{
  "generatedAt": "2026-10-05T12:05:56.334099+00:00",
  "runComplete": true,
  "reviewComplete": true,
  "expectedPerModel": 279,
  "models": {
    "openai": {
      "state": "complete",
      "blocks": {
        "original-uat": {
          "cases": 95,
          "verdicts": {
            "fail": 19,
            "pass": 44,
            "partial": 29,
            "blocked": 3
          },
          "turns": 214,
          "firstTokenP50Ms": 3827,
          "firstTokenP90Ms": 5781,
          "chatFirstTokenP50Ms": 1431,
          "profileP50Ms": 2184,
          "completeP50Ms": 4277,
          "streamedTurns": 125,
          "singleChunkTurns": 89,
          "captureAttempts": 6,
          "acceptedLocal": 6,
          "responseStatuses": {
            "200": 6
          }
        },
        "uat-additions": {
          "cases": 24,
          "verdicts": {
            "fail": 10,
            "pass": 7,
            "partial": 6,
            "blocked": 1
          },
          "turns": 118,
          "firstTokenP50Ms": 3494,
          "firstTokenP90Ms": 5540,
          "chatFirstTokenP50Ms": 1228,
          "profileP50Ms": 2151,
          "completeP50Ms": 3661,
          "streamedTurns": 61,
          "singleChunkTurns": 57,
          "captureAttempts": 2,
          "acceptedLocal": 2,
          "responseStatuses": {
            "200": 2
          }
        },
        "knowledge": {
          "cases": 26,
          "verdicts": {
            "pass": 8,
            "fail": 12,
            "partial": 6
          },
          "turns": 60,
          "firstTokenP50Ms": 3066,
          "firstTokenP90Ms": 5981,
          "chatFirstTokenP50Ms": 1186,
          "profileP50Ms": 1916,
          "completeP50Ms": 3194,
          "streamedTurns": 42,
          "singleChunkTurns": 18,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "regression": {
          "cases": 38,
          "verdicts": {
            "fail": 13,
            "pass": 18,
            "partial": 7
          },
          "turns": 88,
          "firstTokenP50Ms": 3503,
          "firstTokenP90Ms": 5421,
          "chatFirstTokenP50Ms": 1288,
          "profileP50Ms": 1799,
          "completeP50Ms": 3687,
          "streamedTurns": 68,
          "singleChunkTurns": 20,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "feedback": {
          "cases": 14,
          "verdicts": {
            "pass": 7,
            "fail": 4,
            "partial": 3
          },
          "turns": 48,
          "firstTokenP50Ms": 2806,
          "firstTokenP90Ms": 5513,
          "chatFirstTokenP50Ms": 408,
          "profileP50Ms": 2055,
          "completeP50Ms": 2910,
          "streamedTurns": 8,
          "singleChunkTurns": 40,
          "captureAttempts": 6,
          "acceptedLocal": 5,
          "responseStatuses": {
            "200": 5,
            "503": 1
          }
        },
        "ticket-precision": {
          "cases": 12,
          "verdicts": {
            "pass": 5,
            "fail": 6,
            "partial": 1
          },
          "turns": 25,
          "firstTokenP50Ms": 2537,
          "firstTokenP90Ms": 4196,
          "chatFirstTokenP50Ms": 380,
          "profileP50Ms": 2078,
          "completeP50Ms": 2598,
          "streamedTurns": 5,
          "singleChunkTurns": 20,
          "captureAttempts": 1,
          "acceptedLocal": 1,
          "responseStatuses": {
            "200": 1
          }
        },
        "handoff-regression": {
          "cases": 4,
          "verdicts": {
            "fail": 1,
            "pass": 2,
            "blocked": 1
          },
          "turns": 17,
          "firstTokenP50Ms": 4116,
          "firstTokenP90Ms": 5619,
          "chatFirstTokenP50Ms": 2174,
          "profileP50Ms": 2176,
          "completeP50Ms": 4117,
          "streamedTurns": 6,
          "singleChunkTurns": 11,
          "captureAttempts": 2,
          "acceptedLocal": 2,
          "responseStatuses": {
            "200": 2
          }
        },
        "workflow-router": {
          "cases": 12,
          "verdicts": {
            "pass": 11,
            "fail": 1
          },
          "turns": 23,
          "firstTokenP50Ms": 2946,
          "firstTokenP90Ms": 4402,
          "chatFirstTokenP50Ms": 1055,
          "profileP50Ms": 2044,
          "completeP50Ms": 3095,
          "streamedTurns": 7,
          "singleChunkTurns": 16,
          "captureAttempts": 6,
          "acceptedLocal": 6,
          "responseStatuses": {
            "200": 6
          }
        },
        "flow-selector": {
          "cases": 23,
          "verdicts": {
            "pass": 19,
            "fail": 2,
            "blocked": 1,
            "partial": 1
          },
          "turns": 55,
          "firstTokenP50Ms": 3272,
          "firstTokenP90Ms": 5363,
          "chatFirstTokenP50Ms": 1173,
          "profileP50Ms": 2146,
          "completeP50Ms": 3316,
          "streamedTurns": 19,
          "singleChunkTurns": 36,
          "captureAttempts": 9,
          "acceptedLocal": 8,
          "responseStatuses": {
            "200": 8,
            "503": 1
          }
        },
        "response-policy": {
          "cases": 18,
          "verdicts": {
            "pass": 17,
            "partial": 1
          },
          "turns": 18,
          "firstTokenP50Ms": 2545,
          "firstTokenP90Ms": 5191,
          "chatFirstTokenP50Ms": 46,
          "profileP50Ms": 2288,
          "completeP50Ms": 2545,
          "streamedTurns": 6,
          "singleChunkTurns": 12,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "latest-e2e": {
          "cases": 13,
          "verdicts": {
            "pass": 13
          },
          "turns": 0,
          "firstTokenP50Ms": null,
          "firstTokenP90Ms": null,
          "chatFirstTokenP50Ms": null,
          "profileP50Ms": null,
          "completeP50Ms": null,
          "streamedTurns": 0,
          "singleChunkTurns": 0,
          "captureAttempts": 6,
          "acceptedLocal": 4,
          "responseStatuses": {
            "200": 4,
            "503": 2
          }
        }
      },
      "all": {
        "cases": 279,
        "verdicts": {
          "fail": 68,
          "pass": 151,
          "partial": 54,
          "blocked": 6
        },
        "turns": 666,
        "firstTokenP50Ms": 3360,
        "firstTokenP90Ms": 5567,
        "chatFirstTokenP50Ms": 1197,
        "profileP50Ms": 2101,
        "completeP50Ms": 3494,
        "streamedTurns": 347,
        "singleChunkTurns": 319,
        "captureAttempts": 38,
        "acceptedLocal": 34,
        "responseStatuses": {
          "200": 34,
          "503": 4
        }
      }
    },
    "claude": {
      "state": "complete",
      "blocks": {
        "original-uat": {
          "cases": 95,
          "verdicts": {
            "fail": 17,
            "partial": 25,
            "blocked": 5,
            "pass": 48
          },
          "turns": 224,
          "firstTokenP50Ms": 4122,
          "firstTokenP90Ms": 8464,
          "chatFirstTokenP50Ms": 1759,
          "profileP50Ms": 2158,
          "completeP50Ms": 5718,
          "streamedTurns": 125,
          "singleChunkTurns": 99,
          "captureAttempts": 4,
          "acceptedLocal": 4,
          "responseStatuses": {
            "200": 4
          }
        },
        "uat-additions": {
          "cases": 24,
          "verdicts": {
            "pass": 7,
            "partial": 5,
            "fail": 12
          },
          "turns": 118,
          "firstTokenP50Ms": 3802,
          "firstTokenP90Ms": 8243,
          "chatFirstTokenP50Ms": 1594,
          "profileP50Ms": 2153,
          "completeP50Ms": 5335,
          "streamedTurns": 63,
          "singleChunkTurns": 55,
          "captureAttempts": 4,
          "acceptedLocal": 4,
          "responseStatuses": {
            "200": 4
          }
        },
        "knowledge": {
          "cases": 26,
          "verdicts": {
            "pass": 14,
            "fail": 9,
            "partial": 3
          },
          "turns": 63,
          "firstTokenP50Ms": 3761,
          "firstTokenP90Ms": 8646,
          "chatFirstTokenP50Ms": 1745,
          "profileP50Ms": 1955,
          "completeP50Ms": 4496,
          "streamedTurns": 47,
          "singleChunkTurns": 16,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "regression": {
          "cases": 38,
          "verdicts": {
            "fail": 9,
            "pass": 27,
            "partial": 2
          },
          "turns": 92,
          "firstTokenP50Ms": 3952,
          "firstTokenP90Ms": 7431,
          "chatFirstTokenP50Ms": 1775,
          "profileP50Ms": 2075,
          "completeP50Ms": 4500,
          "streamedTurns": 75,
          "singleChunkTurns": 17,
          "captureAttempts": 1,
          "acceptedLocal": 1,
          "responseStatuses": {
            "200": 1
          }
        },
        "feedback": {
          "cases": 14,
          "verdicts": {
            "pass": 8,
            "fail": 5,
            "partial": 1
          },
          "turns": 48,
          "firstTokenP50Ms": 2834,
          "firstTokenP90Ms": 11034,
          "chatFirstTokenP50Ms": 398,
          "profileP50Ms": 2052,
          "completeP50Ms": 2835,
          "streamedTurns": 7,
          "singleChunkTurns": 41,
          "captureAttempts": 6,
          "acceptedLocal": 5,
          "responseStatuses": {
            "200": 5,
            "503": 1
          }
        },
        "ticket-precision": {
          "cases": 12,
          "verdicts": {
            "partial": 2,
            "fail": 6,
            "pass": 4
          },
          "turns": 25,
          "firstTokenP50Ms": 2904,
          "firstTokenP90Ms": 6634,
          "chatFirstTokenP50Ms": 416,
          "profileP50Ms": 2317,
          "completeP50Ms": 2904,
          "streamedTurns": 5,
          "singleChunkTurns": 20,
          "captureAttempts": 1,
          "acceptedLocal": 1,
          "responseStatuses": {
            "200": 1
          }
        },
        "handoff-regression": {
          "cases": 4,
          "verdicts": {
            "fail": 3,
            "blocked": 1
          },
          "turns": 17,
          "firstTokenP50Ms": 3089,
          "firstTokenP90Ms": 14063,
          "chatFirstTokenP50Ms": 546,
          "profileP50Ms": 2203,
          "completeP50Ms": 3090,
          "streamedTurns": 4,
          "singleChunkTurns": 13,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "workflow-router": {
          "cases": 12,
          "verdicts": {
            "pass": 10,
            "fail": 1,
            "partial": 1
          },
          "turns": 23,
          "firstTokenP50Ms": 3591,
          "firstTokenP90Ms": 6930,
          "chatFirstTokenP50Ms": 1592,
          "profileP50Ms": 2222,
          "completeP50Ms": 4759,
          "streamedTurns": 7,
          "singleChunkTurns": 16,
          "captureAttempts": 6,
          "acceptedLocal": 6,
          "responseStatuses": {
            "200": 6
          }
        },
        "flow-selector": {
          "cases": 23,
          "verdicts": {
            "pass": 20,
            "fail": 2,
            "partial": 1
          },
          "turns": 55,
          "firstTokenP50Ms": 4018,
          "firstTokenP90Ms": 9010,
          "chatFirstTokenP50Ms": 1707,
          "profileP50Ms": 2235,
          "completeP50Ms": 4718,
          "streamedTurns": 22,
          "singleChunkTurns": 33,
          "captureAttempts": 10,
          "acceptedLocal": 9,
          "responseStatuses": {
            "200": 9,
            "503": 1
          }
        },
        "response-policy": {
          "cases": 18,
          "verdicts": {
            "pass": 18
          },
          "turns": 18,
          "firstTokenP50Ms": 3014,
          "firstTokenP90Ms": 8248,
          "chatFirstTokenP50Ms": 47,
          "profileP50Ms": 2606,
          "completeP50Ms": 3014,
          "streamedTurns": 6,
          "singleChunkTurns": 12,
          "captureAttempts": 0,
          "acceptedLocal": 0,
          "responseStatuses": {}
        },
        "latest-e2e": {
          "cases": 13,
          "verdicts": {
            "pass": 12,
            "fail": 1
          },
          "turns": 0,
          "firstTokenP50Ms": null,
          "firstTokenP90Ms": null,
          "chatFirstTokenP50Ms": null,
          "profileP50Ms": null,
          "completeP50Ms": null,
          "streamedTurns": 0,
          "singleChunkTurns": 0,
          "captureAttempts": 7,
          "acceptedLocal": 4,
          "responseStatuses": {
            "200": 4,
            "503": 3
          }
        }
      },
      "all": {
        "cases": 279,
        "verdicts": {
          "fail": 65,
          "partial": 40,
          "blocked": 6,
          "pass": 168
        },
        "turns": 683,
        "firstTokenP50Ms": 3843,
        "firstTokenP90Ms": 8362,
        "chatFirstTokenP50Ms": 1623,
        "profileP50Ms": 2146,
        "completeP50Ms": 4628,
        "streamedTurns": 361,
        "singleChunkTurns": 322,
        "captureAttempts": 39,
        "acceptedLocal": 34,
        "responseStatuses": {
          "200": 34,
          "503": 5
        }
      }
    }
  },
  "pairedCases": 279,
  "verdictTransitions": [
    {
      "openai": "blocked",
      "claude": "blocked",
      "count": 3
    },
    {
      "openai": "blocked",
      "claude": "fail",
      "count": 1
    },
    {
      "openai": "blocked",
      "claude": "partial",
      "count": 1
    },
    {
      "openai": "blocked",
      "claude": "pass",
      "count": 1
    },
    {
      "openai": "fail",
      "claude": "fail",
      "count": 46
    },
    {
      "openai": "fail",
      "claude": "partial",
      "count": 8
    },
    {
      "openai": "fail",
      "claude": "pass",
      "count": 14
    },
    {
      "openai": "partial",
      "claude": "blocked",
      "count": 1
    },
    {
      "openai": "partial",
      "claude": "fail",
      "count": 6
    },
    {
      "openai": "partial",
      "claude": "partial",
      "count": 22
    },
    {
      "openai": "partial",
      "claude": "pass",
      "count": 25
    },
    {
      "openai": "pass",
      "claude": "blocked",
      "count": 2
    },
    {
      "openai": "pass",
      "claude": "fail",
      "count": 12
    },
    {
      "openai": "pass",
      "claude": "partial",
      "count": 9
    },
    {
      "openai": "pass",
      "claude": "pass",
      "count": 128
    }
  ]
}
