Agent Maturity Compass API - v1.2.0
    Preparing search index...

    Interface LiveScoreDrift

    interface LiveScoreDrift {
        actionFixationRateIncrease0to1: number;
        adkDeploymentReadinessDrop0to1: number;
        adkEvalPassRateDrop0to1: number;
        adkEvidenceCoverageDrop0to1: number;
        adkGraphCoverageDrop0to1: number;
        adkStreamingStabilityDrop0to1: number;
        adkToolCallSuccessRateDrop0to1: number;
        agentEvalHarnessCostIncreaseRatio: number;
        agentEvalHarnessEvidenceCoverageDrop0to1: number;
        agentEvalHarnessHallucinationIncrease0to1: number;
        agentEvalHarnessLatencyP95IncreaseRatio: number;
        agentEvalHarnessToolSuccessDrop0to1: number;
        agentEvalHarnessTraceCoverageDrop0to1: number;
        agentEvalObservabilityConfigCoverageDrop0to1: number;
        agentEvalObservabilityEvidenceCoverageDrop0to1: number;
        agentEvalObservabilityTelemetryCoverageDrop0to1: number;
        agenticSearchCitationCoverageDrop0to1: number;
        agenticSearchPlanningScoreDrop0to1: number;
        agenticSearchQueryDecompositionDrop0to1: number;
        agenticSearchRelevanceDrop0to1: number;
        agenticSearchSynthesisDrop0to1: number;
        agenticSearchTraceCoverageDrop0to1: number;
        agentSecurityAttackEffectivenessIncrease0to1: number;
        agentSecurityAuditTrailIntegrityDrop0to1: number;
        agentSecurityEvidenceCoverageDrop0to1: number;
        agentSecurityFalsePositiveRateIncrease0to1: number;
        agentSecurityLatencyP95IncreaseRatio: number;
        agentSecurityPolicyDecisionAccuracyDrop0to1: number;
        agentSecuritySecretScrubRateDrop0to1: number;
        agentSecuritySourceOriginCoverageDrop0to1: number;
        agentSecurityTaintPropagationCoverageDrop0to1: number;
        agentTestingEvidenceCoverageDrop0to1: number;
        agentTestingFaultInjectionCoverageDrop0to1: number;
        agentTestingMethodologyCoverageDrop0to1: number;
        agentTestingObservabilitySignalCoverageDrop0to1: number;
        agentTestingResiliencePassRateDrop0to1: number;
        agentTestingSafetyRegressionRateIncrease0to1: number;
        agentTestingScenarioCoverageDrop0to1: number;
        artifactAccuracyDrop0to1: number;
        authorityHandoffRateDrop0to1: number;
        backdoorAgentAttackSuccessRateIncrease0to1: number;
        backdoorAgentCleanAccuracyDrop0to1: number;
        backdoorAgentEvidenceCoverageDrop0to1: number;
        backdoorAgentTrajectoryCoverageDrop0to1: number;
        backdoorAgentTriggerPersistenceIncrease0to1: number;
        backdoorAgentTriggerPropagationIncrease0to1: number;
        chaosDropIncrease0to1: number;
        chaosEvidenceCoverageDrop0to1: number;
        chaosFailureTraceCoverageDrop0to1: number;
        chaosImprovementEvalCoverageDrop0to1: number;
        chaosProductionReliabilityDrop0to1: number;
        chaosRecoveryPassRateDrop0to1: number;
        chaosResilienceScoreDrop0to1: number;
        controlCorrectabilityDrop0to1: number;
        controlInterpretabilityDrop0to1: number;
        controlInterruptibilityDrop0to1: number;
        controlReversibilityDrop0to1: number;
        costIncreaseRatio: number;
        cpuAgenticCpuUtilizationIncrease0to1: number;
        cpuAgenticEvidenceCoverageDrop0to1: number;
        cpuAgenticFrameworkOverheadShareIncrease0to1: number;
        cpuAgenticGpuUtilizationDrop0to1: number;
        cpuAgenticLatencyP50IncreaseRatio: number;
        cpuAgenticLatencyP95IncreaseRatio: number;
        cpuAgenticLatencyP99IncreaseRatio: number;
        cpuAgenticLlmInferenceShareShift0to1: number;
        cpuAgenticMemoryIncreaseRatio: number;
        cpuAgenticThroughputDropRatio: number;
        cpuAgenticToolExecutionShareIncrease0to1: number;
        ctfCheckpointCompletionDrop0to1: number;
        ctfCompetitionImpactIncrease0to1: number;
        ctfContaminationRiskIncrease0to1: number;
        ctfExternalSearchUseRateIncrease0to1: number;
        ctfFirstCorrectFlagForwardingRateDrop0to1: number;
        ctfFlagSolveRateDrop0to1: number;
        ctfIndependenceViolationRate0to1: number;
        ctfIsolationViolationRate0to1: number;
        ctfPartialCreditScoreDrop0to1: number;
        ctfTraceCoverageRateDrop0to1: number;
        divergenceMomentumDrop0to1: number;
        documentDatasetEvidenceCoverageDrop0to1: number;
        documentDatasetMemoryIncreaseRatio: number;
        documentDatasetNumericMismatchRateIncrease0to1: number;
        documentDatasetNumGuardCoverageDrop0to1: number;
        documentDatasetQaAccuracyDrop0to1: number;
        documentDatasetRagFaithfulnessDrop0to1: number;
        documentDatasetSummaryQualityDrop0to1: number;
        documentDatasetThroughputDropRatio: number;
        documentDatasetTokenSavingsDropRatio: number;
        driftStatistic: number;
        errorAttributionRateIncrease0to1: number;
        errorRateIncrease0to1: number;
        evalTechniqueAlgorithmicFeedbackCoverageDrop0to1: number;
        evalTechniqueDynamicGroundTruthPassRateDrop0to1: number;
        evalTechniqueEvidenceCoverageDrop0to1: number;
        evalTechniqueExactMatchAccuracyDrop0to1: number;
        evalTechniqueLlmJudgeAgreementDrop0to1: number;
        evalTechniquePairwiseWinRateDrop0to1: number;
        evalTechniqueRagContextRelevanceDrop0to1: number;
        evalTechniqueRagFaithfulnessDrop0to1: number;
        evalTechniqueRealtimeFeedbackDrop0to1: number;
        evalTechniqueSimulationGoalCompletionDrop0to1: number;
        evalTechniqueStructuredValidationDrop0to1: number;
        evalTechniqueToolImprovementDrop0to1: number;
        evalTechniqueToolPrecisionDrop0to1: number;
        evalTechniqueTrajectoryMatchRateDrop0to1: number;
        formatQualityDrop0to1: number;
        formulaIntegrityDrop0to1: number;
        genomicsExpertCurationCoverageDrop0to1: number;
        genomicsFormatConformanceRateDrop0to1: number;
        genomicsPreprocessingQualityDrop0to1: number;
        genomicsReferenceCoverageDrop0to1: number;
        genomicsSelectionAccuracyDrop0to1: number;
        genomicsStatisticalAnalysisAccuracyDrop0to1: number;
        hedraRagEvidenceCoverageDrop0to1: number;
        hedraRagLatencyP95IncreaseRatio: number;
        hedraRagMemoryIncreaseRatio: number;
        hedraRagReplayPassRateDrop0to1: number;
        hedraRagThroughputDropRatio: number;
        interactionTurnMeanShiftRatio: number;
        invalidActionRateIncrease0to1: number;
        kiteDocumentCountDropRatio: number;
        kiteEvidenceCoverageDrop0to1: number;
        kiteGradeDrop0to10: number;
        kiteNormalizedGradeDrop0to1: number;
        kiteQuestionCountDropRatio: number;
        latencyIncreaseRatio: number;
        legalAgentCitationCoverageDrop0to1: number;
        legalAgentEvidenceCoverageDrop0to1: number;
        legalAgentFinalSuccessDrop0to1: number;
        legalAgentProcessRateDrop0to1: number;
        legalAgentTokenCostIncreaseRatio: number;
        legalAgentToolUseAccuracyDrop0to1: number;
        llmRagBiasRiskIncrease0to1: number;
        llmRagEvalSuiteEvidenceCoverageDrop0to1: number;
        llmRagHallucinationRateIncrease0to1: number;
        llmRagSemanticSimilarityDrop0to1: number;
        localSystemEvidenceCoverageDrop0to1: number;
        localSystemGhostDriverDetectionCoverageDrop0to1: number;
        localSystemLocalOnlyPrivacyCoverageDrop0to1: number;
        localSystemProactiveAlertCoverageDrop0to1: number;
        localSystemProcessIdentityCoverageDrop0to1: number;
        localSystemThermalBaselineDeviationIncrease0to1: number;
        localSystemVoltageSpcAnomalyRateIncrease0to1: number;
        naviBenchCrashRateIncrease0to1: number;
        naviBenchEvidenceCoverageDrop0to1: number;
        naviBenchExcludingCrashedScoreDrop0to1: number;
        naviBenchLowerBoundScoreDrop0to1: number;
        naviBenchStepCountIncreaseRatio: number;
        naviBenchStepLimitViolationRateIncrease0to1: number;
        naviBenchTaskSuccessDrop0to1: number;
        naviBenchTrajectoryCoverageDrop0to1: number;
        naviBenchVisualizationCoverageDrop0to1: number;
        noMiraclAbstentionAccuracyDrop0to1: number;
        noMiraclErrorRateIncrease0to1: number;
        noMiraclEvidenceCoverageDrop0to1: number;
        noMiraclHallucinationRateIncrease0to1: number;
        noMiraclLanguageCoverageDrop0to1: number;
        noMiraclRelevanceAccuracyDrop0to1: number;
        noMiraclSubsetCoverageDrop0to1: number;
        observabilityDeterministicCheckPassRateDrop0to1: number;
        observabilityEvidenceCoverageDrop0to1: number;
        observabilityReportCoverageDrop0to1: number;
        observabilityResolutionScoreDrop0to1: number;
        observabilityRubricScoreDrop0to1: number;
        observabilityScenarioClockAlignmentRateDrop0to1: number;
        observabilityTraceCoverageDrop0to1: number;
        offPathAttemptMeanDropRatio: number;
        ollamaMetricsEvidenceCoverageDrop0to1: number;
        ollamaMetricsLoadedModelCountDropRatio: number;
        ollamaMetricsModelLoadedRateDrop0to1: number;
        ollamaMetricsModelRamIncreaseRatio: number;
        ollamaMetricsRequestDurationP95IncreaseRatio: number;
        ollamaMetricsRequestErrorRateIncrease0to1: number;
        ollamaMetricsTimePerTokenIncreaseRatio: number;
        osUniverseAutoValidationPassDrop0to1: number;
        osUniverseEvidenceCoverageDrop0to1: number;
        osUniverseStepCountIncreaseRatio: number;
        osUniverseStepLimitViolationRateIncrease0to1: number;
        osUniverseTaskSuccessDrop0to1: number;
        osUniverseValidationErrorRateIncrease0to1: number;
        passRateDrop0to1: number;
        personaBehaviorCoverageDrop0to1: number;
        personaHumanLikenessDrop0to1: number;
        personaTaskGoalPreservationDrop0to1: number;
        physicianBenchArtifactCoverageDrop0to1: number;
        physicianBenchCheckpointPassRateDrop0to1: number;
        physicianBenchClinicalActionSafetyDrop0to1: number;
        physicianBenchDocumentationQualityDrop0to1: number;
        physicianBenchEvidenceCoverageDrop0to1: number;
        physicianBenchFhirDataAccessAccuracyDrop0to1: number;
        physicianBenchTaskSuccessRateDrop0to1: number;
        physicianBenchTrajectoryCoverageDrop0to1: number;
        piArenaAgentTaskSuccessRateDrop0to1: number;
        piArenaAttackSuccessRateIncrease0to1: number;
        piArenaDefenseBlockRateDrop0to1: number;
        piArenaEvidenceCoverageDrop0to1: number;
        piArenaFalsePositiveRateIncrease0to1: number;
        piArenaToolCallSuccessRateDrop0to1: number;
        pokerEvalAllInAdjBbPer100Drop: number;
        pokerEvalBbPer100Drop: number;
        pokerEvalEvBbPer100Drop: number;
        pokerEvalEvidenceCoverageDrop0to1: number;
        pokerEvalHandCountDropRatio: number;
        pokerEvalVpipShift0to1: number;
        privacyPeerExposureRateIncrease0to1: number;
        privacySafeguardActiveRateDrop0to1: number;
        privacySensitiveDisclosureRateIncrease0to1: number;
        privacySocialPressureIncrease0to1: number;
        privacyWebDataMinimizationPassRateDrop0to1: number;
        privacyWebEvidenceCoverageDrop0to1: number;
        privacyWebLeakageRateIncrease0to1: number;
        privacyWebModalLeakageDeltaIncrease0to1: number;
        privacyWebSensitiveFieldExposureIncreaseRatio: number;
        privacyWebTaskSuccessRateDrop0to1: number;
        privacyWebUnnecessaryDisclosureRateIncrease0to1: number;
        processDefectRateIncrease0to1: number;
        ragAccuracyDrop0to1: number;
        ragAnswerSupportCoverageDrop0to1: number;
        ragCitationCoverageDrop0to1: number;
        ragCompletenessDrop0to1: number;
        ragDatasetBuilderEvidenceCoverageDrop0to1: number;
        ragGeneratedDataFinalCoverageDrop0to1: number;
        ragGenerationCostIncreaseRatio: number;
        ragHallucinationRateIncrease0to1: number;
        ragHumanVerificationCoverageDrop0to1: number;
        ragNumericalAccuracyDrop0to1: number;
        ragPassageGroundingCoverageDrop0to1: number;
        ragQuestionCountDropRatio: number;
        ragRetrievalTopKMeanShiftRatio: number;
        ragSourceDocumentCountDropRatio: number;
        ragStrategyEvidenceCoverageDrop0to1: number;
        ragUtilizationDrop0to1: number;
        recoveryBenchContextCoverageDrop0to1: number;
        recoveryBenchCorruptedEnvironmentCoverageDrop0to1: number;
        recoveryBenchEvidenceCoverageDrop0to1: number;
        recoveryBenchFailureTraceCoverageDrop0to1: number;
        recoveryBenchRecoveryRewardDrop0to1: number;
        recoveryBenchRecoverySuccessRateDrop0to1: number;
        recoveryBenchReplayIntegrityRateDrop0to1: number;
        redTeamAttackCoverageDrop0to1: number;
        redTeamComplianceDrop0to1: number;
        redTeamDatasetCoverageDrop0to1: number;
        redTeamGuardCoverageDrop0to1: number;
        redTeamGuardScoreDrop0to1: number;
        redTeamTaxonomyCoverageDrop0to1: number;
        redTeamUnsafeResponseRateIncrease0to1: number;
        refusalRateIncrease0to1: number;
        researchGymArtifactCoverageDrop0to1: number;
        researchGymBudgetOverrunRateIncrease0to1: number;
        researchGymInspectionPassRateDrop0to1: number;
        researchGymScoreImprovementDrop0to1: number;
        researchGymSubtaskCompletionDrop0to1: number;
        researchGymViolationRateIncrease0to1: number;
        sapAgentEvalEnterpriseContextCoverageDrop0to1: number;
        sapAgentEvalEvidenceCoverageDrop0to1: number;
        sapAgentEvalObjectiveCoverageDrop0to1: number;
        sapAgentEvalProcessCoverageDrop0to1: number;
        scalingLawEvidenceCoverageDrop0to1: number;
        scalingLawNmaeIncrease: number;
        scalingLawNmseIncrease: number;
        scalingLawR2Drop: number;
        scoreDrop0to1: number;
        socialHarmPrevalenceIncrease0to1: number;
        socialLexicalDiversityDrop0to1: number;
        socialSemanticAlignmentDrop0to1: number;
        socialSentimentMeanShift: number;
        solutionPathMeanDropRatio: number;
        strandsBenchmarkHarnessCostIncreaseRatio: number;
        strandsBenchmarkHarnessEvidenceCoverageDrop0to1: number;
        strandsBenchmarkHarnessLatencyP95IncreaseRatio: number;
        strandsBenchmarkHarnessPatchApplyRateDrop0to1: number;
        strandsBenchmarkHarnessTaskSuccessDrop0to1: number;
        strandsBenchmarkHarnessTestPassRateDrop0to1: number;
        strandsBenchmarkHarnessTrajectoryCoverageDrop0to1: number;
        toolAnswerVerificationDrop0to1: number;
        toolCallMeanShiftRatio: number;
        toolCallValidityDrop0to1: number;
        toolEvalImprovementDrop0to1: number;
        toolJudgeAgreementDrop0to1: number;
        toolRolloutDiversityDrop0to1: number;
        toolUseRewardDrop0to1: number;
        tradingClaimValidationFailureIncrease0to1: number;
        tradingDrawdownIncrease0to1: number;
        tradingMemoryRetrievalHitRateDrop0to1: number;
        tradingPnlDropPct: number;
        tradingProviderFallbackRateIncrease0to1: number;
        tradingRiskLimitViolationIncrease0to1: number;
        tradingRiskRewardDropRatio: number;
        tradingVisionChartAgreementDrop0to1: number;
        tradingWinRateDrop0to1: number;
        webOperatorLlmEvaluationDrop0to1: number;
        webOperatorMismatchRateIncrease0to1: number;
        webOperatorReplayCoverageDrop0to1: number;
        webOperatorSelfReportOverclaimIncrease0to1: number;
        webOperatorStepLimitViolationRateIncrease0to1: number;
        webOperatorTaskReliabilityDrop0to1: number;
        webOperatorTaskTimeIncreaseRatio: number;
    }
    Index
    actionFixationRateIncrease0to1 adkDeploymentReadinessDrop0to1 adkEvalPassRateDrop0to1 adkEvidenceCoverageDrop0to1 adkGraphCoverageDrop0to1 adkStreamingStabilityDrop0to1 adkToolCallSuccessRateDrop0to1 agentEvalHarnessCostIncreaseRatio agentEvalHarnessEvidenceCoverageDrop0to1 agentEvalHarnessHallucinationIncrease0to1 agentEvalHarnessLatencyP95IncreaseRatio agentEvalHarnessToolSuccessDrop0to1 agentEvalHarnessTraceCoverageDrop0to1 agentEvalObservabilityConfigCoverageDrop0to1 agentEvalObservabilityEvidenceCoverageDrop0to1 agentEvalObservabilityTelemetryCoverageDrop0to1 agenticSearchCitationCoverageDrop0to1 agenticSearchPlanningScoreDrop0to1 agenticSearchQueryDecompositionDrop0to1 agenticSearchRelevanceDrop0to1 agenticSearchSynthesisDrop0to1 agenticSearchTraceCoverageDrop0to1 agentSecurityAttackEffectivenessIncrease0to1 agentSecurityAuditTrailIntegrityDrop0to1 agentSecurityEvidenceCoverageDrop0to1 agentSecurityFalsePositiveRateIncrease0to1 agentSecurityLatencyP95IncreaseRatio agentSecurityPolicyDecisionAccuracyDrop0to1 agentSecuritySecretScrubRateDrop0to1 agentSecuritySourceOriginCoverageDrop0to1 agentSecurityTaintPropagationCoverageDrop0to1 agentTestingEvidenceCoverageDrop0to1 agentTestingFaultInjectionCoverageDrop0to1 agentTestingMethodologyCoverageDrop0to1 agentTestingObservabilitySignalCoverageDrop0to1 agentTestingResiliencePassRateDrop0to1 agentTestingSafetyRegressionRateIncrease0to1 agentTestingScenarioCoverageDrop0to1 artifactAccuracyDrop0to1 authorityHandoffRateDrop0to1 backdoorAgentAttackSuccessRateIncrease0to1 backdoorAgentCleanAccuracyDrop0to1 backdoorAgentEvidenceCoverageDrop0to1 backdoorAgentTrajectoryCoverageDrop0to1 backdoorAgentTriggerPersistenceIncrease0to1 backdoorAgentTriggerPropagationIncrease0to1 chaosDropIncrease0to1 chaosEvidenceCoverageDrop0to1 chaosFailureTraceCoverageDrop0to1 chaosImprovementEvalCoverageDrop0to1 chaosProductionReliabilityDrop0to1 chaosRecoveryPassRateDrop0to1 chaosResilienceScoreDrop0to1 controlCorrectabilityDrop0to1 controlInterpretabilityDrop0to1 controlInterruptibilityDrop0to1 controlReversibilityDrop0to1 costIncreaseRatio cpuAgenticCpuUtilizationIncrease0to1 cpuAgenticEvidenceCoverageDrop0to1 cpuAgenticFrameworkOverheadShareIncrease0to1 cpuAgenticGpuUtilizationDrop0to1 cpuAgenticLatencyP50IncreaseRatio cpuAgenticLatencyP95IncreaseRatio cpuAgenticLatencyP99IncreaseRatio cpuAgenticLlmInferenceShareShift0to1 cpuAgenticMemoryIncreaseRatio cpuAgenticThroughputDropRatio cpuAgenticToolExecutionShareIncrease0to1 ctfCheckpointCompletionDrop0to1 ctfCompetitionImpactIncrease0to1 ctfContaminationRiskIncrease0to1 ctfExternalSearchUseRateIncrease0to1 ctfFirstCorrectFlagForwardingRateDrop0to1 ctfFlagSolveRateDrop0to1 ctfIndependenceViolationRate0to1 ctfIsolationViolationRate0to1 ctfPartialCreditScoreDrop0to1 ctfTraceCoverageRateDrop0to1 divergenceMomentumDrop0to1 documentDatasetEvidenceCoverageDrop0to1 documentDatasetMemoryIncreaseRatio documentDatasetNumericMismatchRateIncrease0to1 documentDatasetNumGuardCoverageDrop0to1 documentDatasetQaAccuracyDrop0to1 documentDatasetRagFaithfulnessDrop0to1 documentDatasetSummaryQualityDrop0to1 documentDatasetThroughputDropRatio documentDatasetTokenSavingsDropRatio driftStatistic errorAttributionRateIncrease0to1 errorRateIncrease0to1 evalTechniqueAlgorithmicFeedbackCoverageDrop0to1 evalTechniqueDynamicGroundTruthPassRateDrop0to1 evalTechniqueEvidenceCoverageDrop0to1 evalTechniqueExactMatchAccuracyDrop0to1 evalTechniqueLlmJudgeAgreementDrop0to1 evalTechniquePairwiseWinRateDrop0to1 evalTechniqueRagContextRelevanceDrop0to1 evalTechniqueRagFaithfulnessDrop0to1 evalTechniqueRealtimeFeedbackDrop0to1 evalTechniqueSimulationGoalCompletionDrop0to1 evalTechniqueStructuredValidationDrop0to1 evalTechniqueToolImprovementDrop0to1 evalTechniqueToolPrecisionDrop0to1 evalTechniqueTrajectoryMatchRateDrop0to1 formatQualityDrop0to1 formulaIntegrityDrop0to1 genomicsExpertCurationCoverageDrop0to1 genomicsFormatConformanceRateDrop0to1 genomicsPreprocessingQualityDrop0to1 genomicsReferenceCoverageDrop0to1 genomicsSelectionAccuracyDrop0to1 genomicsStatisticalAnalysisAccuracyDrop0to1 hedraRagEvidenceCoverageDrop0to1 hedraRagLatencyP95IncreaseRatio hedraRagMemoryIncreaseRatio hedraRagReplayPassRateDrop0to1 hedraRagThroughputDropRatio interactionTurnMeanShiftRatio invalidActionRateIncrease0to1 kiteDocumentCountDropRatio kiteEvidenceCoverageDrop0to1 kiteGradeDrop0to10 kiteNormalizedGradeDrop0to1 kiteQuestionCountDropRatio latencyIncreaseRatio legalAgentCitationCoverageDrop0to1 legalAgentEvidenceCoverageDrop0to1 legalAgentFinalSuccessDrop0to1 legalAgentProcessRateDrop0to1 legalAgentTokenCostIncreaseRatio legalAgentToolUseAccuracyDrop0to1 llmRagBiasRiskIncrease0to1 llmRagEvalSuiteEvidenceCoverageDrop0to1 llmRagHallucinationRateIncrease0to1 llmRagSemanticSimilarityDrop0to1 localSystemEvidenceCoverageDrop0to1 localSystemGhostDriverDetectionCoverageDrop0to1 localSystemLocalOnlyPrivacyCoverageDrop0to1 localSystemProactiveAlertCoverageDrop0to1 localSystemProcessIdentityCoverageDrop0to1 localSystemThermalBaselineDeviationIncrease0to1 localSystemVoltageSpcAnomalyRateIncrease0to1 naviBenchCrashRateIncrease0to1 naviBenchEvidenceCoverageDrop0to1 naviBenchExcludingCrashedScoreDrop0to1 naviBenchLowerBoundScoreDrop0to1 naviBenchStepCountIncreaseRatio naviBenchStepLimitViolationRateIncrease0to1 naviBenchTaskSuccessDrop0to1 naviBenchTrajectoryCoverageDrop0to1 naviBenchVisualizationCoverageDrop0to1 noMiraclAbstentionAccuracyDrop0to1 noMiraclErrorRateIncrease0to1 noMiraclEvidenceCoverageDrop0to1 noMiraclHallucinationRateIncrease0to1 noMiraclLanguageCoverageDrop0to1 noMiraclRelevanceAccuracyDrop0to1 noMiraclSubsetCoverageDrop0to1 observabilityDeterministicCheckPassRateDrop0to1 observabilityEvidenceCoverageDrop0to1 observabilityReportCoverageDrop0to1 observabilityResolutionScoreDrop0to1 observabilityRubricScoreDrop0to1 observabilityScenarioClockAlignmentRateDrop0to1 observabilityTraceCoverageDrop0to1 offPathAttemptMeanDropRatio ollamaMetricsEvidenceCoverageDrop0to1 ollamaMetricsLoadedModelCountDropRatio ollamaMetricsModelLoadedRateDrop0to1 ollamaMetricsModelRamIncreaseRatio ollamaMetricsRequestDurationP95IncreaseRatio ollamaMetricsRequestErrorRateIncrease0to1 ollamaMetricsTimePerTokenIncreaseRatio osUniverseAutoValidationPassDrop0to1 osUniverseEvidenceCoverageDrop0to1 osUniverseStepCountIncreaseRatio osUniverseStepLimitViolationRateIncrease0to1 osUniverseTaskSuccessDrop0to1 osUniverseValidationErrorRateIncrease0to1 passRateDrop0to1 personaBehaviorCoverageDrop0to1 personaHumanLikenessDrop0to1 personaTaskGoalPreservationDrop0to1 physicianBenchArtifactCoverageDrop0to1 physicianBenchCheckpointPassRateDrop0to1 physicianBenchClinicalActionSafetyDrop0to1 physicianBenchDocumentationQualityDrop0to1 physicianBenchEvidenceCoverageDrop0to1 physicianBenchFhirDataAccessAccuracyDrop0to1 physicianBenchTaskSuccessRateDrop0to1 physicianBenchTrajectoryCoverageDrop0to1 piArenaAgentTaskSuccessRateDrop0to1 piArenaAttackSuccessRateIncrease0to1 piArenaDefenseBlockRateDrop0to1 piArenaEvidenceCoverageDrop0to1 piArenaFalsePositiveRateIncrease0to1 piArenaToolCallSuccessRateDrop0to1 pokerEvalAllInAdjBbPer100Drop pokerEvalBbPer100Drop pokerEvalEvBbPer100Drop pokerEvalEvidenceCoverageDrop0to1 pokerEvalHandCountDropRatio pokerEvalVpipShift0to1 privacyPeerExposureRateIncrease0to1 privacySafeguardActiveRateDrop0to1 privacySensitiveDisclosureRateIncrease0to1 privacySocialPressureIncrease0to1 privacyWebDataMinimizationPassRateDrop0to1 privacyWebEvidenceCoverageDrop0to1 privacyWebLeakageRateIncrease0to1 privacyWebModalLeakageDeltaIncrease0to1 privacyWebSensitiveFieldExposureIncreaseRatio privacyWebTaskSuccessRateDrop0to1 privacyWebUnnecessaryDisclosureRateIncrease0to1 processDefectRateIncrease0to1 ragAccuracyDrop0to1 ragAnswerSupportCoverageDrop0to1 ragCitationCoverageDrop0to1 ragCompletenessDrop0to1 ragDatasetBuilderEvidenceCoverageDrop0to1 ragGeneratedDataFinalCoverageDrop0to1 ragGenerationCostIncreaseRatio ragHallucinationRateIncrease0to1 ragHumanVerificationCoverageDrop0to1 ragNumericalAccuracyDrop0to1 ragPassageGroundingCoverageDrop0to1 ragQuestionCountDropRatio ragRetrievalTopKMeanShiftRatio ragSourceDocumentCountDropRatio ragStrategyEvidenceCoverageDrop0to1 ragUtilizationDrop0to1 recoveryBenchContextCoverageDrop0to1 recoveryBenchCorruptedEnvironmentCoverageDrop0to1 recoveryBenchEvidenceCoverageDrop0to1 recoveryBenchFailureTraceCoverageDrop0to1 recoveryBenchRecoveryRewardDrop0to1 recoveryBenchRecoverySuccessRateDrop0to1 recoveryBenchReplayIntegrityRateDrop0to1 redTeamAttackCoverageDrop0to1 redTeamComplianceDrop0to1 redTeamDatasetCoverageDrop0to1 redTeamGuardCoverageDrop0to1 redTeamGuardScoreDrop0to1 redTeamTaxonomyCoverageDrop0to1 redTeamUnsafeResponseRateIncrease0to1 refusalRateIncrease0to1 researchGymArtifactCoverageDrop0to1 researchGymBudgetOverrunRateIncrease0to1 researchGymInspectionPassRateDrop0to1 researchGymScoreImprovementDrop0to1 researchGymSubtaskCompletionDrop0to1 researchGymViolationRateIncrease0to1 sapAgentEvalEnterpriseContextCoverageDrop0to1 sapAgentEvalEvidenceCoverageDrop0to1 sapAgentEvalObjectiveCoverageDrop0to1 sapAgentEvalProcessCoverageDrop0to1 scalingLawEvidenceCoverageDrop0to1 scalingLawNmaeIncrease scalingLawNmseIncrease scalingLawR2Drop scoreDrop0to1 socialHarmPrevalenceIncrease0to1 socialLexicalDiversityDrop0to1 socialSemanticAlignmentDrop0to1 socialSentimentMeanShift solutionPathMeanDropRatio strandsBenchmarkHarnessCostIncreaseRatio strandsBenchmarkHarnessEvidenceCoverageDrop0to1 strandsBenchmarkHarnessLatencyP95IncreaseRatio strandsBenchmarkHarnessPatchApplyRateDrop0to1 strandsBenchmarkHarnessTaskSuccessDrop0to1 strandsBenchmarkHarnessTestPassRateDrop0to1 strandsBenchmarkHarnessTrajectoryCoverageDrop0to1 toolAnswerVerificationDrop0to1 toolCallMeanShiftRatio toolCallValidityDrop0to1 toolEvalImprovementDrop0to1 toolJudgeAgreementDrop0to1 toolRolloutDiversityDrop0to1 toolUseRewardDrop0to1 tradingClaimValidationFailureIncrease0to1 tradingDrawdownIncrease0to1 tradingMemoryRetrievalHitRateDrop0to1 tradingPnlDropPct tradingProviderFallbackRateIncrease0to1 tradingRiskLimitViolationIncrease0to1 tradingRiskRewardDropRatio tradingVisionChartAgreementDrop0to1 tradingWinRateDrop0to1 webOperatorLlmEvaluationDrop0to1 webOperatorMismatchRateIncrease0to1 webOperatorReplayCoverageDrop0to1 webOperatorSelfReportOverclaimIncrease0to1 webOperatorStepLimitViolationRateIncrease0to1 webOperatorTaskReliabilityDrop0to1 webOperatorTaskTimeIncreaseRatio
    actionFixationRateIncrease0to1: number
    adkDeploymentReadinessDrop0to1: number
    adkEvalPassRateDrop0to1: number
    adkEvidenceCoverageDrop0to1: number
    adkGraphCoverageDrop0to1: number
    adkStreamingStabilityDrop0to1: number
    adkToolCallSuccessRateDrop0to1: number
    agentEvalHarnessCostIncreaseRatio: number
    agentEvalHarnessEvidenceCoverageDrop0to1: number
    agentEvalHarnessHallucinationIncrease0to1: number
    agentEvalHarnessLatencyP95IncreaseRatio: number
    agentEvalHarnessToolSuccessDrop0to1: number
    agentEvalHarnessTraceCoverageDrop0to1: number
    agentEvalObservabilityConfigCoverageDrop0to1: number
    agentEvalObservabilityEvidenceCoverageDrop0to1: number
    agentEvalObservabilityTelemetryCoverageDrop0to1: number
    agenticSearchCitationCoverageDrop0to1: number
    agenticSearchPlanningScoreDrop0to1: number
    agenticSearchQueryDecompositionDrop0to1: number
    agenticSearchRelevanceDrop0to1: number
    agenticSearchSynthesisDrop0to1: number
    agenticSearchTraceCoverageDrop0to1: number
    agentSecurityAttackEffectivenessIncrease0to1: number
    agentSecurityAuditTrailIntegrityDrop0to1: number
    agentSecurityEvidenceCoverageDrop0to1: number
    agentSecurityFalsePositiveRateIncrease0to1: number
    agentSecurityLatencyP95IncreaseRatio: number
    agentSecurityPolicyDecisionAccuracyDrop0to1: number
    agentSecuritySecretScrubRateDrop0to1: number
    agentSecuritySourceOriginCoverageDrop0to1: number
    agentSecurityTaintPropagationCoverageDrop0to1: number
    agentTestingEvidenceCoverageDrop0to1: number
    agentTestingFaultInjectionCoverageDrop0to1: number
    agentTestingMethodologyCoverageDrop0to1: number
    agentTestingObservabilitySignalCoverageDrop0to1: number
    agentTestingResiliencePassRateDrop0to1: number
    agentTestingSafetyRegressionRateIncrease0to1: number
    agentTestingScenarioCoverageDrop0to1: number
    artifactAccuracyDrop0to1: number
    authorityHandoffRateDrop0to1: number
    backdoorAgentAttackSuccessRateIncrease0to1: number
    backdoorAgentCleanAccuracyDrop0to1: number
    backdoorAgentEvidenceCoverageDrop0to1: number
    backdoorAgentTrajectoryCoverageDrop0to1: number
    backdoorAgentTriggerPersistenceIncrease0to1: number
    backdoorAgentTriggerPropagationIncrease0to1: number
    chaosDropIncrease0to1: number
    chaosEvidenceCoverageDrop0to1: number
    chaosFailureTraceCoverageDrop0to1: number
    chaosImprovementEvalCoverageDrop0to1: number
    chaosProductionReliabilityDrop0to1: number
    chaosRecoveryPassRateDrop0to1: number
    chaosResilienceScoreDrop0to1: number
    controlCorrectabilityDrop0to1: number
    controlInterpretabilityDrop0to1: number
    controlInterruptibilityDrop0to1: number
    controlReversibilityDrop0to1: number
    costIncreaseRatio: number
    cpuAgenticCpuUtilizationIncrease0to1: number
    cpuAgenticEvidenceCoverageDrop0to1: number
    cpuAgenticFrameworkOverheadShareIncrease0to1: number
    cpuAgenticGpuUtilizationDrop0to1: number
    cpuAgenticLatencyP50IncreaseRatio: number
    cpuAgenticLatencyP95IncreaseRatio: number
    cpuAgenticLatencyP99IncreaseRatio: number
    cpuAgenticLlmInferenceShareShift0to1: number
    cpuAgenticMemoryIncreaseRatio: number
    cpuAgenticThroughputDropRatio: number
    cpuAgenticToolExecutionShareIncrease0to1: number
    ctfCheckpointCompletionDrop0to1: number
    ctfCompetitionImpactIncrease0to1: number
    ctfContaminationRiskIncrease0to1: number
    ctfExternalSearchUseRateIncrease0to1: number
    ctfFirstCorrectFlagForwardingRateDrop0to1: number
    ctfFlagSolveRateDrop0to1: number
    ctfIndependenceViolationRate0to1: number
    ctfIsolationViolationRate0to1: number
    ctfPartialCreditScoreDrop0to1: number
    ctfTraceCoverageRateDrop0to1: number
    divergenceMomentumDrop0to1: number
    documentDatasetEvidenceCoverageDrop0to1: number
    documentDatasetMemoryIncreaseRatio: number
    documentDatasetNumericMismatchRateIncrease0to1: number
    documentDatasetNumGuardCoverageDrop0to1: number
    documentDatasetQaAccuracyDrop0to1: number
    documentDatasetRagFaithfulnessDrop0to1: number
    documentDatasetSummaryQualityDrop0to1: number
    documentDatasetThroughputDropRatio: number
    documentDatasetTokenSavingsDropRatio: number
    driftStatistic: number
    errorAttributionRateIncrease0to1: number
    errorRateIncrease0to1: number
    evalTechniqueAlgorithmicFeedbackCoverageDrop0to1: number
    evalTechniqueDynamicGroundTruthPassRateDrop0to1: number
    evalTechniqueEvidenceCoverageDrop0to1: number
    evalTechniqueExactMatchAccuracyDrop0to1: number
    evalTechniqueLlmJudgeAgreementDrop0to1: number
    evalTechniquePairwiseWinRateDrop0to1: number
    evalTechniqueRagContextRelevanceDrop0to1: number
    evalTechniqueRagFaithfulnessDrop0to1: number
    evalTechniqueRealtimeFeedbackDrop0to1: number
    evalTechniqueSimulationGoalCompletionDrop0to1: number
    evalTechniqueStructuredValidationDrop0to1: number
    evalTechniqueToolImprovementDrop0to1: number
    evalTechniqueToolPrecisionDrop0to1: number
    evalTechniqueTrajectoryMatchRateDrop0to1: number
    formatQualityDrop0to1: number
    formulaIntegrityDrop0to1: number
    genomicsExpertCurationCoverageDrop0to1: number
    genomicsFormatConformanceRateDrop0to1: number
    genomicsPreprocessingQualityDrop0to1: number
    genomicsReferenceCoverageDrop0to1: number
    genomicsSelectionAccuracyDrop0to1: number
    genomicsStatisticalAnalysisAccuracyDrop0to1: number
    hedraRagEvidenceCoverageDrop0to1: number
    hedraRagLatencyP95IncreaseRatio: number
    hedraRagMemoryIncreaseRatio: number
    hedraRagReplayPassRateDrop0to1: number
    hedraRagThroughputDropRatio: number
    interactionTurnMeanShiftRatio: number
    invalidActionRateIncrease0to1: number
    kiteDocumentCountDropRatio: number
    kiteEvidenceCoverageDrop0to1: number
    kiteGradeDrop0to10: number
    kiteNormalizedGradeDrop0to1: number
    kiteQuestionCountDropRatio: number
    latencyIncreaseRatio: number
    legalAgentCitationCoverageDrop0to1: number
    legalAgentEvidenceCoverageDrop0to1: number
    legalAgentFinalSuccessDrop0to1: number
    legalAgentProcessRateDrop0to1: number
    legalAgentTokenCostIncreaseRatio: number
    legalAgentToolUseAccuracyDrop0to1: number
    llmRagBiasRiskIncrease0to1: number
    llmRagEvalSuiteEvidenceCoverageDrop0to1: number
    llmRagHallucinationRateIncrease0to1: number
    llmRagSemanticSimilarityDrop0to1: number
    localSystemEvidenceCoverageDrop0to1: number
    localSystemGhostDriverDetectionCoverageDrop0to1: number
    localSystemLocalOnlyPrivacyCoverageDrop0to1: number
    localSystemProactiveAlertCoverageDrop0to1: number
    localSystemProcessIdentityCoverageDrop0to1: number
    localSystemThermalBaselineDeviationIncrease0to1: number
    localSystemVoltageSpcAnomalyRateIncrease0to1: number
    naviBenchCrashRateIncrease0to1: number
    naviBenchEvidenceCoverageDrop0to1: number
    naviBenchExcludingCrashedScoreDrop0to1: number
    naviBenchLowerBoundScoreDrop0to1: number
    naviBenchStepCountIncreaseRatio: number
    naviBenchStepLimitViolationRateIncrease0to1: number
    naviBenchTaskSuccessDrop0to1: number
    naviBenchTrajectoryCoverageDrop0to1: number
    naviBenchVisualizationCoverageDrop0to1: number
    noMiraclAbstentionAccuracyDrop0to1: number
    noMiraclErrorRateIncrease0to1: number
    noMiraclEvidenceCoverageDrop0to1: number
    noMiraclHallucinationRateIncrease0to1: number
    noMiraclLanguageCoverageDrop0to1: number
    noMiraclRelevanceAccuracyDrop0to1: number
    noMiraclSubsetCoverageDrop0to1: number
    observabilityDeterministicCheckPassRateDrop0to1: number
    observabilityEvidenceCoverageDrop0to1: number
    observabilityReportCoverageDrop0to1: number
    observabilityResolutionScoreDrop0to1: number
    observabilityRubricScoreDrop0to1: number
    observabilityScenarioClockAlignmentRateDrop0to1: number
    observabilityTraceCoverageDrop0to1: number
    offPathAttemptMeanDropRatio: number
    ollamaMetricsEvidenceCoverageDrop0to1: number
    ollamaMetricsLoadedModelCountDropRatio: number
    ollamaMetricsModelLoadedRateDrop0to1: number
    ollamaMetricsModelRamIncreaseRatio: number
    ollamaMetricsRequestDurationP95IncreaseRatio: number
    ollamaMetricsRequestErrorRateIncrease0to1: number
    ollamaMetricsTimePerTokenIncreaseRatio: number
    osUniverseAutoValidationPassDrop0to1: number
    osUniverseEvidenceCoverageDrop0to1: number
    osUniverseStepCountIncreaseRatio: number
    osUniverseStepLimitViolationRateIncrease0to1: number
    osUniverseTaskSuccessDrop0to1: number
    osUniverseValidationErrorRateIncrease0to1: number
    passRateDrop0to1: number
    personaBehaviorCoverageDrop0to1: number
    personaHumanLikenessDrop0to1: number
    personaTaskGoalPreservationDrop0to1: number
    physicianBenchArtifactCoverageDrop0to1: number
    physicianBenchCheckpointPassRateDrop0to1: number
    physicianBenchClinicalActionSafetyDrop0to1: number
    physicianBenchDocumentationQualityDrop0to1: number
    physicianBenchEvidenceCoverageDrop0to1: number
    physicianBenchFhirDataAccessAccuracyDrop0to1: number
    physicianBenchTaskSuccessRateDrop0to1: number
    physicianBenchTrajectoryCoverageDrop0to1: number
    piArenaAgentTaskSuccessRateDrop0to1: number
    piArenaAttackSuccessRateIncrease0to1: number
    piArenaDefenseBlockRateDrop0to1: number
    piArenaEvidenceCoverageDrop0to1: number
    piArenaFalsePositiveRateIncrease0to1: number
    piArenaToolCallSuccessRateDrop0to1: number
    pokerEvalAllInAdjBbPer100Drop: number
    pokerEvalBbPer100Drop: number
    pokerEvalEvBbPer100Drop: number
    pokerEvalEvidenceCoverageDrop0to1: number
    pokerEvalHandCountDropRatio: number
    pokerEvalVpipShift0to1: number
    privacyPeerExposureRateIncrease0to1: number
    privacySafeguardActiveRateDrop0to1: number
    privacySensitiveDisclosureRateIncrease0to1: number
    privacySocialPressureIncrease0to1: number
    privacyWebDataMinimizationPassRateDrop0to1: number
    privacyWebEvidenceCoverageDrop0to1: number
    privacyWebLeakageRateIncrease0to1: number
    privacyWebModalLeakageDeltaIncrease0to1: number
    privacyWebSensitiveFieldExposureIncreaseRatio: number
    privacyWebTaskSuccessRateDrop0to1: number
    privacyWebUnnecessaryDisclosureRateIncrease0to1: number
    processDefectRateIncrease0to1: number
    ragAccuracyDrop0to1: number
    ragAnswerSupportCoverageDrop0to1: number
    ragCitationCoverageDrop0to1: number
    ragCompletenessDrop0to1: number
    ragDatasetBuilderEvidenceCoverageDrop0to1: number
    ragGeneratedDataFinalCoverageDrop0to1: number
    ragGenerationCostIncreaseRatio: number
    ragHallucinationRateIncrease0to1: number
    ragHumanVerificationCoverageDrop0to1: number
    ragNumericalAccuracyDrop0to1: number
    ragPassageGroundingCoverageDrop0to1: number
    ragQuestionCountDropRatio: number
    ragRetrievalTopKMeanShiftRatio: number
    ragSourceDocumentCountDropRatio: number
    ragStrategyEvidenceCoverageDrop0to1: number
    ragUtilizationDrop0to1: number
    recoveryBenchContextCoverageDrop0to1: number
    recoveryBenchCorruptedEnvironmentCoverageDrop0to1: number
    recoveryBenchEvidenceCoverageDrop0to1: number
    recoveryBenchFailureTraceCoverageDrop0to1: number
    recoveryBenchRecoveryRewardDrop0to1: number
    recoveryBenchRecoverySuccessRateDrop0to1: number
    recoveryBenchReplayIntegrityRateDrop0to1: number
    redTeamAttackCoverageDrop0to1: number
    redTeamComplianceDrop0to1: number
    redTeamDatasetCoverageDrop0to1: number
    redTeamGuardCoverageDrop0to1: number
    redTeamGuardScoreDrop0to1: number
    redTeamTaxonomyCoverageDrop0to1: number
    redTeamUnsafeResponseRateIncrease0to1: number
    refusalRateIncrease0to1: number
    researchGymArtifactCoverageDrop0to1: number
    researchGymBudgetOverrunRateIncrease0to1: number
    researchGymInspectionPassRateDrop0to1: number
    researchGymScoreImprovementDrop0to1: number
    researchGymSubtaskCompletionDrop0to1: number
    researchGymViolationRateIncrease0to1: number
    sapAgentEvalEnterpriseContextCoverageDrop0to1: number
    sapAgentEvalEvidenceCoverageDrop0to1: number
    sapAgentEvalObjectiveCoverageDrop0to1: number
    sapAgentEvalProcessCoverageDrop0to1: number
    scalingLawEvidenceCoverageDrop0to1: number
    scalingLawNmaeIncrease: number
    scalingLawNmseIncrease: number
    scalingLawR2Drop: number
    scoreDrop0to1: number
    socialHarmPrevalenceIncrease0to1: number
    socialLexicalDiversityDrop0to1: number
    socialSemanticAlignmentDrop0to1: number
    socialSentimentMeanShift: number
    solutionPathMeanDropRatio: number
    strandsBenchmarkHarnessCostIncreaseRatio: number
    strandsBenchmarkHarnessEvidenceCoverageDrop0to1: number
    strandsBenchmarkHarnessLatencyP95IncreaseRatio: number
    strandsBenchmarkHarnessPatchApplyRateDrop0to1: number
    strandsBenchmarkHarnessTaskSuccessDrop0to1: number
    strandsBenchmarkHarnessTestPassRateDrop0to1: number
    strandsBenchmarkHarnessTrajectoryCoverageDrop0to1: number
    toolAnswerVerificationDrop0to1: number
    toolCallMeanShiftRatio: number
    toolCallValidityDrop0to1: number
    toolEvalImprovementDrop0to1: number
    toolJudgeAgreementDrop0to1: number
    toolRolloutDiversityDrop0to1: number
    toolUseRewardDrop0to1: number
    tradingClaimValidationFailureIncrease0to1: number
    tradingDrawdownIncrease0to1: number
    tradingMemoryRetrievalHitRateDrop0to1: number
    tradingPnlDropPct: number
    tradingProviderFallbackRateIncrease0to1: number
    tradingRiskLimitViolationIncrease0to1: number
    tradingRiskRewardDropRatio: number
    tradingVisionChartAgreementDrop0to1: number
    tradingWinRateDrop0to1: number
    webOperatorLlmEvaluationDrop0to1: number
    webOperatorMismatchRateIncrease0to1: number
    webOperatorReplayCoverageDrop0to1: number
    webOperatorSelfReportOverclaimIncrease0to1: number
    webOperatorStepLimitViolationRateIncrease0to1: number
    webOperatorTaskReliabilityDrop0to1: number
    webOperatorTaskTimeIncreaseRatio: number