{
  "schemaVersion": "kingy-independent-measurement-completion-eligibility-review/1.0.0",
  "checkedAt": "2026-10-17T21:34:17.115994+00:00",
  "readOnlyInterpretation": true,
  "gateLiteral": "all 240 original attempts complete and integrity sealed",
  "gateSource": "quality-gate-plan.json",
  "measurementCompletionGateSatisfied": false,
  "gateStatus": "UNMET_VERIFIED_COMPLETION",
  "launchEligibleFromThisStudy": false,
  "recommendedReleaseAction": "Hold candidate1.5.4 under the predeclared ifAnyFails rule. Preserve the existing active workflow and prior studies. A full honest retained-results report may still be published after accounting/seal, validity/privacy/public-readback work.",
  "basis": [
    "B3 original23 was submitted once and retained byte-exact as host-interrupted/incomplete, with no saved final-generation block, assistant completion timestamp or verified documented final-answer event.",
    "The capture-only procedure administratively records the incomplete attempt; it does not create a model completion or establish that a usable answer finished within90seconds.",
    "The predeclared gate requires completion plus sealed integrity. Retention/accounting and hash integrity are necessary but cannot substitute for missing completion evidence.",
    "The protocol explicitly requires preserving errors, incomplete generations and timeouts in planned denominators. That preservation requirement does not convert an unverified interrupted original into a verified completion or waive a separate launch gate.",
    "Even if the remaining originals are each executed once and all240 are retained/sealed, this preserved original23 prevents a claim of240 verified completed responses. Kingy numeric pass rates alone cannot override this distinct predeclared gate."
  ],
  "notARequirementForPerfectAccuracy": "This interpretation does not require every answer to be factually correct or supported; those outcomes have separate numeric thresholds. It requires that the stated measurement-completion condition be established. The host-interrupted original has no verified terminal answer outcome.",
  "sealerInterpretation": {
    "all240OriginalsRetained": "Original accounting/integrity evidence only.",
    "allPass": "Checks implemented by the sealer only; native manifest count/hash/order checks do not independently establish final-generation completion of all native originals.",
    "postHocReinterpretationForbidden": "Do not relabel all240Retained as all240Complete or silently weaken the launch criterion."
  },
  "knownInterruptedOriginal": {
    "system": "B",
    "repetition": 3,
    "scheduledPosition": 23,
    "questionId": "V19-Q09",
    "retainedSha256": "d98481153cec34f759ee58a1dc1a429f8238d08857a0c2811ae13e2ce90b6a6c",
    "assistantCompletionTimestampPropertyPresent": false,
    "finalGenerationBlockPresent": false,
    "finalAnswerCompletionVerified": false,
    "helperDeadlineExpired": true,
    "originalReplayed": false,
    "atMostPossibleVerifiedCompletionsAtStudyEnd": 239
  },
  "currentAccountingCountsOnly": {
    "A": {
      "1": 40,
      "2": 40,
      "3": 40
    },
    "B": {
      "1": 40,
      "2": 40,
      "3": 23
    }
  },
  "doesNotAssertOtherAttemptsVerifiedCompletions": true,
  "noAllowedRepairByReplay": [
    "Do not resubmit/continue/regenerate original23, splice replacement answers, edit the native record, or change scorer aliases/results.",
    "Later newly generated or discovered late output cannot retroactively establish a within90second first-attempt completion or replace the preserved original."
  ],
  "validityAudit": {
    "status": "PENDING",
    "started": false,
    "conditionToStart": "All240 original attempts accounted and integrity sealed, followed by explicit root signal to inspect fresh references/results.",
    "scoreValidityPassed": null,
    "scoreValidityAndMeasurementCompletionAreSeparate": true,
    "currentReviewDoesNotPredictAnyQualityScores": true
  },
  "claimBoundaries": [
    "Do not claim240 successful or verified-complete model responses from accounting alone.",
    "Report original23 as host-interrupted/incomplete/unverified and preserve its primary denominator treatment.",
    "Do not use this row to prove competitor task-quality or reliability disadvantage. Any exclusion sensitivity is supplemental and cannot replace primary frozen results."
  ],
  "ifAnyFailsLiteral": "Publish retained results and limitations; do not launch candidate; preserve previous studies.",
  "gateModified": false,
  "candidateOrScorerModified": false,
  "scoresOrOriginalResultsChanged": false,
  "questionReferenceAnswerOrGradeViewed": false,
  "nativeUiActions": 0,
  "modelCalls": 0,
  "inputSha256": {
    "quality-gate-plan.json": "106c5dd5aeb8224c8b713a62d7018d77932c0fb4430ca493c9032df9808def4b",
    "method/COMPARISON-PROTOCOL.md": "a0638194c552ef306b218d1c43a3c6adb3798507e02406c792496ef7072e3069",
    "seal-measurement.py": "7dc2fef344233624aa77fa3d374b0863bf6f1356b5fb0d2b659aa57decd945d1",
    "native-transport-amendment/ax-menu-r2/host-gap-capture-only-execution.json": "baff3cbecabae5ca557e870b215b740d53285dabeef2145734cb3f0bdd0720f3",
    "native-transport-amendment/ax-menu-r2/host-gap-boot-observation.json": "49fe85301e1601d4d0811b5ceacbb5dbbb4ff30534d9fe35c44f362ff35b8e1c",
    "native-transport-amendment/ax-menu-r2/host-gap-native-preflight-review.json": "2af2814fee61d1cbe6b68613dffa534bb04a8d581777b269b549cc4621d7faf2",
    "sessions/B3/native-23.conversation.json": "d98481153cec34f759ee58a1dc1a429f8238d08857a0c2811ae13e2ce90b6a6c"
  },
  "independentFinalAuditWritten": false
}