@tangle-network/agent-eval 0.102.1 → 0.103.1

package/dist/index.js

@@ -9,7 +9,7 @@ import {
   checkBehavioralCanary,
   checkCanaries,
   runBehavioralCanaries
-} from "./chunk-6Q2DYRWV.js";
+} from "./chunk-TCPP4Z5Q.js";
 import {
   classifyEuAiRisk,
   euAiActReport,
@@ -34,15 +34,22 @@ import {
   summarizeRecords
 } from "./chunk-STGVSCDH.js";
 import {
+  assertProductBenchmarkRun,
+  buildProductBenchmarkManifest,
+  exportProductBenchmark,
+  exportProductBenchmarkRuns,
   findProductBenchmarkArtifacts,
   productBenchmarkIntegrityFailures,
+  productBenchmarkMutableSurfaces,
+  productBenchmarkRepoIdentity,
   productBenchmarkSplits,
   readProductBenchmarkManifest,
   readProductBenchmarkRecords,
+  runRecordToProductBenchmarkRecord,
   validateProductBenchmarkManifest,
   validateProductBenchmarkRecord,
   validateProductBenchmarkRun
-} from "./chunk-63MBSQTX.js";
+} from "./chunk-RQNOLV3I.js";
 import {
   CODING_HARNESSES,
   HARNESS_NATIVE_MODEL,
@@ -65,7 +72,7 @@ import {
   llmJudge,
   parseCorrectnessResponse,
   verifyCompletion
-} from "./chunk-CMJSTXUR.js";
+} from "./chunk-YFYIOSNV.js";
 import {
   DEFAULT_MUTATION_PRIMITIVES,
   DEFAULT_RED_TEAM_CORPUS,
@@ -89,7 +96,7 @@ import {
   scoreRedTeamOutput,
   surfaceContentHash,
   toolNamesForRun
-} from "./chunk-Z3FLN24V.js";
+} from "./chunk-2B4HPFXN.js";
 import {
   BackendIntegrityError,
   assertRealBackend,
@@ -99,7 +106,7 @@ import {
   fileVerdictCache,
   inMemoryVerdictCache,
   summarizeBackendIntegrity
-} from "./chunk-QIT2XZ4E.js";
+} from "./chunk-VNM52AGA.js";
 import {
   MODEL_PRICING,
   MetricsCollector,
@@ -121,7 +128,7 @@ import {
   computeToolUseMetrics,
   iqr,
   welchsTTest
-} from "./chunk-IH7LYRHL.js";
+} from "./chunk-BXZVBX3D.js";
 import {
   analyzeSeries
 } from "./chunk-BOD4O7OF.js";
@@ -152,7 +159,7 @@ import {
   defaultIsMaterial,
   diffFindings,
   runSemanticConceptJudge
-} from "./chunk-JU6ZX3CX.js";
+} from "./chunk-SMCACT4Z.js";
 import {
   LockedJsonlAppender,
   Mutex
@@ -177,7 +184,7 @@ import {
   policyEditsFromFindings,
   scorePolicyEditReadiness,
   validatePolicyEdit
-} from "./chunk-IN3SHQML.js";
+} from "./chunk-LMSQ6EFA.js";
 import {
   AnalystRegistry,
   DEFAULT_TRACE_ANALYST_KINDS,
@@ -198,7 +205,7 @@ import {
   runProposeReview,
   runProposeReviewAsControlLoop,
   scoreFromEvals
-} from "./chunk-B2TMQM62.js";
+} from "./chunk-BHCFJGL4.js";
 import {
   allCriticalPassed,
   errorStreakDetector,
@@ -217,10 +224,10 @@ import {
   evaluateReleaseConfidence,
   judgeReplayGate,
   renderReleaseReport
-} from "./chunk-VDGPPGE3.js";
+} from "./chunk-UMEAR2FI.js";
 import {
   runEvalCampaign
-} from "./chunk-6FIAJHCU.js";
+} from "./chunk-6YVAN6R3.js";
 import {
   LlmCallError,
   LlmClient,
@@ -243,7 +250,7 @@ import {
   paretoChart,
   researchReport,
   summaryTable
-} from "./chunk-6GT4NI4V.js";
+} from "./chunk-CLS3374R.js";
 import {
   attributeCounterfactuals,
   runCounterfactual
@@ -292,7 +299,7 @@ import {
   weightedMean,
   wilcoxonSignedRank,
   wilson
-} from "./chunk-HRGUJTER.js";
+} from "./chunk-HYGRFL7C.js";
 import {
   FileSystemTraceStore,
   InMemoryTraceStore,
@@ -323,7 +330,7 @@ import {
   scoreTraceInsightReadiness,
   tokenizeDomainWords,
   traceAnalystOnRunComplete
-} from "./chunk-7RBJANJD.js";
+} from "./chunk-3722PKPB.js";
 import {
   TRACE_ANALYST_ACTOR_DESCRIPTION,
   TRACE_ANALYST_ACTOR_DESCRIPTION_VERSION,
@@ -345,7 +352,7 @@ import {
   runFailureClass,
   runsForScenario,
   toolSpans
-} from "./chunk-UFSG7ACU.js";
+} from "./chunk-JZXGWLK5.js";
 import {
   FAILURE_CLASSES,
   TRACE_SCHEMA_VERSION,
@@ -408,7 +415,7 @@ import {
   parseRunRecordSafe,
   roundTripRunRecord,
   validateRunRecord
-} from "./chunk-2NSLDY4B.js";
+} from "./chunk-J5MUFXEY.js";
 import {
   TraceEmitter,
   llmSpanFromProvider
@@ -426,7 +433,7 @@ import {
   toAgentProfileJson,
   validateAgentProfileCell,
   verifyAgentProfileCell
-} from "./chunk-ABOIVNXL.js";
+} from "./chunk-AGYDMORK.js";
 import {
   canonicalize,
   evaluateHypothesis,
@@ -5490,7 +5497,7 @@ function matchText(actual, matcher) {
 }
 function resolveToolName(span) {
   if (typeof span.toolName === "string") return span.toolName;
-  const fromAttr = span.attributes?.["tool.name"] ?? span.attributes?.["toolName"];
+  const fromAttr = span.attributes?.["tool.name"] ?? span.attributes?.toolName;
   if (typeof fromAttr === "string") return fromAttr;
   if (span.kind === "tool" && typeof span.name === "string") return span.name;
   return void 0;
@@ -10144,6 +10151,7 @@ export {
   assertLlmRoute,
   assertModelsServed,
   assertNoHiddenLeak,
+  assertProductBenchmarkRun,
   assertRealBackend,
   assertRecordIntegrity,
   assertReleaseConfidence,
@@ -10169,6 +10177,7 @@ export {
   buildAgreementJudge,
   buildDefaultAnalystRegistry,
   buildDriverSystemPrompt,
+  buildProductBenchmarkManifest,
   buildReflectionPrompt,
   buildReviewerPrompt,
   buildTraceAnalystTools,
@@ -10283,6 +10292,8 @@ export {
   expandMatrix,
   expandProfileAxes,
   expectAgent,
+  exportProductBenchmark,
+  exportProductBenchmarkRuns,
   exportRewardModel,
   exportRunAsOtlp,
   extractAssetUrls,
@@ -10437,6 +10448,8 @@ export {
   printDriverSummary,
   probeLlm,
   productBenchmarkIntegrityFailures,
+  productBenchmarkMutableSurfaces,
+  productBenchmarkRepoIdentity,
   productBenchmarkSplits,
   profile_exports as profile,
   projectOtlpFlatLine,
@@ -10498,6 +10511,7 @@ export {
   runLiveProof,
   runProposeReview,
   runProposeReviewAsControlLoop,
+  runRecordToProductBenchmarkRecord,
   runReferenceReplay,
   runScore,
   runSelfPlay,