@tangle-network/agent-eval 0.181.0 → 0.182.0

package/dist/cli.js

@@ -1,9 +1,9 @@
 #!/usr/bin/env node
 import { i as runRolloutReleaseCli } from "./hf-dataset-D8_RNIis.js";
 import { t as createChatClient } from "./chat-client-CkmjYlfB.js";
-import { t as runAnalystBenchmarkCommand } from "./benchmark-command-B57n9vjz.js";
+import { t as runAnalystBenchmarkCommand } from "./benchmark-command-7X7HPNHh.js";
 import { a as runRpcBatch, o as runRpcOnce, p as handleVersion, r as startServerAsync, s as buildOpenApi } from "./server-D_cjseFN.js";
-import { t as runSupervisorRunCommand } from "./report-command-V1ecVgAv.js";
+import { t as runSupervisorRunCommand } from "./report-command-BvqaxlVG.js";
 import { writeFileSync } from "node:fs";
 //#region src/cli-config.ts
 /**

package/dist/contract/index.js

@@ -1,25 +1,25 @@
 import { s as ValidationError } from "../errors-Dngq5h35.js";
 import { a as hashCanonical } from "../canonical-DPyQ_rpt.js";
 import { r as pairedBootstrap } from "../paired-tests-C8iCsioC.js";
-import { s as decidePairedPromotion } from "../run-record-Br-Yzt_k.js";
-import { Z as campaignSplitDigest, it as heldoutSignificance } from "../campaign-evidence-B8oF9xQ6.js";
-import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-Ddu33JH9.js";
+import { s as decidePairedPromotion } from "../run-record-DPa2PeHx.js";
+import { Z as campaignSplitDigest, it as heldoutSignificance } from "../campaign-evidence-PYPUftRy.js";
+import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-DSIPB4hi.js";
 import { a as parseRunRecordSafe } from "../run-record-DualPTn2.js";
-import { X as runCampaign, Y as runEval, a as transientDispatchFailure, c as compareOptimizationMethods, h as runImprovementLoop, t as llmJudge, z as defaultProductionGate } from "../llm-judge-DEFZeSiu.js";
+import { X as runCampaign, Y as runEval, a as transientDispatchFailure, c as compareOptimizationMethods, h as runImprovementLoop, t as llmJudge, z as defaultProductionGate } from "../llm-judge-BPFR12VJ.js";
 import { i as isModelPriced, n as estimateCost } from "../metrics-Qv-cpptD.js";
 import { i as CostLedger } from "../cost-ledger-B1qx30B4.js";
 import { f as mapConcurrentRange } from "../journal-Cs9f7385.js";
 import { C as inMemoryCampaignStorage, S as fsCampaignStorage } from "../external-optimizer-subprocess-D4dzUBZI.js";
 import { i as makeProposalFinding } from "../types-DQ0e2E7y.js";
 import { K as classifyOtlpSpanRole, q as isOtlpModelCall } from "../kind-factory-D5HOW3R0.js";
-import { t as buildDefaultAnalystRegistry } from "../default-registry-aL7xUrUz.js";
+import { t as buildDefaultAnalystRegistry } from "../default-registry-DHWwIwa8.js";
 import { LLM_MODEL_ATTR_KEYS, SPAN_KIND_ATTR_KEYS } from "../trace-attributes.js";
 import { t as extractUsage } from "../extract-usage-BrQ8mCLX.js";
 import { t as createChatClient } from "../chat-client-CkmjYlfB.js";
 import { n as InMemoryOutcomeStore, t as FileSystemOutcomeStore } from "../outcome-store-BXlkwMPR.js";
 import { i as summarizeTraceErrors, n as recordAggregateMeasurements, r as summarizeExecutionMeasurements, t as readTaskFailureLabels } from "../task-failure-attributes-B072KO_n.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-CL_0aArC.js";
-import { n as paretoPolicy, r as paretoSignificanceGate, t as buildEvidenceVector } from "../promotion-policy-CDMMxzb6.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-B2IEaCeF.js";
+import { n as paretoPolicy, r as paretoSignificanceGate, t as buildEvidenceVector } from "../promotion-policy-BKFrl7CC.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";
 import { createReadStream } from "node:fs";

package/dist/{default-registry-aL7xUrUz.js → default-registry-DHWwIwa8.js}

@@ -4,7 +4,7 @@ import { a as assertValidAnalystUsageReceipt, c as validateUsageSettlementTimeou
 import { B as findingSubjectGrammarPromptFor, X as snapshotExactExecutionPlan, Y as snapshotExactExecutionComponentIdentity, k as spanEpochMillis, t as createTraceAnalyst } from "./kind-factory-D5HOW3R0.js";
 import { LLM_CONTEXT_TOKENS, LLM_INPUT_TOKEN_ATTR_KEYS, LLM_OUTPUT_TOKEN_ATTR_KEYS, TOOL_NAME_ATTR_KEYS } from "./trace-attributes.js";
 import { t as executionTrackByLane } from "./execution-tracks-CpgFPpS5.js";
-import { t as analyzeSupervisorRunIntegrity } from "./integrity-DH5ng72x.js";
+import { t as analyzeSupervisorRunIntegrity } from "./integrity-CgQce0FV.js";
 import { randomUUID } from "node:crypto";
 import { z } from "zod";
 //#region src/feedback-trajectory-review.ts
@@ -2363,4 +2363,4 @@ function buildDefaultAnalystRegistry(options = {}) {
 //#endregion
 export { completedAnalystReviewQuality as _, KNOWLEDGE_POISONING_KIND_SPEC as a, validateAnalystReviewDecisions as b, FAILURE_MODE_KIND_SPEC as c, emitControlIntegrityFindings as d, behavioralAnalyst as f, assertUniqueFindingIds as g, analystRunDigest as h, DEFAULT_TRACE_ANALYST_KINDS as i, CONTROL_INTEGRITY_ANALYST as l, analystFindingDigest as m, AnalystRegistry as n, KNOWLEDGE_GAP_KIND_SPEC as o, deriveEfficiencyFindings as p, ExactAnalystRunExecutionError as r, IMPROVEMENT_KIND_SPEC as s, buildDefaultAnalystRegistry as t, ControlIntegrityAnalyst as u, readAnalystReview as v, snapshotAnalystRun as y };
 
-//# sourceMappingURL=default-registry-aL7xUrUz.js.map
+//# sourceMappingURL=default-registry-DHWwIwa8.js.map