@tangle-network/agent-eval 0.113.0 → 0.114.0

package/dist/{chunk-CX5RBMPW.js → chunk-VHPD6AXX.js}

@@ -12,7 +12,7 @@ import {
   runImprovementLoop,
   surfaceContentHash,
   surfaceHash
-} from "./chunk-F2NV5PQ6.js";
+} from "./chunk-N6MTC3GK.js";
 import {
   estimateCost,
   isModelPriced
@@ -23,7 +23,7 @@ import {
   planCampaignRun,
   runCampaign,
   summarizeBackendIntegrity
-} from "./chunk-XNCF3JU3.js";
+} from "./chunk-FAOEFFRT.js";
 import {
   admitPolicyEdit,
   applyPolicyEditToSurface,
@@ -41,7 +41,7 @@ import {
   mulberry32,
   pairedBootstrap,
   weightedComposite
-} from "./chunk-VC43KQCK.js";
+} from "./chunk-PJQFMIOX.js";
 import {
   analyzeTraces
 } from "./chunk-RPDDVKI7.js";
@@ -4966,4 +4966,4 @@ export {
   verifyCodeSurface,
   resolveWorktreePath
 };
-//# sourceMappingURL=chunk-CX5RBMPW.js.map
+//# sourceMappingURL=chunk-VHPD6AXX.js.map

package/dist/{chunk-BOLYPSAE.js → chunk-WBOGKYM4.js}

@@ -1,9 +1,9 @@
 import {
   summarizeBackendIntegrity
-} from "./chunk-XNCF3JU3.js";
+} from "./chunk-FAOEFFRT.js";
 import {
   paretoChart
-} from "./chunk-EMKORATZ.js";
+} from "./chunk-DPZAEKA6.js";
 import {
   cohensD,
   pairedBootstrap,
@@ -12,7 +12,7 @@ import {
   pearsonR,
   requiredSampleSize,
   spearmanR
-} from "./chunk-VC43KQCK.js";
+} from "./chunk-PJQFMIOX.js";
 import {
   llmSpans
 } from "./chunk-MHNQWM4I.js";
@@ -822,4 +822,4 @@ export {
   HoldoutAuditor,
   analyzeRuns
 };
-//# sourceMappingURL=chunk-BOLYPSAE.js.map
+//# sourceMappingURL=chunk-WBOGKYM4.js.map

package/dist/{chunk-3LDPUYAC.js → chunk-X4UCIOTZ.js}

@@ -1,7 +1,7 @@
 import {
   pearsonR,
   spearmanR
-} from "./chunk-VC43KQCK.js";
+} from "./chunk-PJQFMIOX.js";
 
 // src/meta-eval/rubric-predictive-validity.ts
 async function rubricPredictiveValidity(input) {
@@ -133,4 +133,4 @@ function makeRng(seed) {
 export {
   rubricPredictiveValidity
 };
-//# sourceMappingURL=chunk-3LDPUYAC.js.map
+//# sourceMappingURL=chunk-X4UCIOTZ.js.map

package/dist/contract/index.d.ts

@@ -1,7 +1,7 @@
 import { S as Scenario, M as MutableSurface, D as DispatchContext, b as JudgeConfig, f as SurfaceProposer, g as Gate, L as LabeledScenarioStore, C as CampaignResult, h as GateDecision } from '../types-CgSlO6wT.js';
 export { i as CampaignAggregates, j as CampaignArtifactWriter, k as CampaignCellResult, l as CampaignCostMeter, d as CampaignTraceWriter, m as CodeSurface, c as Dispatch, n as GateContext, G as GateResult, o as GenerationCandidate, e as GenerationRecord, a as JudgeDimension, J as JudgeScore, p as Mutator, O as OptimizationProposer, q as OptimizerConfig, r as SessionScript } from '../types-CgSlO6wT.js';
-import { L as LoopProvenanceRecord, P as PowerPreflight, R as RunEvalOptions } from '../provenance-DChJ2oZ4.js';
-export { A as AxisEvidence, a as AxisVerdict, B as BuildEvidenceVectorOptions, D as DefaultProductionGateOptions, E as EvidenceVector, b as EvolutionaryProposerOptions, H as HeldOutGateOptions, O as ObjectiveSource, c as ParetoSignificanceGateOptions, d as PromotionObjective, e as PromotionPolicy, f as buildEvidenceVector, g as composeGate, h as defaultProductionGate, i as evolutionaryProposer, j as heldOutGate, p as paretoPolicy, k as paretoSignificanceGate, r as runEval } from '../provenance-DChJ2oZ4.js';
+import { L as LoopProvenanceRecord, P as PowerPreflight, R as RunEvalOptions } from '../provenance-BZmpWmn4.js';
+export { A as AxisEvidence, a as AxisVerdict, B as BuildEvidenceVectorOptions, D as DefaultProductionGateOptions, E as EvidenceVector, b as EvolutionaryProposerOptions, H as HeldOutGateOptions, O as ObjectiveSource, c as ParetoSignificanceGateOptions, d as PromotionObjective, e as PromotionPolicy, f as buildEvidenceVector, g as composeGate, h as defaultProductionGate, i as evolutionaryProposer, j as heldOutGate, p as paretoPolicy, k as paretoSignificanceGate, r as runEval } from '../provenance-BZmpWmn4.js';
 import { R as RunOptimizationOptions, a as RunImprovementLoopResult } from '../gepa-DolL_Fko.js';
 export { G as GepaProposerOptions, b as RunCampaignOptions, c as RunImprovementLoopOptions, g as gepaProposer, r as runCampaign, d as runImprovementLoop } from '../gepa-DolL_Fko.js';
 import { C as CampaignStorage } from '../storage-Dw_f7WMt.js';
@@ -16,7 +16,7 @@ export { A as AnalystFinding } from '../kind-factory-20hcaYpf.js';
 import { A as AnalyzeRunsOptions } from '../analyze-runs-Dmz6LA9e.js';
 export { a as analyzeRuns } from '../analyze-runs-Dmz6LA9e.js';
 export { C as CodeAgentSessionDiagnostic, a as CodeAgentSessionIntakeOptions, b as CodeAgentSessionIntakeResult, c as CodeAgentSessionMetrics, d as CodeAgentSessionSource, P as ParsedCodeAgentJsonl, f as fromClaudeCodeSession, e as fromCodexSession, g as fromKimiCodeSession, h as fromOpenCodeSession, i as fromPiSession, j as fromPigraphSession, p as parseCodeAgentJsonl } from '../code-agent-session-yitf9I-F.js';
-import '../statistics-CDfpwIdX.js';
+import '../statistics-oUbOJe-S.js';
 import '../judge-calibration-7C-IDmKr.js';
 import '../types-C7DGg5ex.js';
 import '@tangle-network/tcloud';

package/dist/contract/index.js

@@ -12,7 +12,7 @@ import {
 } from "../chunk-ZZUXHH3R.js";
 import {
   analyzeRuns
-} from "../chunk-BOLYPSAE.js";
+} from "../chunk-WBOGKYM4.js";
 import {
   buildEvidenceVector,
   composeGate,
@@ -28,13 +28,13 @@ import {
   runImprovementLoop,
   surfaceContentHash,
   surfaceHash
-} from "../chunk-F2NV5PQ6.js";
+} from "../chunk-N6MTC3GK.js";
 import "../chunk-VI2UW6B6.js";
 import {
   fsCampaignStorage,
   inMemoryCampaignStorage,
   runCampaign
-} from "../chunk-XNCF3JU3.js";
+} from "../chunk-FAOEFFRT.js";
 import {
   buildDefaultAnalystRegistry
 } from "../chunk-LVTGFSHF.js";
@@ -43,9 +43,9 @@ import {
   FileSystemOutcomeStore,
   InMemoryOutcomeStore
 } from "../chunk-3RF76KTD.js";
-import "../chunk-7QPK6W7R.js";
-import "../chunk-EMKORATZ.js";
-import "../chunk-VC43KQCK.js";
+import "../chunk-ARU2PZFM.js";
+import "../chunk-DPZAEKA6.js";
+import "../chunk-PJQFMIOX.js";
 import "../chunk-MHNQWM4I.js";
 import "../chunk-GGE4NNQT.js";
 import {

package/dist/index.d.ts

@@ -22,8 +22,8 @@ import { T as TraceEmitter } from './emitter-BRchAAAx.js';
 export { R as RunCompleteHook, a as RunCompleteHookContext, S as SpanHandle, b as TraceEmitterOptions, l as llmSpanFromProvider } from './emitter-BRchAAAx.js';
 import { h as ReleaseConfidenceThresholds, f as ReleaseConfidenceScorecard } from './release-report-oBfOz8ku.js';
 export { A as ActionableSideInfo, o as AsiSeverity, B as BootstrapOptions, a as BootstrapResult, J as JudgeReplayGateArgs, R as ReleaseConfidenceAxis, b as ReleaseConfidenceAxisName, c as ReleaseConfidenceInput, d as ReleaseConfidenceIssue, e as ReleaseConfidenceMetrics, g as ReleaseConfidenceStatus, i as ReleaseTraceEvidence, j as RenderReleaseReportOptions, V as Verdict, k as assertReleaseConfidence, l as bootstrapCi, m as evaluateReleaseConfidence, n as judgeReplayGate, r as renderReleaseReport } from './release-report-oBfOz8ku.js';
-import { P as PairedBootstrapOptions, M as McNemarResult, R as RiskDifferenceResult, a as PairedBootstrapResult } from './statistics-CDfpwIdX.js';
-export { c as CliffsMagnitude, d as CorpusAgreementOptions, e as CorpusAgreementPerDimension, C as CorpusAgreementReport, f as CorpusScoreRecord, g as EProcess, h as EProcessOptions, E as EProcessState, i as EProcessStep, j as ProportionInterval, W as WeightedCompositeInput, k as WeightedCompositeResult, b as benjaminiHochberg, l as bonferroni, m as cliffsDelta, n as cohensD, o as confidenceInterval, q as corpusInterRaterAgreement, r as corpusInterRaterAgreementFromJudgeScores, s as eProcess, t as holm, u as interRaterReliability, v as interpretCliffs, x as mannWhitneyU, y as mcnemar, z as mcnemarPower, A as mcnemarRequiredN, B as mulberry32, D as normalizeScores, p as pairedBootstrap, F as pairedMde, G as pairedRiskDifference, H as pairedTTest, I as partialCredit, J as passAtK, K as pearsonR, L as ranks, N as requiredSampleSize, O as spearmanR, Q as weightedComposite, S as weightedMean, w as wilcoxonSignedRank, T as wilson } from './statistics-CDfpwIdX.js';
+import { P as PairedBootstrapOptions, M as McNemarResult, R as RiskDifferenceResult, a as PairedBootstrapResult } from './statistics-oUbOJe-S.js';
+export { c as CliffsMagnitude, d as CorpusAgreementOptions, e as CorpusAgreementPerDimension, C as CorpusAgreementReport, f as CorpusScoreRecord, g as EProcess, h as EProcessOptions, E as EProcessState, i as EProcessStep, j as PairedSignTestResult, k as ProportionInterval, S as SignTestAlternative, W as WeightedCompositeInput, l as WeightedCompositeResult, b as benjaminiHochberg, m as bonferroni, n as cliffsDelta, o as cohensD, q as confidenceInterval, r as corpusInterRaterAgreement, s as corpusInterRaterAgreementFromJudgeScores, t as eProcess, u as holm, v as interRaterReliability, x as interpretCliffs, y as mannWhitneyU, z as mcnemar, A as mcnemarPower, B as mcnemarRequiredN, D as mulberry32, F as normalizeScores, p as pairedBootstrap, G as pairedMde, H as pairedRiskDifference, I as pairedSignTest, J as pairedTTest, K as partialCredit, L as passAtK, N as pearsonR, O as ranks, Q as requiredSampleSize, T as spearmanR, U as weightedComposite, V as weightedMean, w as wilcoxonSignedRank, X as wilson } from './statistics-oUbOJe-S.js';
 import { OtelExporter, OtelExportConfig } from './traces.js';
 export { CaptureFetchContext, CaptureFetchOptions, DEFAULT_REDACTION_RULES, ExportableSpan, ExtractedUsage, FlattenOtlpOptions, LLM_CACHED_TOKENS, LLM_CACHED_TOKEN_ATTR_KEYS, LLM_COST_ATTR_KEYS, LLM_COST_USD, LLM_INPUT_TOKENS, LLM_INPUT_TOKEN_ATTR_KEYS, LLM_MODEL_ATTR_KEYS, LLM_MODEL_NAME, LLM_OUTPUT_TOKENS, LLM_OUTPUT_TOKEN_ATTR_KEYS, OPENINFERENCE_SPAN_KIND, OTEL_AGENT_EVAL_SCOPE, OtlpExport, OtlpFileTraceStore, OtlpFileTraceStoreOptions, OtlpFlatLine, OtlpResourceSpans, OtlpSpan, OtlpToRunRecordsOptions, OtlpTraceRunRecord, ProjectedOtlpSpan, REDACTION_VERSION, RedactionReport, RedactionRule, ReplayCache, ReplayCacheEntry, ReplayCacheMissError, ReplayCacheStats, ReplayFetchOptions, SPAN_KIND_ATTR_KEYS, SpanNotFoundError, TOOL_NAME, TOOL_NAME_ATTR_KEYS, TRACE_ANALYST_ACTOR_DESCRIPTION, TRACE_ANALYST_ACTOR_DESCRIPTION_VERSION, TRACE_ANALYST_SUBAGENT_DESCRIPTION, TraceAggregate, TraceAnalystHookOptions, TraceFileMissingError, TraceInsightContext, TraceInsightFinding, TraceInsightPanelRole, TraceInsightPromptInput, TraceInsightQualityGate, TraceInsightQuestion, TraceInsightReadiness, TraceInsightSuite, TraceInsightTask, TraceNotFoundError, TraceStoreSource, TraceStoreToOtlpOptions, TracesToOtlpResult, asNumber, asString, buildTraceAnalystTools, buildTraceInsightContext, buildTraceInsightPrompt, captureFetchToRawSink, convertTraceStoresToOtlp, createOtelExporter, createOtelTracingStore, createReplayFetch, defaultTraceInsightPanel, describeTraceInsightScope, domainEvidencePattern, exportRunAsOtlp, extractOtlpAttributes, extractUsage, extractUsageFromResponse, extractUsageFromSse, firstNumberAttr, firstStringAttr, flattenOtlpExportToNdjson, inferDomainKeywords, inferOtlpKind, iterateRawCalls, otelRunCompleteHook, otlpToRunRecords, otlpToTraceRunRecords, planTraceInsightQuestions, projectOtlpFlatLine, readOtlpStatus, redactString, redactValue, scoreTraceInsightReadiness, stringField, tokenizeDomainWords, traceAnalystFunctionGroup, traceAnalystOnRunComplete, traceSpanKindToOpenInferenceKind } from './traces.js';
 import { a as AnalyzeTracesInput, A as AnalyzeTracesOptions, b as AnalyzeTracesResult } from './analyst-C8HHvfJp.js';

package/dist/index.js

@@ -9,12 +9,12 @@ import {
   checkBehavioralCanary,
   checkCanaries,
   runBehavioralCanaries
-} from "./chunk-BOLYPSAE.js";
+} from "./chunk-WBOGKYM4.js";
 import {
   BENCHMARK_SPLIT_SEED,
   benchmarks_exports,
   deterministicSplit
-} from "./chunk-VLNGPUJJ.js";
+} from "./chunk-3LXTCTWL.js";
 import {
   DEFAULT_RULES,
   buildTrajectory,
@@ -23,7 +23,7 @@ import {
   computeToolUseMetrics,
   iqr,
   welchsTTest
-} from "./chunk-YWGKQARF.js";
+} from "./chunk-NYFUT3B3.js";
 import {
   analyzeSeries
 } from "./chunk-BOD4O7OF.js";
@@ -59,7 +59,7 @@ import {
   llmJudge,
   parseCorrectnessResponse,
   verifyCompletion
-} from "./chunk-CX5RBMPW.js";
+} from "./chunk-VHPD6AXX.js";
 import {
   DEFAULT_MUTATION_PRIMITIVES,
   DEFAULT_RED_TEAM_CORPUS,
@@ -83,7 +83,7 @@ import {
   scoreRedTeamOutput,
   surfaceContentHash,
   toolNamesForRun
-} from "./chunk-F2NV5PQ6.js";
+} from "./chunk-N6MTC3GK.js";
 import {
   MODEL_PRICING,
   MetricsCollector,
@@ -102,7 +102,7 @@ import {
   fileVerdictCache,
   inMemoryVerdictCache,
   summarizeBackendIntegrity
-} from "./chunk-XNCF3JU3.js";
+} from "./chunk-FAOEFFRT.js";
 import {
   DEFAULT_COMPLEXITY_WEIGHTS,
   FindingsStore,
@@ -179,11 +179,11 @@ import {
   evaluateReleaseConfidence,
   judgeReplayGate,
   renderReleaseReport
-} from "./chunk-RT2AFUXM.js";
+} from "./chunk-MOXWMGPC.js";
 import {
   runEvalCampaign
-} from "./chunk-WZOIFFG2.js";
-import "./chunk-7QPK6W7R.js";
+} from "./chunk-LOBMT6SB.js";
+import "./chunk-ARU2PZFM.js";
 import {
   evaluateInterimReleaseConfidence,
   pairedEvalueSequence
@@ -194,7 +194,7 @@ import {
   paretoChart,
   researchReport,
   summaryTable
-} from "./chunk-EMKORATZ.js";
+} from "./chunk-DPZAEKA6.js";
 import {
   benjaminiHochberg,
   bonferroni,
@@ -219,6 +219,7 @@ import {
   pairedBootstrap,
   pairedMde,
   pairedRiskDifference,
+  pairedSignTest,
   pairedTTest,
   partialCredit,
   passAtK,
@@ -233,7 +234,7 @@ import {
   weightedMean,
   wilcoxonSignedRank,
   wilson
-} from "./chunk-VC43KQCK.js";
+} from "./chunk-PJQFMIOX.js";
 import {
   FileSystemTraceStore,
   InMemoryTraceStore,
@@ -12003,6 +12004,7 @@ export {
   pairedEvalueSequence,
   pairedMde,
   pairedRiskDifference,
+  pairedSignTest,
   pairedTTest,
   paraphraseRobustness,
   paraphraseRobustnessScenarios,