@tangle-network/agent-eval 0.145.1 → 0.145.3

package/dist/cli.js

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
 import { o as runRolloutReleaseCli } from "./hf-dataset-XggBupCr.js";
-import { n as runAnalystBenchmarkCommand } from "./benchmark-command-Dr4cX8QG.js";
+import { n as runAnalystBenchmarkCommand } from "./benchmark-command-Iaqlp3DU.js";
 import { a as runRpcBatch, o as runRpcOnce, p as handleVersion, r as startServerAsync, s as buildOpenApi } from "./server-ulsOdrTI.js";
 import { writeFileSync } from "node:fs";
 //#region src/cli-config.ts

package/dist/contract/index.js

@@ -1,12 +1,12 @@
 import { s as ValidationError } from "../errors-Dngq5h35.js";
 import { r as pairedBootstrap } from "../paired-tests-BHIhYVdu.js";
-import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-iqjT--ZZ.js";
+import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-CAL5HOz9.js";
 import { i as parseRunRecordSafe } from "../run-record-BvHPVS-i.js";
-import { G as runEval, K as runCampaign, f as runImprovementLoop, nt as campaignSplitDigest, t as llmJudge, x as compareOptimizationMethods, z as defaultProductionGate } from "../llm-judge-dZ8P6nGI.js";
+import { G as runEval, K as runCampaign, f as runImprovementLoop, nt as campaignSplitDigest, t as llmJudge, x as compareOptimizationMethods, z as defaultProductionGate } from "../llm-judge-C2lQBv7h.js";
 import { i as isModelPriced, n as estimateCost } from "../metrics-Cl0L1KUy.js";
 import { i as CostLedger } from "../cost-ledger-BSe92yAV.js";
 import { d as mapConcurrentRange } from "../ledger-core-BmZt19oQ.js";
-import { S as inMemoryCampaignStorage, x as fsCampaignStorage } from "../external-optimizer-subprocess-DrJ9hR8u.js";
+import { S as inMemoryCampaignStorage, x as fsCampaignStorage } from "../external-optimizer-subprocess-BhKYK0Jv.js";
 import { a as heldoutSignificance, s as decidePairedPromotion } from "../power-preflight-DEw-uC7q.js";
 import { r as makeProposalFinding } from "../types-BI4fT3HN.js";
 import { $ as classifyOtlpSpanRole, et as isOtlpModelCall } from "../kind-factory-CPmSd58s.js";
@@ -15,7 +15,7 @@ import { LLM_MODEL_ATTR_KEYS, SPAN_KIND_ATTR_KEYS } from "../trace-attributes.js
 import { t as extractUsage } from "../extract-usage-BrQ8mCLX.js";
 import { n as InMemoryOutcomeStore, t as FileSystemOutcomeStore } from "../outcome-store-ChBKlTd_.js";
 import { i as summarizeTraceErrors, n as recordAggregateMeasurements, r as summarizeExecutionMeasurements, t as readTaskFailureLabels } from "../task-failure-attributes-CpQ4y5RD.js";
-import { a as externalTextOptimizationMethod, c as REFERENCE_EQUIVALENCE_INPUT_LIMITS, d as runReferenceEquivalenceJudge, i as composeGate, l as REFERENCE_EQUIVALENCE_JUDGE_VERSION, n as gepaOptimizationMethod, r as heldOutGate, t as skillOptOptimizationMethod, u as createReferenceEquivalenceJudge } from "../skillopt-optimization-method-jjdnc3YK.js";
+import { a as externalTextOptimizationMethod, c as REFERENCE_EQUIVALENCE_INPUT_LIMITS, d as runReferenceEquivalenceJudge, i as composeGate, l as REFERENCE_EQUIVALENCE_JUDGE_VERSION, n as gepaOptimizationMethod, r as heldOutGate, t as skillOptOptimizationMethod, u as createReferenceEquivalenceJudge } from "../skillopt-optimization-method-B__Jnto4.js";
 import { n as paretoPolicy, r as paretoSignificanceGate, t as buildEvidenceVector } from "../promotion-policy-xzA40Evo.js";
 import { createHash } from "node:crypto";
 import { agentCandidateBenchmarkSuiteSchema, agentCandidateBenchmarkTaskSchema, agentCandidateBundleSchema, agentCandidateEvaluationPolicySchema, agentCandidateExperimentSchema, agentImprovementMeasuredComparisonSchema, agentProfileImprovementExperimentSchema, agentProfileImprovementMeasuredComparisonSchema, agentProfileImprovementRunCellSchema, agentProfileImprovementRunReceiptSchema, agentProfileImprovementSuiteInputsSchema, agentProfileImprovementSuiteSchema, agentProfileImprovementTaskSchema, candidateExecutionEvidenceSchema, canonicalCandidateDigest, canonicalCandidateJson, numbersApproximatelyEqual, omitTopLevelDigest } from "@tangle-network/agent-interface";

package/dist/{define-agent-eval-iqjT--ZZ.js → define-agent-eval-CAL5HOz9.js}

@@ -9,9 +9,9 @@ import { r as welchsTTest } from "./baseline-BhPRQBVn.js";
 import { r as observedSplitScore } from "./reward-nw2xZGZG.js";
 import { r as modelHasSnapshot, s as validateRunRecord } from "./run-record-BvHPVS-i.js";
 import { r as paretoChart } from "./summary-report-Blysd6Z2.js";
-import { A as surfaceHash, G as runEval, J as resolveRunDir, f as runImprovementLoop, k as surfaceContentHash, o as emitLoopProvenance, s as loopProvenanceArgsFromResult, y as assertOptimizationResult, z as defaultProductionGate } from "./llm-judge-dZ8P6nGI.js";
+import { A as surfaceHash, G as runEval, J as resolveRunDir, f as runImprovementLoop, k as surfaceContentHash, o as emitLoopProvenance, s as loopProvenanceArgsFromResult, y as assertOptimizationResult, z as defaultProductionGate } from "./llm-judge-C2lQBv7h.js";
 import { c as campaignCellTaskScore, l as campaignCellToRunRecord, o as campaignCellExecutionEvidence, s as campaignCellJudgeDimensions } from "./reward-hacking-DNgjilrV.js";
-import { S as inMemoryCampaignStorage, b as createRunCostLedger, x as fsCampaignStorage } from "./external-optimizer-subprocess-DrJ9hR8u.js";
+import { S as inMemoryCampaignStorage, b as createRunCostLedger, x as fsCampaignStorage } from "./external-optimizer-subprocess-BhKYK0Jv.js";
 import { t as powerPreflight } from "./power-preflight-DEw-uC7q.js";
 import { t as createHostedClient } from "./client-LIuo-KPv.js";
 //#region src/contamination-guard.ts
@@ -1491,4 +1491,4 @@ function requirePositiveInteger(value, field) {
 //#endregion
 export { summarizeExecution as a, analyzeRuns as i, SelfImproveRunError as n, checkCanaries as o, selfImprove as r, defineAgentEval as t };
 
-//# sourceMappingURL=define-agent-eval-iqjT--ZZ.js.map
+//# sourceMappingURL=define-agent-eval-CAL5HOz9.js.map