@tangle-network/agent-eval 0.209.1 → 0.211.0

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 (109) hide show
  1. package/CHANGELOG.md +20 -0
  2. package/README.md +1 -1
  3. package/dist/adapters/http.d.ts +1 -1
  4. package/dist/analyst/index.d.ts +2 -2
  5. package/dist/analyst/index.js +1 -1
  6. package/dist/{analyze-runs-0cdYMgob.js → analyze-runs-3xpby3p2.js} +4 -4
  7. package/dist/analyze-runs-3xpby3p2.js.map +1 -0
  8. package/dist/{benchmark-command-Cve7mHNC.js → benchmark-command-S8tCtkxh.js} +3 -3
  9. package/dist/{benchmark-command-Cve7mHNC.js.map → benchmark-command-S8tCtkxh.js.map} +1 -1
  10. package/dist/benchmarks/index.d.ts +1 -1
  11. package/dist/benchmarks/index.js +2 -2
  12. package/dist/campaign/index.d.ts +2 -2
  13. package/dist/campaign/index.js +9 -9
  14. package/dist/{campaign-evidence-B5qsCHEn.js → campaign-evidence-OcUciD-N.js} +2 -2
  15. package/dist/campaign-evidence-OcUciD-N.js.map +1 -0
  16. package/dist/{campaign-CvhmiLKa.js → campaign-wvvJA1ps.js} +10 -10
  17. package/dist/{campaign-CvhmiLKa.js.map → campaign-wvvJA1ps.js.map} +1 -1
  18. package/dist/claude-jsonl-dZir12m6.js +86 -0
  19. package/dist/claude-jsonl-dZir12m6.js.map +1 -0
  20. package/dist/cli.js +4 -4
  21. package/dist/contract/index.d.ts +5 -5
  22. package/dist/contract/index.js +90 -14
  23. package/dist/contract/index.js.map +1 -1
  24. package/dist/counterfactual-DtMWvn7D.js.map +1 -1
  25. package/dist/{define-agent-eval-_nyV5nev.d.ts → define-agent-eval-B_zZY2AZ.d.ts} +3 -3
  26. package/dist/{define-agent-eval-_nyV5nev.d.ts.map → define-agent-eval-B_zZY2AZ.d.ts.map} +1 -1
  27. package/dist/{define-agent-eval-Csd47ur_.js → define-agent-eval-CoXXDCMu.js} +4 -4
  28. package/dist/{define-agent-eval-Csd47ur_.js.map → define-agent-eval-CoXXDCMu.js.map} +1 -1
  29. package/dist/diagnosis/index.js +1 -1
  30. package/dist/{eval-campaign-C31Qbv77.js → eval-campaign-4byTtqrf.js} +2 -2
  31. package/dist/{eval-campaign-C31Qbv77.js.map → eval-campaign-4byTtqrf.js.map} +1 -1
  32. package/dist/experiment/index.d.ts +1 -1
  33. package/dist/experiment/index.js +6 -6
  34. package/dist/feedback-trajectory-CX01700m.d.ts.map +1 -1
  35. package/dist/{index-H-u7f-w3.d.ts → index-D0_l25We.d.ts} +20 -41
  36. package/dist/{index-H-u7f-w3.d.ts.map → index-D0_l25We.d.ts.map} +1 -1
  37. package/dist/{index-BHh9Z9El.d.ts → index-DnZdBxCL.d.ts} +9 -6
  38. package/dist/{index-BHh9Z9El.d.ts.map → index-DnZdBxCL.d.ts.map} +1 -1
  39. package/dist/{index-Ddi5M-TQ.d.ts → index-xnhRdqik.d.ts} +3 -14
  40. package/dist/{index-Ddi5M-TQ.d.ts.map → index-xnhRdqik.d.ts.map} +1 -1
  41. package/dist/index.d.ts +6 -6
  42. package/dist/index.d.ts.map +1 -1
  43. package/dist/index.js +14 -14
  44. package/dist/index.js.map +1 -1
  45. package/dist/judge-gate-B1_O0QIs.d.ts +454 -0
  46. package/dist/judge-gate-B1_O0QIs.d.ts.map +1 -0
  47. package/dist/judge-gate-B1qYnkpi.js +658 -0
  48. package/dist/judge-gate-B1qYnkpi.js.map +1 -0
  49. package/dist/{llm-judge-C0256VT_.js → llm-judge-Tr7MBUnf.js} +6 -6
  50. package/dist/llm-judge-Tr7MBUnf.js.map +1 -0
  51. package/dist/meta-eval/index.d.ts +2 -132
  52. package/dist/meta-eval/index.d.ts.map +1 -1
  53. package/dist/meta-eval/index.js +2 -275
  54. package/dist/meta-eval/index.js.map +1 -1
  55. package/dist/openapi.json +1 -1
  56. package/dist/{paired-arms-B5v-X7No.js → paired-arms-DzKn0TfY.js} +2 -2
  57. package/dist/{paired-arms-B5v-X7No.js.map → paired-arms-DzKn0TfY.js.map} +1 -1
  58. package/dist/{paired-binary-b2L3WKqr.js → paired-binary-Q36bhn2R.js} +13 -9
  59. package/dist/paired-binary-Q36bhn2R.js.map +1 -0
  60. package/dist/{paired-promotion-decision-2deJ7sDX.js → paired-promotion-decision-BN6IQTiu.js} +2 -2
  61. package/dist/{paired-promotion-decision-2deJ7sDX.js.map → paired-promotion-decision-BN6IQTiu.js.map} +1 -1
  62. package/dist/{produced-state-CgB9g0VX.js → produced-state-CKybuR9F.js} +3 -3
  63. package/dist/produced-state-CKybuR9F.js.map +1 -0
  64. package/dist/{promotion-policy-CCmyEH_c.js → promotion-policy-CWW35UWq.js} +3 -3
  65. package/dist/{promotion-policy-CCmyEH_c.js.map → promotion-policy-CWW35UWq.js.map} +1 -1
  66. package/dist/reporting.js +2 -2
  67. package/dist/rl.js +3 -3
  68. package/dist/rollout/index.d.ts +2 -2
  69. package/dist/rollout/index.js +3 -3
  70. package/dist/{rollout-BZqQc0YY.js → rollout-D9bKKBnZ.js} +29 -164
  71. package/dist/rollout-D9bKKBnZ.js.map +1 -0
  72. package/dist/search/index.js +2 -2
  73. package/dist/{search-kernel-Bfe3kzRk.js → search-kernel-BIXWM7YV.js} +3 -3
  74. package/dist/search-kernel-BIXWM7YV.js.map +1 -0
  75. package/dist/{search-summary-DK1bwTAf.js → search-summary-nwrtXxC_.js} +3 -3
  76. package/dist/{search-summary-DK1bwTAf.js.map → search-summary-nwrtXxC_.js.map} +1 -1
  77. package/dist/{sequential-BOzBND4t.js → sequential-D-1PPCmM.js} +2 -2
  78. package/dist/{sequential-BOzBND4t.js.map → sequential-D-1PPCmM.js.map} +1 -1
  79. package/dist/{skillopt-optimization-method-BCojvoP1.js → skillopt-optimization-method-DujPOiwD.js} +4 -4
  80. package/dist/{skillopt-optimization-method-BCojvoP1.js.map → skillopt-optimization-method-DujPOiwD.js.map} +1 -1
  81. package/dist/{statistical-heldout-BP6iEJvR.js → statistical-heldout-1Gbmex7v.js} +2 -2
  82. package/dist/{statistical-heldout-BP6iEJvR.js.map → statistical-heldout-1Gbmex7v.js.map} +1 -1
  83. package/dist/{statistical-heldout-D0rX_IDl.d.ts → statistical-heldout-8RGPIeyo.d.ts} +14 -10
  84. package/dist/statistical-heldout-8RGPIeyo.d.ts.map +1 -0
  85. package/dist/{summary-report-BsaEFEwA.js → summary-report-DtdRucM0.js} +2 -2
  86. package/dist/{summary-report-BsaEFEwA.js.map → summary-report-DtdRucM0.js.map} +1 -1
  87. package/dist/supervisor-run/index.d.ts.map +1 -1
  88. package/dist/supervisor-run/index.js +34 -69
  89. package/dist/supervisor-run/index.js.map +1 -1
  90. package/dist/{tree-CZ78tiPg.js → tree-kc1d80yF.js} +3 -3
  91. package/dist/{tree-CZ78tiPg.js.map → tree-kc1d80yF.js.map} +1 -1
  92. package/docs/concepts.md +14 -0
  93. package/docs/experiment.md +10 -0
  94. package/docs/trace-analysis.md +3 -0
  95. package/package.json +4 -3
  96. package/dist/analyze-runs-0cdYMgob.js.map +0 -1
  97. package/dist/campaign-evidence-B5qsCHEn.js.map +0 -1
  98. package/dist/claude-jsonl-CG3JUbTJ.js +0 -211
  99. package/dist/claude-jsonl-CG3JUbTJ.js.map +0 -1
  100. package/dist/llm-judge-C0256VT_.js.map +0 -1
  101. package/dist/paired-binary-b2L3WKqr.js.map +0 -1
  102. package/dist/produced-state-CgB9g0VX.js.map +0 -1
  103. package/dist/rollout-BZqQc0YY.js.map +0 -1
  104. package/dist/search-kernel-Bfe3kzRk.js.map +0 -1
  105. package/dist/series-convergence-CjO2QdRW.js +0 -43
  106. package/dist/series-convergence-CjO2QdRW.js.map +0 -1
  107. package/dist/series-convergence-hEfKdoi-.d.ts +0 -126
  108. package/dist/series-convergence-hEfKdoi-.d.ts.map +0 -1
  109. package/dist/statistical-heldout-D0rX_IDl.d.ts.map +0 -1
@@ -1,25 +1,25 @@
1
1
  import { s as ValidationError, t as AgentEvalError } from "./errors-Dngq5h35.js";
2
2
  import { a as hashCanonical, i as compareCodeUnits, r as canonicalString } from "./canonical-BsZVgBhW.js";
3
- import { c as agentProfileId, d as harnessAxisOf, i as verifyCompletion, l as agentProfileModelId, s as agentProfileHash, t as extractProducedState } from "./produced-state-CgB9g0VX.js";
3
+ import { c as agentProfileId, d as harnessAxisOf, i as verifyCompletion, l as agentProfileModelId, s as agentProfileHash, t as extractProducedState } from "./produced-state-CKybuR9F.js";
4
4
  import { o as buildAgentProfileCell } from "./agent-profile-cell-BYc0bxsL.js";
5
5
  import { r as pairedBootstrap } from "./paired-tests-jbwPdSz5.js";
6
- import { t as comparePairedArms } from "./paired-arms-B5v-X7No.js";
7
- import { o as pairHoldout } from "./statistical-heldout-BP6iEJvR.js";
6
+ import { t as comparePairedArms } from "./paired-arms-DzKn0TfY.js";
7
+ import { o as pairHoldout } from "./statistical-heldout-1Gbmex7v.js";
8
8
  import { c as validateRunRecord, i as modelHasSnapshot, n as UNKNOWN_MODEL, o as runTaskScore } from "./run-record-DlTZ_lqe.js";
9
9
  import { n as contentHash, t as canonicalJson } from "./verdict-cache-BQEsVy9n.js";
10
- import { at as resolveRunDir, d as combineComparisonCosts, it as planCampaignRun, lt as cellCachePath, rt as runCampaign, u as executeOptimizationMethod, ut as computeManifestHash, v as labelTrustRank } from "./llm-judge-C0256VT_.js";
10
+ import { at as resolveRunDir, d as combineComparisonCosts, it as planCampaignRun, lt as cellCachePath, rt as runCampaign, u as executeOptimizationMethod, ut as computeManifestHash, v as labelTrustRank } from "./llm-judge-Tr7MBUnf.js";
11
11
  import { n as campaignCellToRunRecord, r as projectCampaignCellQuality, t as campaignCellCostProvenance } from "./run-record-ES07wfz9.js";
12
12
  import { t as CostAccountingIncompleteError } from "./cost-ledger-_IbkejCF.js";
13
- import { B as assertCampaignDesign, N as surfaceContentHash, R as assertRealBackend, W as campaignScenarioIdentity, k as assertCodeSurfaceIdentity, z as summarizeBackendIntegrity } from "./campaign-evidence-B5qsCHEn.js";
13
+ import { B as assertCampaignDesign, N as surfaceContentHash, R as assertRealBackend, W as campaignScenarioIdentity, k as assertCodeSurfaceIdentity, z as summarizeBackendIntegrity } from "./campaign-evidence-OcUciD-N.js";
14
14
  import { m as mapConcurrent } from "./journal-DvKCvjty.js";
15
15
  import { S as fsCampaignStorage } from "./external-optimizer-subprocess-C0EsxEbI.js";
16
- import "./search-kernel-Bfe3kzRk.js";
16
+ import "./search-kernel-BIXWM7YV.js";
17
17
  import "./search-ledger-aMTNQn_m.js";
18
- import "./search-summary-DK1bwTAf.js";
18
+ import "./search-summary-nwrtXxC_.js";
19
19
  import { a as scoreAnalystFindings } from "./benchmark-BywPeAUo.js";
20
20
  import "./external-optimizer-process-D2Txd-7D.js";
21
- import "./skillopt-optimization-method-BCojvoP1.js";
22
- import "./sequential-BOzBND4t.js";
21
+ import "./skillopt-optimization-method-DujPOiwD.js";
22
+ import "./sequential-D-1PPCmM.js";
23
23
  import { createHash } from "node:crypto";
24
24
  import { closeSync, constants, existsSync, fstatSync, lstatSync, mkdirSync, mkdtempSync, openSync, readFileSync, readSync, readdirSync, readlinkSync, realpathSync, rmSync, statSync, writeFileSync } from "node:fs";
25
25
  import { z } from "zod";
@@ -3436,4 +3436,4 @@ function resolveWorktreePath(surface, worktreeDir) {
3436
3436
  //#endregion
3437
3437
  export { loadEvalFixtureScenarios as A, FsLabeledScenarioStore as C, neutralizationGate as D, rolloutArgumentDiff as E, analyzeCrossSurfaceInteractions as M, buildTraceAnalystSurfaceDispatch as N, discoverEvalFixtures as O, traceAnalystQualityJudge as P, neutralizeText as S, classifyUngroundedLiterals as T, scoreUserStory as _, autoevalsScorerJudge as a, scopedOptimizationMethod as b, selectDiscriminative as c, finalizeProfileMatrix as d, runProfileMatrixSegment as f, renderScoreboardMarkdown as g, makePlaybackDispatch as h, verifyCodeSurface as i, planEvalFixtureRun as j, loadEvalFixture as k, createRetainedDispatch as l, runProfileMatrix as m, gitWorktreeAdapter as n, phoenixEvaluatorJudge as o, ProfileMatrixError as p, resolveWorktreePath as r, scoreDiscrimination as s, WorktreeAdapterError as t, createProfileMatrixPlan as u, scoreboardSummary as v, LabeledScenarioStoreError as w, sequentialOptimizationMethod as x, userStoryScoreboard as y };
3438
3438
 
3439
- //# sourceMappingURL=campaign-CvhmiLKa.js.map
3439
+ //# sourceMappingURL=campaign-wvvJA1ps.js.map