@tangle-network/agent-eval 0.185.0 → 0.186.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (87) hide show
  1. package/CHANGELOG.md +8 -0
  2. package/dist/adapters/http.d.ts +1 -1
  3. package/dist/analyst/index.d.ts +5 -5
  4. package/dist/analyst/index.js +1 -1
  5. package/dist/{analyze-runs-sTygaDyD.js → analyze-runs-DCY7LMt4.js} +2 -2
  6. package/dist/{analyze-runs-sTygaDyD.js.map → analyze-runs-DCY7LMt4.js.map} +1 -1
  7. package/dist/{analyze-runs-CrYhqXop.d.ts → analyze-runs-DapYZz02.d.ts} +3 -3
  8. package/dist/{analyze-runs-CrYhqXop.d.ts.map → analyze-runs-DapYZz02.d.ts.map} +1 -1
  9. package/dist/{benchmark-command-kvq7yFY4.js → benchmark-command-tBgs-_AM.js} +2 -2
  10. package/dist/{benchmark-command-kvq7yFY4.js.map → benchmark-command-tBgs-_AM.js.map} +1 -1
  11. package/dist/benchmarks/index.d.ts +1 -1
  12. package/dist/benchmarks/index.js +2 -2
  13. package/dist/campaign/index.d.ts +2 -2
  14. package/dist/campaign/index.js +7 -7
  15. package/dist/{campaign-eJgtRs-W.js → campaign-DlxdTOvw.js} +7 -7
  16. package/dist/{campaign-eJgtRs-W.js.map → campaign-DlxdTOvw.js.map} +1 -1
  17. package/dist/{campaign-evidence-BsrCmBSE.js → campaign-evidence-IF1p0oDa.js} +2 -2
  18. package/dist/{campaign-evidence-BsrCmBSE.js.map → campaign-evidence-IF1p0oDa.js.map} +1 -1
  19. package/dist/cli.js +1 -1
  20. package/dist/{client-Bc8LANCG.d.ts → client-DCc7a8ZC.d.ts} +2 -2
  21. package/dist/{client-Bc8LANCG.d.ts.map → client-DCc7a8ZC.d.ts.map} +1 -1
  22. package/dist/contract/index.d.ts +8 -8
  23. package/dist/contract/index.js +8 -8
  24. package/dist/{default-registry-CXv9kHMt.d.ts → default-registry-DMxeb4GD.d.ts} +3 -3
  25. package/dist/{default-registry-CXv9kHMt.d.ts.map → default-registry-DMxeb4GD.d.ts.map} +1 -1
  26. package/dist/{define-agent-eval-D4uumjl3.js → define-agent-eval-C0EqJ9Gz.js} +4 -4
  27. package/dist/{define-agent-eval-D4uumjl3.js.map → define-agent-eval-C0EqJ9Gz.js.map} +1 -1
  28. package/dist/{define-agent-eval-DE67ECH1.d.ts → define-agent-eval-CGBwPPaq.d.ts} +5 -5
  29. package/dist/{define-agent-eval-DE67ECH1.d.ts.map → define-agent-eval-CGBwPPaq.d.ts.map} +1 -1
  30. package/dist/diagnosis/index.d.ts +1 -1
  31. package/dist/diagnosis/index.js +1 -1
  32. package/dist/{engine-cXUK-WHS.d.ts → engine-DNlKhIm0.d.ts} +2 -2
  33. package/dist/{engine-cXUK-WHS.d.ts.map → engine-DNlKhIm0.d.ts.map} +1 -1
  34. package/dist/experiment/index.d.ts +2 -2
  35. package/dist/experiment/index.js +4 -4
  36. package/dist/hosted/index.d.ts +2 -2
  37. package/dist/{index-ww8736Am.d.ts → index-BRsrfSHT.d.ts} +4 -4
  38. package/dist/{index-ww8736Am.d.ts.map → index-BRsrfSHT.d.ts.map} +1 -1
  39. package/dist/{index-C8_iU8KL.d.ts → index-C8jCAOTT.d.ts} +3 -3
  40. package/dist/{index-C8_iU8KL.d.ts.map → index-C8jCAOTT.d.ts.map} +1 -1
  41. package/dist/{index-BQWfAFeH.d.ts → index-YZrbnBSV.d.ts} +8 -8
  42. package/dist/{index-BQWfAFeH.d.ts.map → index-YZrbnBSV.d.ts.map} +1 -1
  43. package/dist/index.d.ts +171 -14
  44. package/dist/index.d.ts.map +1 -1
  45. package/dist/index.js +291 -7
  46. package/dist/index.js.map +1 -1
  47. package/dist/{insight-report-C3Bfl762.d.ts → insight-report-DpOo-3iL.d.ts} +2 -2
  48. package/dist/{insight-report-C3Bfl762.d.ts.map → insight-report-DpOo-3iL.d.ts.map} +1 -1
  49. package/dist/{llm-judge-FWWhEb31.js → llm-judge-BVso5x-u.js} +4 -4
  50. package/dist/{llm-judge-FWWhEb31.js.map → llm-judge-BVso5x-u.js.map} +1 -1
  51. package/dist/openapi.json +1 -1
  52. package/dist/{paired-promotion-decision-ClJh_9nK.js → paired-promotion-decision-6cn3bLc3.js} +9 -3
  53. package/dist/paired-promotion-decision-6cn3bLc3.js.map +1 -0
  54. package/dist/{paired-promotion-decision-DPsMQm-0.d.ts → paired-promotion-decision-DPKz8ZnF.d.ts} +13 -1
  55. package/dist/{paired-promotion-decision-DPsMQm-0.d.ts.map → paired-promotion-decision-DPKz8ZnF.d.ts.map} +1 -1
  56. package/dist/{produced-state-CmB02tr2.js → produced-state-Cn9ECr4v.js} +3 -3
  57. package/dist/{produced-state-CmB02tr2.js.map → produced-state-Cn9ECr4v.js.map} +1 -1
  58. package/dist/{promotion-policy-DdrYWHDh.js → promotion-policy-CfMNUro2.js} +3 -3
  59. package/dist/{promotion-policy-DdrYWHDh.js.map → promotion-policy-CfMNUro2.js.map} +1 -1
  60. package/dist/{release-confidence-DsETiO4f.d.ts → release-confidence-CSp3MsU4.d.ts} +2 -2
  61. package/dist/{release-confidence-DsETiO4f.d.ts.map → release-confidence-CSp3MsU4.d.ts.map} +1 -1
  62. package/dist/reporting.d.ts +3 -3
  63. package/dist/{researcher-CJkH-6Ty.d.ts → researcher-D7E92STT.d.ts} +2 -2
  64. package/dist/{researcher-CJkH-6Ty.d.ts.map → researcher-D7E92STT.d.ts.map} +1 -1
  65. package/dist/rl.d.ts +2 -2
  66. package/dist/rl.js +1 -1
  67. package/dist/{sequential-PgLEqcTJ.js → sequential-lwJOTRON.js} +2 -2
  68. package/dist/{sequential-PgLEqcTJ.js.map → sequential-lwJOTRON.js.map} +1 -1
  69. package/dist/{skillopt-optimization-method-FSk8b5FQ.js → skillopt-optimization-method-Cc9SVLqi.js} +4 -4
  70. package/dist/{skillopt-optimization-method-FSk8b5FQ.js.map → skillopt-optimization-method-Cc9SVLqi.js.map} +1 -1
  71. package/dist/{statistical-heldout-Bu4oNOJh.d.ts → statistical-heldout-DTmhTW8P.d.ts} +2 -2
  72. package/dist/{statistical-heldout-Bu4oNOJh.d.ts.map → statistical-heldout-DTmhTW8P.d.ts.map} +1 -1
  73. package/dist/{statistical-heldout-CeWT_W7y.js → statistical-heldout-GcLIPxk4.js} +2 -2
  74. package/dist/{statistical-heldout-CeWT_W7y.js.map → statistical-heldout-GcLIPxk4.js.map} +1 -1
  75. package/dist/{store-tool-spans-BrdfPaVh.d.ts → store-tool-spans-B3bX9yqn.d.ts} +3 -3
  76. package/dist/{store-tool-spans-BrdfPaVh.d.ts.map → store-tool-spans-B3bX9yqn.d.ts.map} +1 -1
  77. package/dist/{summary-report-BzFS5UhZ.d.ts → summary-report-Iy6kXo8x.d.ts} +2 -2
  78. package/dist/{summary-report-BzFS5UhZ.d.ts.map → summary-report-Iy6kXo8x.d.ts.map} +1 -1
  79. package/dist/{tool-groups-eOOgYGaM.d.ts → tool-groups-veQYPiTy.d.ts} +2 -2
  80. package/dist/tool-groups-veQYPiTy.d.ts.map +1 -0
  81. package/dist/trace-repair/index.d.ts +1 -1
  82. package/dist/traces.d.ts +2 -2
  83. package/docs/design/statistics-decisions.md +6 -0
  84. package/docs/public-api.md +2 -0
  85. package/package.json +1 -1
  86. package/dist/paired-promotion-decision-ClJh_9nK.js.map +0 -1
  87. package/dist/tool-groups-eOOgYGaM.d.ts.map +0 -1
@@ -1,7 +1,7 @@
1
1
  import { o as FailureClass } from "./schema-CR5cpjQ3.js";
2
2
  import { l as RunTerminalOutcome } from "./run-record-D9wS35fK.js";
3
3
  import { r as ContinuousAgreement } from "./judge-calibration-DFtEMlde.js";
4
- import { i as ParetoFigureSpec, t as GainDistributionBin } from "./summary-report-BzFS5UhZ.js";
4
+ import { i as ParetoFigureSpec, t as GainDistributionBin } from "./summary-report-Iy6kXo8x.js";
5
5
  //#region src/contract/insight-report.d.ts
6
6
  interface InsightReport {
7
7
  /** Number of runs analyzed. */
@@ -407,4 +407,4 @@ interface Recommendation {
407
407
  }
408
408
  //#endregion
409
409
  export { FailureClusterInsight as a, JudgeInsight as c, Recommendation as d, ReleaseSummary as f, FailureClassTally as i, LiftInsight as l, TokenUsageInsight as m, ExecutionErrorOutcomeCell as n, InsightReport as o, ScalarDistribution as p, ExecutionInsight as r, InterRaterInsight as s, CostProvenanceSummary as t, OutcomeCorrelationInsight as u };
410
- //# sourceMappingURL=insight-report-C3Bfl762.d.ts.map
410
+ //# sourceMappingURL=insight-report-DpOo-3iL.d.ts.map
@@ -1 +1 @@
1
- {"version":3,"file":"insight-report-C3Bfl762.d.ts","names":[],"sources":["../src/contract/insight-report.ts"],"mappings":";;;;;UAqCiB;;EAEf;;;;EAKA,WAAW;;EAGX,WAAW;;;EAIX,cAAc,eAAe;;EAG7B;IACE,MAAM;IACN,QAAQ;;;;IAIR,aAAa;;;;;IAKb;MAAa;MAAe;;;;;;EAM9B,QAAQ,eAAe;;;;EAKvB,aAAa;;;;EAKb,OAAO;;;EAIP,kBAAkB;;;EAIlB,gBAAgB;;;;;EAMhB,qBAAqB;;;;;EAMrB,SAAS;;;;;;EAOT,wBAAwB;;;;EAKxB,iBAAiB;;;EAIjB,iBAAiB;;UAGF;EACf;IAAY;IAAW;;EACvB;IAAa;IAAW;;EACxB;IAAc;;EACd;;UAGe;;EAEf,YAAY;;EAEZ,SAAS;;;EAGT,YAAY;;;EAGZ;IACE;IACA,YAAY;IACZ,SAAS;IACT;;;EAGF,QAAQ;IAAQ;IAAe;;;;;EAI/B;IACE;IACA;IACA;;;;;EAKF;IACE;;;IAGA;;IAEA;;IAEA;;IAEA;;IAEA;;;;;IAKA,mBAAmB,OAAO,oBAAoB;;;;EAIhD;IACE;IACA;IACA;IACA;IACA;;;UAIa;;EAEf;;EAEA;;EAEA;;UAGe;EACf,OAAO;EACP,QAAQ;EACR,WAAW;EACX,QAAQ;EACR,YAAY;EACZ;IACE;IACA;IACA;IACA;IACA;;;;UAOa;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA,WAAW;;;;;EAKX,WAAW;IAAQ;IAAe;;;UAGnB;;EAEf;;EAEA;;;EAGA,cAAc;;;EAGd;;;EAGA;;;EAGA;;UAGe;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA,SAAS;;EAET,mBAAmB;IACjB;IACA,SAAS;MAAQ;MAAe;;IAChC;;;UAIa;EACf;EACA;;EAEA;;EAEA;;;EAGA;;EAEA;;EAEA;;EAEA;;EAEA;;;;;;;;;;;EAWA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;UAGe;;EAEf,UAAU;IACR;IACA;;IAEA;;IAEA;;IAEA;;EAEF;;;;;UAMe;;EAEf,cAAc;;EAEd;;EAEA;;UAGe;;EAEf;;;EAGA;EACA,UAAU;IAAQ;IAAe;IAAgB;;;UAGlC;;;EAGf;;EAEA;;EAEA;;EAEA;;EAEA;IACE;IACA;IACA;;;UAIa;;;EAGf;EACA,MAAM;IACJ;IACA;IACA;;;;EAIF;;UAGQ;;EAER;;EAEA;;;;EAIA;;EAEA;EACA;;KAGU,eACP;EACC;;EAEA;;EAEA;;EAEA;;EAEA;MAED;EACC;;EAEA;EACA;EACA;EACA;;UAGW;;EAEf;EACA;;EAEA;;;;EAIA,SAAS,eAAe;;;EAGxB;;EAEA;;EAEA;;UAGe;EACf;EACA;EACA;EACA;;EAEA"}
1
+ {"version":3,"file":"insight-report-DpOo-3iL.d.ts","names":[],"sources":["../src/contract/insight-report.ts"],"mappings":";;;;;UAqCiB;;EAEf;;;;EAKA,WAAW;;EAGX,WAAW;;;EAIX,cAAc,eAAe;;EAG7B;IACE,MAAM;IACN,QAAQ;;;;IAIR,aAAa;;;;;IAKb;MAAa;MAAe;;;;;;EAM9B,QAAQ,eAAe;;;;EAKvB,aAAa;;;;EAKb,OAAO;;;EAIP,kBAAkB;;;EAIlB,gBAAgB;;;;;EAMhB,qBAAqB;;;;;EAMrB,SAAS;;;;;;EAOT,wBAAwB;;;;EAKxB,iBAAiB;;;EAIjB,iBAAiB;;UAGF;EACf;IAAY;IAAW;;EACvB;IAAa;IAAW;;EACxB;IAAc;;EACd;;UAGe;;EAEf,YAAY;;EAEZ,SAAS;;;EAGT,YAAY;;;EAGZ;IACE;IACA,YAAY;IACZ,SAAS;IACT;;;EAGF,QAAQ;IAAQ;IAAe;;;;;EAI/B;IACE;IACA;IACA;;;;;EAKF;IACE;;;IAGA;;IAEA;;IAEA;;IAEA;;IAEA;;;;;IAKA,mBAAmB,OAAO,oBAAoB;;;;EAIhD;IACE;IACA;IACA;IACA;IACA;;;UAIa;;EAEf;;EAEA;;EAEA;;UAGe;EACf,OAAO;EACP,QAAQ;EACR,WAAW;EACX,QAAQ;EACR,YAAY;EACZ;IACE;IACA;IACA;IACA;IACA;;;;UAOa;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA,WAAW;;;;;EAKX,WAAW;IAAQ;IAAe;;;UAGnB;;EAEf;;EAEA;;;EAGA,cAAc;;;EAGd;;;EAGA;;;EAGA;;UAGe;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA,SAAS;;EAET,mBAAmB;IACjB;IACA,SAAS;MAAQ;MAAe;;IAChC;;;UAIa;EACf;EACA;;EAEA;;EAEA;;;EAGA;;EAEA;;EAEA;;EAEA;;EAEA;;;;;;;;;;;EAWA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;UAGe;;EAEf,UAAU;IACR;IACA;;IAEA;;IAEA;;IAEA;;EAEF;;;;;UAMe;;EAEf,cAAc;;EAEd;;EAEA;;UAGe;;EAEf;;;EAGA;EACA,UAAU;IAAQ;IAAe;IAAgB;;;UAGlC;;;EAGf;;EAEA;;EAEA;;EAEA;;EAEA;IACE;IACA;IACA;;;UAIa;;;EAGf;EACA,MAAM;IACJ;IACA;IACA;;;;EAIF;;UAGQ;;EAER;;EAEA;;;;EAIA;;EAEA;EACA;;KAGU,eACP;EACC;;EAEA;;EAEA;;EAEA;;EAEA;MAED;EACC;;EAEA;EACA;EACA;EACA;;UAGW;;EAEf;EACA;;EAEA;;;;EAIA,SAAS,eAAe;;;EAGxB;;EAEA;;EAEA;;UAGe;EACf;EACA;EACA;EACA;;EAEA"}
@@ -1,13 +1,13 @@
1
1
  import { i as JudgeError, s as ValidationError } from "./errors-Dngq5h35.js";
2
2
  import { a as hashCanonical, i as compareCodeUnits, r as canonicalString, t as LEDGER_HASH_PATTERN } from "./canonical-BsZVgBhW.js";
3
3
  import { o as summarizeNumberSeries, s as weightedComposite, t as confidenceInterval } from "./descriptive-1V17A-qa.js";
4
- import { t as decidePairedPromotion } from "./paired-promotion-decision-ClJh_9nK.js";
5
- import { a as heldoutSignificance, i as dimensionRegressions, o as pairHoldout } from "./statistical-heldout-CeWT_W7y.js";
4
+ import { t as decidePairedPromotion } from "./paired-promotion-decision-6cn3bLc3.js";
5
+ import { a as heldoutSignificance, i as dimensionRegressions, o as pairHoldout } from "./statistical-heldout-GcLIPxk4.js";
6
6
  import { i as modelHasSnapshot } from "./run-record-tg_5H2IK.js";
7
7
  import { n as contentHash } from "./verdict-cache-BQEsVy9n.js";
8
8
  import { o as projectCampaignCellQuality, t as campaignCellCostProvenance } from "./run-record-CNNzWGpB.js";
9
9
  import { i as CostLedger, t as CostAccountingIncompleteError } from "./cost-ledger-_IbkejCF.js";
10
- import { $ as formatCoverageFailures, A as FinalEvidenceConflictError, B as surfaceDispatchRef, C as campaignMeanCompositeOrNull, H as surfaceHashMatches, J as assertCompleteCampaign, K as assertCampaignDesign, N as defineEvaluationClaim, P as summarizeEvaluationUnits, R as renderSurfaceDiff, S as campaignMeanComposite, U as BackendIntegrityError, V as surfaceHash, X as campaignScenarioIdentity, Y as campaignCoverage, Z as campaignSplitDigest, b as assertFiniteRankKey, j as FinalEvidenceError, q as assertCampaignSplitIdentity, t as createCampaignEvidenceReceipt, w as compareRankKeys, x as campaignBreakdown, z as surfaceContentHash } from "./campaign-evidence-BsrCmBSE.js";
10
+ import { $ as formatCoverageFailures, A as FinalEvidenceConflictError, B as surfaceDispatchRef, C as campaignMeanCompositeOrNull, H as surfaceHashMatches, J as assertCompleteCampaign, K as assertCampaignDesign, N as defineEvaluationClaim, P as summarizeEvaluationUnits, R as renderSurfaceDiff, S as campaignMeanComposite, U as BackendIntegrityError, V as surfaceHash, X as campaignScenarioIdentity, Y as campaignCoverage, Z as campaignSplitDigest, b as assertFiniteRankKey, j as FinalEvidenceError, q as assertCampaignSplitIdentity, t as createCampaignEvidenceReceipt, w as compareRankKeys, x as campaignBreakdown, z as surfaceContentHash } from "./campaign-evidence-IF1p0oDa.js";
11
11
  import { d as mapConcurrent, n as replayLedgerText, t as FileLedgerJournal } from "./journal-BuRLm6ho.js";
12
12
  import "./ledger-core/index.js";
13
13
  import { D as SearchLedgerIntegrityError, E as SearchLedgerError, S as fsCampaignStorage, T as SearchLedgerConflictError, g as isRecord, h as isExternalTextCandidate, w as SEARCH_LEDGER_FILE_CONTEXT, x as createRunCostLedger } from "./external-optimizer-subprocess-DMgmqTB6.js";
@@ -6156,4 +6156,4 @@ function firstString(value) {
6156
6156
  //#endregion
6157
6157
  export { tangleTracesRoot as $, openSearchLedger as A, DEFAULT_RED_TEAM_CORPUS as B, assertSearchHistoryAdmissionOptions as C, verifySearchHistoryArtifact as D, searchHistoryCoverageRow as E, paretoFrontierWithCrowding as F, assertIndependentEvaluationSplit as G, redTeamReport as H, runFinalComparison as I, reserveFinalEvidence as J, captureFinalEvidencePolicy as K, openAutoPr as L, validateSearchLedgerEvent as M, dominates as N, verifySearchHistoryReceipt as O, paretoFrontier as P, resolveRunDir as Q, captureJudge as R, assertCompleteSearchHistory as S, createSearchHistoryReceipt as T, scoreRedTeamOutput as U, redTeamDataset as V, runCanaries as W, runCampaign as X, runEval as Y, planCampaignRun as Z, isProposedCandidate as _, transientDispatchFailure as a, recordCandidatePopulationSearch as b, compareOptimizationMethods as c, combineComparisonCosts as d, readCachedCell as et, costFromLedgerSummary as f, runOptimization as g, runImprovementLoop as h, quotaExhaustedUntil as i, replaySearchLedgerText as j, FileSearchLedger as k, optimizationTokenUsageFromSummary as l, readGepaCandidatePopulationArtifact as m, JudgeParseError as n, cellCachePath as nt, aggregateRunScore as o, assertGepaCandidatePopulationSummary as p, evaluationUnitMap as q, isTransientTransportFailure as r, computeManifestHash as rt, clamp01 as s, llmJudge as t, buildCellSchedule as tt, executeOptimizationMethod as u, labelTrustRank as v, assertSearchHistoryMatchesReplay as w, SearchHistoryRequiredError as x, SearchRecorder as y, defaultProductionGate as z };
6158
6158
 
6159
- //# sourceMappingURL=llm-judge-FWWhEb31.js.map
6159
+ //# sourceMappingURL=llm-judge-BVso5x-u.js.map