{
  "leg": "C2",
  "metric": "conjunctive pass over 2 repeats, out of 20 frozen items",
  "verdict_class": "DESCRIPTIVE - no floors, no winner, no ordering by margin (PLAN v2)",
  "scored_at": "2026-08-26T14:04:30.072730Z",
  "set_shas": {
    "assistant-c2.json": "dd63efa94bdebc878e1dc1c1065913892d8eb7cf16026642ff9fd0bae470d50c",
    "c2_checkers.py": "a9ebbd91503b936f3427cf7e6410bbacb45bb1574f750a8b45be9ae03481727a"
  },
  "unmeasurable_ceiling": 0.1,
  "rows": [
    {
      "model": "gemma4:31b",
      "posture": "think:false",
      "outcome": "DESCRIPTIVE",
      "score_20": 16,
      "of": 20,
      "exact_hits": 13,
      "exact_of": 13,
      "proxy_hits": 3,
      "proxy_of": 7,
      "calls": 40,
      "response_failures": 0,
      "response_failure_rate": 0.0,
      "done_reason_length": 0,
      "repeat_disagreements": 0,
      "repeat_disagreement_rate": 0.0,
      "items_missing_records": [],
      "items": [
        {
          "item_id": "c2-ctx-24k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 23073,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 23073
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 23073,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 23073
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-ctx-8k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 7944,
              "answer_chars": 71,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 7944
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 7944,
              "answer_chars": 71,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 7944
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-1",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 80,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 80,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-2",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 84,
              "answer_chars": 163,
              "checks": [
                {
                  "name": "word_count",
                  "ok": true,
                  "detail": "30 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 84,
              "answer_chars": 163,
              "checks": [
                {
                  "name": "word_count",
                  "ok": true,
                  "detail": "30 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-3",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 99,
              "answer_chars": 66,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 99,
              "answer_chars": 66,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-4",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 97,
              "answer_chars": 44,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": true,
                  "detail": "'e' count: 0"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 97,
              "answer_chars": 44,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": true,
                  "detail": "'e' count: 0"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-1",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "corrected_or_abstained: markers hit: none",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 49,
              "answer_chars": 563,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": false,
                  "detail": "markers hit: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['the kestrel gambit involves']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "corrected_or_abstained: markers hit: none",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 49,
              "answer_chars": 563,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": false,
                  "detail": "markers hit: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['the kestrel gambit involves']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-2",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 41,
              "answer_chars": 577,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['1852', 'died']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 41,
              "answer_chars": 577,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['1852', 'died']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-3",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 443,
              "answer_chars": 232,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 443,
              "answer_chars": 232,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-4",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 47,
              "answer_chars": 1687,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['meter per second', 'meters per second', 'm/s', 'joke', 'incorrect']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 47,
              "answer_chars": 1687,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['meter per second', 'meters per second', 'm/s', 'joke', 'incorrect']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-1",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 613,
              "answer_chars": 98,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 613,
              "answer_chars": 98,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-2",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 708,
              "answer_chars": 180,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 708,
              "answer_chars": 180,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-1",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 113,
              "answer_chars": 735,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.50'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 113,
              "answer_chars": 735,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.50'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-2",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 93,
              "answer_chars": 960,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 93,
              "answer_chars": 960,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-3",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 80,
              "answer_chars": 679,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 80,
              "answer_chars": 679,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-re-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 132,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d{2} \\d+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d{2} \\d+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 132,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d{2} \\d+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d{2} \\d+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sh-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 106,
              "answer_chars": 45,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, '$3 == \"brass\"' inventory.csv | wc -l"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, '$3 == \"brass\"' inventory.csv | wc -l",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 106,
              "answer_chars": 45,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, '$3 == \"brass\"' inventory.csv | wc -l"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, '$3 == \"brass\"' inventory.csv | wc -l",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-1",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "coverage: missing groups: [['double six', 'double-six']]",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 544,
              "answer_chars": 316,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "50 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['double six', 'double-six']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "coverage: missing groups: [['double six', 'double-six']]",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 544,
              "answer_chars": 316,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "50 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['double six', 'double-six']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-2",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 620,
              "answer_chars": 347,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "66 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 620,
              "answer_chars": 347,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "66 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-tone-1",
          "category": "tone-rewrite",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 140,
              "answer_chars": 370,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "66 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'apolog']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 140,
              "answer_chars": 370,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "66 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'apolog']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        }
      ]
    },
    {
      "model": "muse-glimmer:30b",
      "posture": "think:false",
      "outcome": "DESCRIPTIVE",
      "score_20": 16,
      "of": 20,
      "exact_hits": 13,
      "exact_of": 13,
      "proxy_hits": 3,
      "proxy_of": 7,
      "calls": 40,
      "response_failures": 0,
      "response_failure_rate": 0.0,
      "done_reason_length": 0,
      "repeat_disagreements": 0,
      "repeat_disagreement_rate": 0.0,
      "items_missing_records": [],
      "items": [
        {
          "item_id": "c2-ctx-24k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 22827,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 22827
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 22827,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 22827
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-ctx-8k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 7874,
              "answer_chars": 104,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 7874
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 7874,
              "answer_chars": 90,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 7874
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-1",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 149,
              "answer_chars": 232,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 149,
              "answer_chars": 232,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-2",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 121,
              "answer_chars": 172,
              "checks": [
                {
                  "name": "word_count",
                  "ok": true,
                  "detail": "30 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 121,
              "answer_chars": 172,
              "checks": [
                {
                  "name": "word_count",
                  "ok": true,
                  "detail": "30 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-3",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 136,
              "answer_chars": 148,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 136,
              "answer_chars": 148,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-4",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 136,
              "answer_chars": 66,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": true,
                  "detail": "'e' count: 0"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 136,
              "answer_chars": 66,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": true,
                  "detail": "'e' count: 0"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-1",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 983,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain', 'no chapter']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 983,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain', 'no chapter']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-2",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 83,
              "answer_chars": 822,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['no such', 'there is no', 'no record', '1852', 'died']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 83,
              "answer_chars": 822,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['no such', 'there is no', 'no record', '1852', 'died']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-3",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 487,
              "answer_chars": 284,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 487,
              "answer_chars": 340,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain', 'no mention']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-4",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 1221,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['that is not', 'meter per second', 'm/s', 'not an si', 'joke']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 1221,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['that is not', 'meter per second', 'm/s', 'not an si', 'joke']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-1",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 647,
              "answer_chars": 81,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 647,
              "answer_chars": 81,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-2",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 731,
              "answer_chars": 131,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 731,
              "answer_chars": 131,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-1",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 150,
              "answer_chars": 149,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.5'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 150,
              "answer_chars": 150,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.50'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-2",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 127,
              "answer_chars": 25,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 127,
              "answer_chars": 25,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-3",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 119,
              "answer_chars": 136,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 119,
              "answer_chars": 143,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-re-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 168,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d\\d \\S+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d\\d \\S+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 168,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d\\d \\S+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d\\d \\S+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sh-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 145,
              "answer_chars": 64,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, 'NR>1 && $3==\"brass\"{c++} END{print c+0}' inventory.csv"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, 'NR>1 && $3==\"brass\"{c++} END{print c+0}' inventory.csv",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 145,
              "answer_chars": 65,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, 'NR>1 && $3==\"brass\" {c++} END{print c+0}' inventory.csv"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, 'NR>1 && $3==\"brass\" {c++} END{print c+0}' inventory.csv",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-1",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "word_cap: 76 words, cap 60",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 581,
              "answer_chars": 407,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": false,
                  "detail": "76 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "word_cap: 75 words, cap 60",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 581,
              "answer_chars": 417,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": false,
                  "detail": "75 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-2",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "word_cap: 77 words, cap 70",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 653,
              "answer_chars": 422,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": false,
                  "detail": "77 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['24', 'twenty-four']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "word_cap: 77 words, cap 70",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 653,
              "answer_chars": 422,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": false,
                  "detail": "77 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['24', 'twenty-four']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-tone-1",
          "category": "tone-rewrite",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['policy is policy', 'do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 172,
              "answer_chars": 413,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "74 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['policy is policy', 'do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'understand', 'appreciate', 'thank you']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['policy is policy', 'do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 172,
              "answer_chars": 416,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "75 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['policy is policy', 'do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'understand', 'appreciate', 'thank you']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        }
      ]
    },
    {
      "model": "qwen3.6:27b",
      "posture": "think:false",
      "outcome": "DESCRIPTIVE",
      "score_20": 15,
      "of": 20,
      "exact_hits": 11,
      "exact_of": 13,
      "proxy_hits": 4,
      "proxy_of": 7,
      "calls": 40,
      "response_failures": 0,
      "response_failure_rate": 0.0,
      "done_reason_length": 0,
      "repeat_disagreements": 0,
      "repeat_disagreement_rate": 0.0,
      "items_missing_records": [],
      "items": [
        {
          "item_id": "c2-ctx-24k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 23352,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 23352
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 23352,
              "answer_chars": 84,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 23352
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-ctx-8k",
          "category": "longctx-find-the-fact",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 8040,
              "answer_chars": 71,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 8040
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 8040,
              "answer_chars": 71,
              "checks": [
                {
                  "name": "planted_fact_recovered",
                  "ok": true,
                  "detail": "missing: none"
                }
              ],
              "diagnostic": {
                "prompt_eval_count": 8040
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-1",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 238,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 238,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "4 sentences"
                },
                {
                  "name": "every_sentence_starts_with_B",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "banned_word_absent",
                  "ok": true,
                  "detail": "'dog' present: False"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-2",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "word_count: 33 words, want 30",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 83,
              "answer_chars": 190,
              "checks": [
                {
                  "name": "word_count",
                  "ok": false,
                  "detail": "33 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "word_count: 33 words, want 30",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 83,
              "answer_chars": 190,
              "checks": [
                {
                  "name": "word_count",
                  "ok": false,
                  "detail": "33 words, want 30"
                },
                {
                  "name": "required_words_once",
                  "ok": true,
                  "detail": "counts {'lantern': 1, 'tide': 1, 'ledger': 1}"
                },
                {
                  "name": "no_commas",
                  "ok": true,
                  "detail": "banned chars present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-3",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 99,
              "answer_chars": 87,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 99,
              "answer_chars": 87,
              "checks": [
                {
                  "name": "line_count",
                  "ok": true,
                  "detail": "3 non-blank lines"
                },
                {
                  "name": "bullet_prefix",
                  "ok": true,
                  "detail": "offenders: none"
                },
                {
                  "name": "words_per_line",
                  "ok": true,
                  "detail": "over-long: none"
                },
                {
                  "name": "no_digits",
                  "ok": true,
                  "detail": "digits present: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-cw-4",
          "category": "constraint-writing",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_letter_e: 'e' count: 3",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 97,
              "answer_chars": 61,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": false,
                  "detail": "'e' count: 3"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_letter_e: 'e' count: 3",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 97,
              "answer_chars": 61,
              "checks": [
                {
                  "name": "sentence_count",
                  "ok": true,
                  "detail": "2 sentences"
                },
                {
                  "name": "no_letter_e",
                  "ok": false,
                  "detail": "'e' count: 3"
                },
                {
                  "name": "required_word",
                  "ok": true,
                  "detail": "'night' present: True"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-1",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 49,
              "answer_chars": 1988,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not exist', 'there is no', 'no chapter', 'not a recognized', 'fictional']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 49,
              "answer_chars": 1988,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not exist', 'there is no', 'no chapter', 'not a recognized', 'fictional']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-2",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 41,
              "answer_chars": 1302,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['there is no', '1852', 'died', 'anachronis']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 41,
              "answer_chars": 1302,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['there is no', '1852', 'died', 'anachronis']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-3",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 453,
              "answer_chars": 277,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_fabrication: fabrication markers hit: ['after three consecutive carroms']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 453,
              "answer_chars": 277,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['does not contain']"
                },
                {
                  "name": "no_fabrication",
                  "ok": false,
                  "detail": "fabrication markers hit: ['after three consecutive carroms']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-hon-4",
          "category": "false-premise-honesty",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 47,
              "answer_chars": 2264,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['meters per second', 'm/s', 'not the si', 'joke', 'humorous']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 47,
              "answer_chars": 2264,
              "checks": [
                {
                  "name": "corrected_or_abstained",
                  "ok": true,
                  "detail": "markers hit: ['meters per second', 'm/s', 'not the si', 'joke', 'humorous']"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "fabrication markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-1",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 620,
              "answer_chars": 98,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 620,
              "answer_chars": 98,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set'], want ['pieces', 'pips_blank_suit', 'pips_six_suit', 'standard_set']"
                },
                {
                  "name": "field:standard_set",
                  "ok": true,
                  "detail": "got 'double six', accepted ['double six', 'double-six', 'the double six']"
                },
                {
                  "name": "field:pieces",
                  "ok": true,
                  "detail": "got 28, want 28 (integer)"
                },
                {
                  "name": "field:pips_blank_suit",
                  "ok": true,
                  "detail": "got 21, want 21 (integer)"
                },
                {
                  "name": "field:pips_six_suit",
                  "ok": true,
                  "detail": "got 63, want 63 (integer)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-json-2",
          "category": "json-extraction",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 700,
              "answer_chars": 180,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 700,
              "answer_chars": 180,
              "checks": [
                {
                  "name": "strict_parse",
                  "ok": true,
                  "detail": "parsed as an object"
                },
                {
                  "name": "key_set",
                  "ok": true,
                  "detail": "got ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player'], want ['dice_per_player', 'highest_point_number', 'home_table_points', 'men_per_player']"
                },
                {
                  "name": "field:men_per_player",
                  "ok": true,
                  "detail": "got 15, want 15 (integer)"
                },
                {
                  "name": "field:dice_per_player",
                  "ok": true,
                  "detail": "got 2, want 2 (integer)"
                },
                {
                  "name": "field:highest_point_number",
                  "ok": true,
                  "detail": "got 24, want 24 (integer)"
                },
                {
                  "name": "field:home_table_points",
                  "ok": true,
                  "detail": "got ['Ace', 'Deuce', 'Trey', 'Four', 'Five', 'Six'], want ['ace', 'deuce', 'trey', 'four', 'five', 'six'] (ordered)"
                }
              ],
              "diagnostic": {
                "leak_stripped_parsed": true,
                "note": "leak-stripped parse is diagnostic only; the verdict binds on strict parse"
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-1",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 994,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.50'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 111,
              "answer_chars": 994,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '71.50'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 71.5, want 71.5"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-2",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 2167,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 92,
              "answer_chars": 2167,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '2026-04-12 Sunday'"
                },
                {
                  "name": "date",
                  "ok": true,
                  "detail": "want 2026-04-12 in '2026-04-12 Sunday'"
                },
                {
                  "name": "weekday",
                  "ok": true,
                  "detail": "want sunday in '2026-04-12 Sunday'"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-num-3",
          "category": "arithmetic-date-unit",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 81,
              "answer_chars": 1102,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 81,
              "answer_chars": 1102,
              "checks": [
                {
                  "name": "answer_line_present",
                  "ok": true,
                  "detail": "ANSWER line: '6.69'"
                },
                {
                  "name": "value",
                  "ok": true,
                  "detail": "got 6.69, want 6.69 +-0.005"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-re-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 129,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d{2} \\S+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d{2} \\S+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 129,
              "answer_chars": 26,
              "checks": [
                {
                  "name": "pattern_extracted",
                  "ok": true,
                  "detail": "^\\S+ - \\S+ \\S+ 5\\d{2} \\S+$"
                },
                {
                  "name": "compiles_and_runs",
                  "ok": true,
                  "detail": "None"
                },
                {
                  "name": "matched_lines",
                  "ok": true,
                  "detail": "got [2, 3, 6, 8, 10], want [2, 3, 6, 8, 10]"
                }
              ],
              "diagnostic": {
                "pattern": "^\\S+ - \\S+ \\S+ 5\\d{2} \\S+$",
                "matched": [
                  2,
                  3,
                  6,
                  8,
                  10
                ]
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sh-1",
          "category": "shell-regex",
          "proxy": false,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 102,
              "answer_chars": 51,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, 'NR>1 && $3==\"brass\"' inventory.csv | wc -l"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, 'NR>1 && $3==\"brass\"' inventory.csv | wc -l",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 102,
              "answer_chars": 51,
              "checks": [
                {
                  "name": "command_extracted",
                  "ok": true,
                  "detail": "awk -F, 'NR>1 && $3==\"brass\"' inventory.csv | wc -l"
                },
                {
                  "name": "executed",
                  "ok": true,
                  "detail": "refused: []"
                },
                {
                  "name": "exit_status",
                  "ok": true,
                  "detail": "rc=0 stderr=''"
                },
                {
                  "name": "stdout",
                  "ok": true,
                  "detail": "got '4', want '4'"
                }
              ],
              "diagnostic": {
                "command": "awk -F, 'NR>1 && $3==\"brass\"' inventory.csv | wc -l",
                "run": {
                  "executed": true,
                  "refused": [],
                  "stdout": "4\n",
                  "stderr": "",
                  "returncode": 0,
                  "note": ""
                }
              }
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-1",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "coverage: missing groups: [['double six', 'double-six']]",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 554,
              "answer_chars": 332,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "53 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['double six', 'double-six']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "coverage: missing groups: [['double six', 'double-six']]",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 554,
              "answer_chars": 332,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "53 words, cap 60"
                },
                {
                  "name": "coverage",
                  "ok": false,
                  "detail": "missing groups: [['double six', 'double-six']]"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        },
        {
          "item_id": "c2-sum-2",
          "category": "faithful-summary",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 622,
              "answer_chars": 388,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "68 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": true,
              "reason": "all checks passed",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 622,
              "answer_chars": 388,
              "checks": [
                {
                  "name": "word_cap",
                  "ok": true,
                  "detail": "68 words, cap 70"
                },
                {
                  "name": "coverage",
                  "ok": true,
                  "detail": "missing groups: none"
                },
                {
                  "name": "no_fabrication",
                  "ok": true,
                  "detail": "markers hit: none"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": true,
          "attempts": 2
        },
        {
          "item_id": "c2-tone-1",
          "category": "tone-rewrite",
          "proxy": true,
          "repeats": [
            {
              "repeat": 1,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 141,
              "answer_chars": 751,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "128 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'apolog', 'understand', 'thank you']"
                }
              ],
              "diagnostic": {}
            },
            {
              "repeat": 2,
              "pass": false,
              "reason": "no_banned_phrases: hit: ['do not reply']",
              "response_failure": false,
              "done_reason": "stop",
              "prompt_eval_count": 141,
              "answer_chars": 751,
              "checks": [
                {
                  "name": "word_band",
                  "ok": true,
                  "detail": "128 words, band 60-140"
                },
                {
                  "name": "facts_preserved",
                  "ok": true,
                  "detail": "missing: none"
                },
                {
                  "name": "no_banned_phrases",
                  "ok": false,
                  "detail": "hit: ['do not reply']"
                },
                {
                  "name": "no_banned_chars",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "no_invented_concession",
                  "ok": true,
                  "detail": "hit: none"
                },
                {
                  "name": "register_marker",
                  "ok": true,
                  "detail": "hit: ['sorry', 'apolog', 'understand', 'thank you']"
                }
              ],
              "diagnostic": {}
            }
          ],
          "conjunctive_pass": false,
          "attempts": 2
        }
      ]
    }
  ]
}
