,
            10,
            11
          ],
          "in": 43,
          "out": 11824,
          "cr": 1288603,
          "cw": 52118,
          "total": 1352588,
          "usd": 0.5323,
          "wall_s": 84.6,
          "api_calls": 11
        },
        "/pr": {
          "turns": [
            12
          ],
          "in": 1,
          "out": 1372,
          "cr": 112428,
          "cw": 1613,
          "total": 115414,
          "usd": 0.0427,
          "wall_s": 12.4,
          "api_calls": 1
        }
      }
    },
    {
      "run": "f2-spd-r1",
      "task": "F2",
      "cfg": "spd",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 546,
      "code_diff_removed": 14,
      "diff_added": 856,
      "diff_removed": 17,
      "tests_added_suite": 46,
      "tokens": {
        "in": 380,
        "out": 56881,
        "cr": 5074733,
        "cw": 178622,
        "total": 5310616
      },
      "usd": 3.4244,
      "usd_po": 0.09,
      "wall_s": 506.3,
      "driver_turns": 6,
      "questions_routed_to_po": 3,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "writing-plans",
        "test-driven-development"
      ],
      "superpowers_injected": true,
      "commits": 8,
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "stages": {
        "brainstorming": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 18,
          "out": 8103,
          "cr": 280105,
          "cw": 28324,
          "total": 316550,
          "usd": 0.2504,
          "wall_s": 64.3,
          "api_calls": 9
        },
        "writing-plans": {
          "turns": [
            3,
            4,
            5
          ],
          "in": 20,
          "out": 13028,
          "cr": 491569,
          "cw": 24216,
          "total": 528833,
          "usd": 0.3255,
          "wall_s": 89.2,
          "api_calls": 9
        },
        "writing-plans [subagent]": {
          "turns": [
            4
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "executing-plans": {
          "turns": [
            6
          ],
          "in": 4,
          "out": 291,
          "cr": 65551,
          "cw": 7617,
          "total": 73463,
          "usd": 0.0465,
          "wall_s": 4.0,
          "api_calls": 1
        },
        "test-driven-development": {
          "turns": [
            6
          ],
          "in": 80,
          "out": 24030,
          "cr": 3786616,
          "cw": 42748,
          "total": 3853474,
          "usd": 1.1688,
          "wall_s": 345.9,
          "api_calls": 40
        },
        "test-driven-development [subagent]": {
          "turns": [
            6
          ],
          "in": 258,
          "out": 11429,
          "cr": 450892,
          "cw": 75717,
          "total": 538296,
          "usd": 1.6332,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "f2-spd-r2",
      "task": "F2",
      "cfg": "spd",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 401,
      "code_diff_removed": 16,
      "diff_added": 611,
      "diff_removed": 17,
      "tests_added_suite": 20,
      "tokens": {
        "in": 316,
        "out": 56408,
        "cr": 3456411,
        "cw": 156979,
        "total": 3670114
      },
      "usd": 3.0022,
      "usd_po": 0.1126,
      "wall_s": 517.1,
      "driver_turns": 7,
      "questions_routed_to_po": 4,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "test-driven-development"
      ],
      "superpowers_injected": true,
      "commits": 6,
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "stages": {
        "brainstorming": {
          "turns": [
            1,
            2,
            3,
            4
          ],
          "in": 12,
          "out": 3973,
          "cr": 164538,
          "cw": 22707,
          "total": 191230,
          "usd": 0.1635,
          "wall_s": 41.8,
          "api_calls": 6
        },
        "brainstorming [subagent]": {
          "turns": [
            1,
            3
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "writing-plans": {
          "turns": [
            5,
            6
          ],
          "in": 12,
          "out": 15091,
          "cr": 228188,
          "cw": 22849,
          "total": 266140,
          "usd": 0.288,
          "wall_s": 92.3,
          "api_calls": 5
        },
        "writing-plans [subagent]": {
          "turns": [
            6
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "executing-plans": {
          "turns": [
            7
          ],
          "in": 4,
          "out": 282,
          "cr": 58567,
          "cw": 7630,
          "total": 66483,
          "usd": 0.0451,
          "wall_s": 3.0,
          "api_calls": 1
        },
        "test-driven-development": {
          "turns": [
            7
          ],
          "in": 62,
          "out": 22457,
          "cr": 2690340,
          "cw": 43020,
          "total": 2755879,
          "usd": 0.9348,
          "wall_s": 376.6,
          "api_calls": 31
        },
        "test-driven-development [subagent]": {
          "turns": [
            7
          ],
          "in": 226,
          "out": 14605,
          "cr": 314778,
          "cw": 60773,
          "total": 390382,
          "usd": 1.5709,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "mp-r1",
      "task": "F1",
      "cfg": "mp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 349,
      "code_diff_removed": 20,
      "diff_added": 524,
      "diff_removed": 20,
      "tests_added_suite": 29,
      "tokens": {
        "in": 181,
        "out": 60902,
        "cr": 4842442,
        "cw": 194978,
        "total": 5098503
      },
      "usd": 2.2217,
      "usd_po": 0.2058,
      "wall_s": 472.4,
      "driver_turns": 12,
      "questions_routed_to_po": 7,
      "end_reason": "cap_driver_turns",
      "caps": [
        "driver_turns_12"
      ],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": false,
      "commits": 7,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 24,
          "out": 6223,
          "cr": 320351,
          "cw": 22264,
          "total": 348862,
          "usd": 0.2154,
          "wall_s": 59.9,
          "api_calls": 12
        },
        "/to-spec": {
          "turns": [
            4,
            5
          ],
          "in": 10,
          "out": 4516,
          "cr": 153800,
          "cw": 9252,
          "total": 167578,
          "usd": 0.1129,
          "wall_s": 37.6,
          "api_calls": 4
        },
        "/to-tickets": {
          "turns": [
            6,
            7,
            8
          ],
          "in": 16,
          "out": 7382,
          "cr": 349965,
          "cw": 10865,
          "total": 368228,
          "usd": 0.1873,
          "wall_s": 55.8,
          "api_calls": 7
        },
        "/implement": {
          "turns": [
            9,
            10
          ],
          "in": 92,
          "out": 32290,
          "cr": 2890397,
          "cw": 100256,
          "total": 3023035,
          "usd": 1.226,
          "wall_s": 251.2,
          "api_calls": 35
        },
        "/code-review": {
          "turns": [
            11,
            12
          ],
          "in": 39,
          "out": 10491,
          "cr": 1127929,
          "cw": 52341,
          "total": 1190800,
          "usd": 0.4801,
          "wall_s": 67.9,
          "api_calls": 9
        }
      }
    },
    {
      "run": "mp-r2",
      "task": "F1",
      "cfg": "mp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 314,
      "code_diff_removed": 15,
      "diff_added": 527,
      "diff_removed": 15,
      "tests_added_suite": 25,
      "tokens": {
        "in": 154,
        "out": 48104,
        "cr": 3826019,
        "cw": 168538,
        "total": 4042815
      },
      "usd": 1.8045,
      "usd_po": 0.1988,
      "wall_s": 365.0,
      "driver_turns": 12,
      "questions_routed_to_po": 6,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": false,
      "commits": 7,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 16,
          "out": 4727,
          "cr": 187718,
          "cw": 18567,
          "total": 211028,
          "usd": 0.1591,
          "wall_s": 47.0,
          "api_calls": 8
        },
        "/to-spec": {
          "turns": [
            4,
            5
          ],
          "in": 10,
          "out": 4965,
          "cr": 136863,
          "cw": 8781,
          "total": 150619,
          "usd": 0.1122,
          "wall_s": 40.8,
          "api_calls": 4
        },
        "/to-tickets": {
          "turns": [
            6,
            7
          ],
          "in": 8,
          "out": 4035,
          "cr": 126982,
          "cw": 6551,
          "total": 137576,
          "usd": 0.092,
          "wall_s": 29.6,
          "api_calls": 3
        },
        "/implement": {
          "turns": [
            8,
            9
          ],
          "in": 78,
          "out": 24215,
          "cr": 2224544,
          "cw": 81760,
          "total": 2330597,
          "usd": 0.9551,
          "wall_s": 177.1,
          "api_calls": 31
        },
        "/code-review": {
          "turns": [
            10,
            11
          ],
          "in": 38,
          "out": 8813,
          "cr": 1047606,
          "cw": 51098,
          "total": 1107555,
          "usd": 0.4451,
          "wall_s": 58.9,
          "api_calls": 10
        },
        "/pr": {
          "turns": [
            12
          ],
          "in": 4,
          "out": 1349,
          "cr": 102306,
          "cw": 1781,
          "total": 105440,
          "usd": 0.0411,
          "wall_s": 11.6,
          "api_calls": 1
        }
      }
    },
    {
      "run": "mp-r3",
      "task": "F1",
      "cfg": "mp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 395,
      "code_diff_removed": 23,
      "diff_added": 625,
      "diff_removed": 23,
      "tests_added_suite": 29,
      "tokens": {
        "in": 167,
        "out": 61871,
        "cr": 4014893,
        "cw": 210062,
        "total": 4286993
      },
      "usd": 2.1024,
      "usd_po": 0.2105,
      "wall_s": 479.7,
      "driver_turns": 12,
      "questions_routed_to_po": 8,
      "end_reason": "cap_driver_turns",
      "caps": [
        "driver_turns_12"
      ],
      "skills_invoked": [
        "grilling",
        "domain-modeling",
        "tdd",
        "code-review"
      ],
      "superpowers_injected": false,
      "commits": 5,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "/grill-with-docs": {
          "turns": [
            1,
            2,
            3,
            4
          ],
          "in": 26,
          "out": 7047,
          "cr": 358800,
          "cw": 23759,
          "total": 389632,
          "usd": 0.2373,
          "wall_s": 72.9,
          "api_calls": 13
        },
        "/to-spec": {
          "turns": [
            5,
            6
          ],
          "in": 10,
          "out": 5923,
          "cr": 156583,
          "cw": 9319,
          "total": 171835,
          "usd": 0.1278,
          "wall_s": 47.9,
          "api_calls": 4
        },
        "/to-tickets": {
          "turns": [
            7,
            8,
            9
          ],
          "in": 9,
          "out": 6185,
          "cr": 252224,
          "cw": 9644,
          "total": 268062,
          "usd": 0.1509,
          "wall_s": 47.8,
          "api_calls": 5
        },
        "/implement": {
          "turns": [
            10,
            11
          ],
          "in": 82,
          "out": 34139,
          "cr": 2465960,
          "cw": 112881,
          "total": 2613062,
          "usd": 1.1946,
          "wall_s": 245.4,
          "api_calls": 27
        },
        "/code-review": {
          "turns": [
            12
          ],
          "in": 40,
          "out": 8577,
          "cr": 781326,
          "cw": 54459,
          "total": 844402,
          "usd": 0.3917,
          "wall_s": 65.8,
          "api_calls": 5
        }
      }
    },
    {
      "run": "sp-r1",
      "task": "F1",
      "cfg": "sp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 182,
      "code_diff_removed": 15,
      "diff_added": 183,
      "diff_removed": 16,
      "tests_added_suite": 10,
      "tokens": {
        "in": 28,
        "out": 14522,
        "cr": 553287,
        "cw": 45597,
        "total": 613434
      },
      "usd": 0.4383,
      "usd_po": 0.0424,
      "wall_s": 99.4,
      "driver_turns": 3,
      "questions_routed_to_po": 2,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "brainstorming"
      ],
      "superpowers_injected": true,
      "commits": 1,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "(none)": {
          "turns": [
            1
          ],
          "in": 2,
          "out": 209,
          "cr": 13011,
          "cw": 6298,
          "total": 19520,
          "usd": 0.0299,
          "wall_s": 5.4,
          "api_calls": 1
        },
        "brainstorming": {
          "turns": [
            1,
            2
          ],
          "in": 14,
          "out": 9666,
          "cr": 212052,
          "cw": 26168,
          "total": 247900,
          "usd": 0.2438,
          "wall_s": 40.1,
          "api_calls": 7
        },
        "implementation (no skill)": {
          "turns": [
            2,
            3
          ],
          "in": 12,
          "out": 4647,
          "cr": 328224,
          "cw": 13131,
          "total": 346014,
          "usd": 0.1647,
          "wall_s": 52.8,
          "api_calls": 6
        },
        "implementation (no skill) [subagent]": {
          "turns": [
            2
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "sp-r2",
      "task": "F1",
      "cfg": "sp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 212,
      "code_diff_removed": 15,
      "diff_added": 216,
      "diff_removed": 15,
      "tests_added_suite": 13,
      "tokens": {
        "in": 32,
        "out": 18160,
        "cr": 691441,
        "cw": 55475,
        "total": 765108
      },
      "usd": 0.5419,
      "usd_po": 0.0305,
      "wall_s": 119.6,
      "driver_turns": 2,
      "questions_routed_to_po": 1,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "brainstorming",
        "test-driven-development"
      ],
      "superpowers_injected": true,
      "commits": 0,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "(none)": {
          "turns": [
            1
          ],
          "in": 2,
          "out": 227,
          "cr": 13011,
          "cw": 6298,
          "total": 19538,
          "usd": 0.0301,
          "wall_s": 5.7,
          "api_calls": 1
        },
        "brainstorming": {
          "turns": [
            1,
            2
          ],
          "in": 10,
          "out": 2311,
          "cr": 133731,
          "cw": 14200,
          "total": 150252,
          "usd": 0.1067,
          "wall_s": 20.4,
          "api_calls": 5
        },
        "test-driven-development": {
          "turns": [
            2
          ],
          "in": 20,
          "out": 15622,
          "cr": 544699,
          "cw": 34977,
          "total": 595318,
          "usd": 0.4051,
          "wall_s": 93.0,
          "api_calls": 10
        }
      }
    },
    {
      "run": "sp-r3",
      "task": "F1",
      "cfg": "sp",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 267,
      "code_diff_removed": 14,
      "diff_added": 267,
      "diff_removed": 14,
      "tests_added_suite": 14,
      "tokens": {
        "in": 34,
        "out": 17926,
        "cr": 727283,
        "cw": 50017,
        "total": 795260
      },
      "usd": 0.5249,
      "usd_po": 0.0489,
      "wall_s": 135.3,
      "driver_turns": 3,
      "questions_routed_to_po": 2,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "brainstorming"
      ],
      "superpowers_injected": true,
      "commits": 1,
      "models": [
        "claude-sonnet-5-5"
      ],
      "stages": {
        "(none)": {
          "turns": [
            1
          ],
          "in": 2,
          "out": 278,
          "cr": 13011,
          "cw": 6298,
          "total": 19589,
          "usd": 0.0306,
          "wall_s": 4.9,
          "api_calls": 1
        },
        "brainstorming": {
          "turns": [
            1,
            2
          ],
          "in": 18,
          "out": 12135,
          "cr": 306378,
          "cw": 31097,
          "total": 349628,
          "usd": 0.307,
          "wall_s": 61.1,
          "api_calls": 9
        },
        "implementation (no skill)": {
          "turns": [
            2,
            3
          ],
          "in": 14,
          "out": 5513,
          "cr": 407894,
          "cw": 12622,
          "total": 426043,
          "usd": 0.1872,
          "wall_s": 67.8,
          "api_calls": 7
        },
        "implementation (no skill) [subagent]": {
          "turns": [
            2
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "spd-r1",
      "task": "F1",
      "cfg": "spd",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 350,
      "code_diff_removed": 21,
      "diff_added": 516,
      "diff_removed": 21,
      "tests_added_suite": 21,
      "tokens": {
        "in": 340,
        "out": 59243,
        "cr": 3428307,
        "cw": 179396,
        "total": 3667286
      },
      "usd": 3.4118,
      "usd_po": 0.1084,
      "wall_s": 600.8,
      "driver_turns": 6,
      "questions_routed_to_po": 3,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "test-driven-development"
      ],
      "superpowers_injected": true,
      "commits": 4,
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "stages": {
        "brainstorming": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 14,
          "out": 5191,
          "cr": 198912,
          "cw": 25310,
          "total": 229427,
          "usd": 0.193,
          "wall_s": 49.8,
          "api_calls": 7
        },
        "brainstorming [subagent]": {
          "turns": [
            2
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "writing-plans": {
          "turns": [
            4,
            5
          ],
          "in": 12,
          "out": 13010,
          "cr": 243336,
          "cw": 23856,
          "total": 280214,
          "usd": 0.2742,
          "wall_s": 95.1,
          "api_calls": 5
        },
        "executing-plans": {
          "turns": [
            6
          ],
          "in": 4,
          "out": 614,
          "cr": 62177,
          "cw": 7788,
          "total": 70583,
          "usd": 0.0497,
          "wall_s": 4.6,
          "api_calls": 1
        },
        "test-driven-development": {
          "turns": [
            6
          ],
          "in": 52,
          "out": 22476,
          "cr": 2455319,
          "cw": 43404,
          "total": 2521251,
          "usd": 0.8895,
          "wall_s": 448.9,
          "api_calls": 26
        },
        "test-driven-development [subagent]": {
          "turns": [
            6
          ],
          "in": 258,
          "out": 17952,
          "cr": 468563,
          "cw": 79038,
          "total": 565811,
          "usd": 2.0053,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "spd-r2",
      "task": "F1",
      "cfg": "spd",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 317,
      "code_diff_removed": 16,
      "diff_added": 527,
      "diff_removed": 16,
      "tests_added_suite": 36,
      "tokens": {
        "in": 354,
        "out": 53270,
        "cr": 3666473,
        "cw": 166918,
        "total": 3887015
      },
      "usd": 3.3947,
      "usd_po": 0.1118,
      "wall_s": 520.0,
      "driver_turns": 7,
      "questions_routed_to_po": 4,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "test-driven-development"
      ],
      "superpowers_injected": true,
      "commits": 5,
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "stages": {
        "brainstorming": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 10,
          "out": 3724,
          "cr": 134220,
          "cw": 22880,
          "total": 160834,
          "usd": 0.1556,
          "wall_s": 33.6,
          "api_calls": 5
        },
        "writing-plans": {
          "turns": [
            4,
            5,
            6
          ],
          "in": 14,
          "out": 8546,
          "cr": 271872,
          "cw": 17835,
          "total": 298267,
          "usd": 0.2112,
          "wall_s": 63.9,
          "api_calls": 6
        },
        "writing-plans [subagent]": {
          "turns": [
            6
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "executing-plans": {
          "turns": [
            7
          ],
          "in": 4,
          "out": 385,
          "cr": 53726,
          "cw": 7377,
          "total": 61492,
          "usd": 0.0441,
          "wall_s": 3.7,
          "api_calls": 1
        },
        "test-driven-development": {
          "turns": [
            7
          ],
          "in": 68,
          "out": 21023,
          "cr": 2791978,
          "cw": 41898,
          "total": 2854967,
          "usd": 0.9364,
          "wall_s": 415.1,
          "api_calls": 34
        },
        "test-driven-development [subagent]": {
          "turns": [
            7
          ],
          "in": 258,
          "out": 19592,
          "cr": 414677,
          "cw": 76928,
          "total": 511455,
          "usd": 2.0474,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    },
    {
      "run": "spd-r3",
      "task": "F1",
      "cfg": "spd",
      "hidden_pass": 5,
      "hidden_reqs": {
        "test_req1": true,
        "test_req2": true,
        "test_req3": true,
        "test_req4": true,
        "test_req5": true
      },
      "suite_green": true,
      "typecheck_green": true,
      "code_diff_added": 434,
      "code_diff_removed": 20,
      "diff_added": 646,
      "diff_removed": 20,
      "tests_added_suite": 26,
      "tokens": {
        "in": 504,
        "out": 72541,
        "cr": 5615970,
        "cw": 196313,
        "total": 5885328
      },
      "usd": 4.4087,
      "usd_po": 0.1286,
      "wall_s": 701.3,
      "driver_turns": 8,
      "questions_routed_to_po": 5,
      "end_reason": "done",
      "caps": [],
      "skills_invoked": [
        "test-driven-development",
        "finishing-a-development-branch"
      ],
      "superpowers_injected": true,
      "commits": 6,
      "models": [
        "claude-fable-5-1",
        "claude-sonnet-5-5"
      ],
      "stages": {
        "brainstorming": {
          "turns": [
            1,
            2,
            3
          ],
          "in": 10,
          "out": 3567,
          "cr": 130512,
          "cw": 21715,
          "total": 155804,
          "usd": 0.1487,
          "wall_s": 33.0,
          "api_calls": 5
        },
        "writing-plans": {
          "turns": [
            4,
            5,
            6
          ],
          "in": 16,
          "out": 13925,
          "cr": 345667,
          "cw": 25975,
          "total": 385583,
          "usd": 0.3123,
          "wall_s": 101.2,
          "api_calls": 7
        },
        "writing-plans [subagent]": {
          "turns": [
            5,
            6
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "executing-plans": {
          "turns": [
            7
          ],
          "in": 4,
          "out": 516,
          "cr": 60701,
          "cw": 7518,
          "total": 68739,
          "usd": 0.0474,
          "wall_s": 5.5,
          "api_calls": 1
        },
        "test-driven-development": {
          "turns": [
            7
          ],
          "in": 72,
          "out": 24026,
          "cr": 3334129,
          "cw": 46850,
          "total": 3405077,
          "usd": 1.0946,
          "wall_s": 507.1,
          "api_calls": 36
        },
        "finishing-a-development-branch": {
          "turns": [
            7,
            8
          ],
          "in": 16,
          "out": 5297,
          "cr": 962018,
          "cw": 9174,
          "total": 976505,
          "usd": 0.2821,
          "wall_s": 50.7,
          "api_calls": 8
        },
        "test-driven-development [subagent]": {
          "turns": [
            7
          ],
          "in": 386,
          "out": 25210,
          "cr": 782943,
          "cw": 85081,
          "total": 893620,
          "usd": 2.5236,
          "wall_s": 0.0,
          "api_calls": 0
        },
        "finishing-a-development-branch [subagent]": {
          "turns": [
            8
          ],
          "in": 0,
          "out": 0,
          "cr": 0,
          "cw": 0,
          "total": 0,
          "usd": 0.0,
          "wall_s": 0.0,
          "api_calls": 0
        }
      }
    }
  ]
}