@tangle-network/agent-eval 0.213.0 → 0.214.0

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 (97) hide show
  1. package/CHANGELOG.md +5 -0
  2. package/dist/adapters/http.d.ts +2 -2
  3. package/dist/analyst/index.d.ts +6 -6
  4. package/dist/analyst/index.d.ts.map +1 -1
  5. package/dist/analyst/index.js +3 -3
  6. package/dist/{benchmark-command-IMtRf9eK.js → benchmark-command-CMmeV-_Z.js} +7 -6
  7. package/dist/benchmark-command-CMmeV-_Z.js.map +1 -0
  8. package/dist/benchmarks/index.d.ts +4 -4
  9. package/dist/benchmarks/index.js +3 -3
  10. package/dist/campaign/index.d.ts +4 -4
  11. package/dist/campaign/index.js +8 -8
  12. package/dist/{campaign-CrMj5nYM.js → campaign-Cuw3DQKF.js} +9 -9
  13. package/dist/{campaign-CrMj5nYM.js.map → campaign-Cuw3DQKF.js.map} +1 -1
  14. package/dist/cli.js +6 -6
  15. package/dist/contract/index.d.ts +7 -7
  16. package/dist/contract/index.js +4 -4
  17. package/dist/{default-registry-Crk3QEy1.d.ts → default-registry-BAiTePcB.d.ts} +3 -3
  18. package/dist/{default-registry-Crk3QEy1.d.ts.map → default-registry-BAiTePcB.d.ts.map} +1 -1
  19. package/dist/{define-agent-eval-B_zZY2AZ.d.ts → define-agent-eval-CZdBPUkk.d.ts} +5 -5
  20. package/dist/{define-agent-eval-B_zZY2AZ.d.ts.map → define-agent-eval-CZdBPUkk.d.ts.map} +1 -1
  21. package/dist/{define-agent-eval-BmF0c78X.js → define-agent-eval-FGtpqnh4.js} +4 -4
  22. package/dist/{define-agent-eval-BmF0c78X.js.map → define-agent-eval-FGtpqnh4.js.map} +1 -1
  23. package/dist/{dspy-rlm-engine-D7BIsKyZ.js → dspy-rlm-engine-OMBXVN0u.js} +2 -2
  24. package/dist/{dspy-rlm-engine-D7BIsKyZ.js.map → dspy-rlm-engine-OMBXVN0u.js.map} +1 -1
  25. package/dist/{engine-D-YZ32ZJ.d.ts → engine-BzOR5Ygn.d.ts} +2 -2
  26. package/dist/{engine-D-YZ32ZJ.d.ts.map → engine-BzOR5Ygn.d.ts.map} +1 -1
  27. package/dist/evaluation.d.ts +1 -1
  28. package/dist/experiment/index.d.ts +1 -1
  29. package/dist/{external-optimizer-process-CB3RGyr7.js → external-optimizer-process-DCZx5PdG.js} +2 -2
  30. package/dist/{external-optimizer-process-CB3RGyr7.js.map → external-optimizer-process-DCZx5PdG.js.map} +1 -1
  31. package/dist/{external-optimizer-subprocess-DitxECD3.js → external-optimizer-subprocess-CA3aXT8I.js} +2 -2
  32. package/dist/{external-optimizer-subprocess-DitxECD3.js.map → external-optimizer-subprocess-CA3aXT8I.js.map} +1 -1
  33. package/dist/hosted/index.js +1 -1
  34. package/dist/{index-DnZdBxCL.d.ts → index-BTwxP7L0.d.ts} +6 -6
  35. package/dist/{index-DnZdBxCL.d.ts.map → index-BTwxP7L0.d.ts.map} +1 -1
  36. package/dist/{index-UxNTvn-w.d.ts → index-BWiAAWrk.d.ts} +3 -3
  37. package/dist/{index-UxNTvn-w.d.ts.map → index-BWiAAWrk.d.ts.map} +1 -1
  38. package/dist/{index-Bj96W0jZ.d.ts → index-DHKEAv3s.d.ts} +44 -2
  39. package/dist/index-DHKEAv3s.d.ts.map +1 -0
  40. package/dist/{index-xnhRdqik.d.ts → index-D_rfXep9.d.ts} +4 -4
  41. package/dist/{index-xnhRdqik.d.ts.map → index-D_rfXep9.d.ts.map} +1 -1
  42. package/dist/index.d.ts +9 -9
  43. package/dist/index.js +6 -6
  44. package/dist/ledger-core/index.d.ts +2 -2
  45. package/dist/ledger-core/index.js +2 -1
  46. package/dist/ledger-core-B2hahzrt.js +248 -0
  47. package/dist/ledger-core-B2hahzrt.js.map +1 -0
  48. package/dist/{llm-judge-BtMxSe1e.js → llm-judge-Dw0rDpve.js} +5 -5
  49. package/dist/{llm-judge-BtMxSe1e.js.map → llm-judge-Dw0rDpve.js.map} +1 -1
  50. package/dist/{matrix-jHTsHEkR.d.ts → matrix-CQwid3kj.d.ts} +2 -2
  51. package/dist/{matrix-jHTsHEkR.d.ts.map → matrix-CQwid3kj.d.ts.map} +1 -1
  52. package/dist/multishot/golden/index.d.ts +1 -1
  53. package/dist/multishot/index.d.ts +2 -2
  54. package/dist/openapi.json +1 -1
  55. package/dist/{produced-state-CYhim2We.js → produced-state-C1cTYK6w.js} +2 -2
  56. package/dist/{produced-state-CYhim2We.js.map → produced-state-C1cTYK6w.js.map} +1 -1
  57. package/dist/rl.d.ts +2 -2
  58. package/dist/search/index.d.ts +2 -2
  59. package/dist/search/index.js +2 -2
  60. package/dist/{search-kernel-ah2GESe3.d.ts → search-kernel-BWp7hfJn.d.ts} +3 -3
  61. package/dist/{search-kernel-ah2GESe3.d.ts.map → search-kernel-BWp7hfJn.d.ts.map} +1 -1
  62. package/dist/{search-kernel-BIXWM7YV.js → search-kernel-BztlERLv.js} +3 -3
  63. package/dist/{search-kernel-BIXWM7YV.js.map → search-kernel-BztlERLv.js.map} +1 -1
  64. package/dist/{search-ledger-aMTNQn_m.js → search-ledger-d7V8o9Lk.js} +3 -3
  65. package/dist/{search-ledger-aMTNQn_m.js.map → search-ledger-d7V8o9Lk.js.map} +1 -1
  66. package/dist/{search-ledger-file-DY0AXX3W.js → search-ledger-file-BqOo-bVo.js} +2 -2
  67. package/dist/{search-ledger-file-DY0AXX3W.js.map → search-ledger-file-BqOo-bVo.js.map} +1 -1
  68. package/dist/{search-shipper-DKXWuLnp.js → search-shipper-er4pK5_Y.js} +3 -3
  69. package/dist/{search-shipper-DKXWuLnp.js.map → search-shipper-er4pK5_Y.js.map} +1 -1
  70. package/dist/{search-summary-nwrtXxC_.js → search-summary-CBvz3u_S.js} +3 -3
  71. package/dist/{search-summary-nwrtXxC_.js.map → search-summary-CBvz3u_S.js.map} +1 -1
  72. package/dist/{semantic-concept-judge-B4zwq9Uy.js → semantic-concept-judge-sDBNg_mI.js} +2 -2
  73. package/dist/{semantic-concept-judge-B4zwq9Uy.js.map → semantic-concept-judge-sDBNg_mI.js.map} +1 -1
  74. package/dist/{skillopt-optimization-method-Ds4Ck2Dn.js → skillopt-optimization-method-CvaFGm95.js} +5 -5
  75. package/dist/{skillopt-optimization-method-Ds4Ck2Dn.js.map → skillopt-optimization-method-CvaFGm95.js.map} +1 -1
  76. package/dist/{statistical-heldout-8RGPIeyo.d.ts → statistical-heldout-DZaTo3gZ.d.ts} +2 -2
  77. package/dist/{statistical-heldout-8RGPIeyo.d.ts.map → statistical-heldout-DZaTo3gZ.d.ts.map} +1 -1
  78. package/dist/{store-tool-spans-wQ9idsnn.d.ts → store-tool-spans-DuvZ4YWb.d.ts} +3 -3
  79. package/dist/{store-tool-spans-wQ9idsnn.d.ts.map → store-tool-spans-DuvZ4YWb.d.ts.map} +1 -1
  80. package/dist/systemone.d.ts +1 -1
  81. package/dist/{tool-groups-v28A-ZAF.d.ts → tool-groups-DOriCbkR.d.ts} +2 -2
  82. package/dist/tool-groups-DOriCbkR.d.ts.map +1 -0
  83. package/dist/{trace-questions-C26UvI3D.d.ts → trace-questions-RCaYLkwH.d.ts} +2 -2
  84. package/dist/{trace-questions-C26UvI3D.d.ts.map → trace-questions-RCaYLkwH.d.ts.map} +1 -1
  85. package/dist/trace-repair/index.d.ts +1 -1
  86. package/dist/traces.d.ts +2 -2
  87. package/dist/{tree-kc1d80yF.js → tree-rXLW3M_4.js} +3 -3
  88. package/dist/{tree-kc1d80yF.js.map → tree-rXLW3M_4.js.map} +1 -1
  89. package/dist/{types-Dv9sYwBB.d.ts → types-CMnPcbW5.d.ts} +2 -2
  90. package/dist/{types-Dv9sYwBB.d.ts.map → types-CMnPcbW5.d.ts.map} +1 -1
  91. package/dist/{types-tWFY6bON.d.ts → types-Dj0Qv2qi.d.ts} +2 -2
  92. package/dist/{types-tWFY6bON.d.ts.map → types-Dj0Qv2qi.d.ts.map} +1 -1
  93. package/docs/verified-labels-flywheel.md +2 -2
  94. package/package.json +1 -1
  95. package/dist/benchmark-command-IMtRf9eK.js.map +0 -1
  96. package/dist/index-Bj96W0jZ.d.ts.map +0 -1
  97. package/dist/tool-groups-v28A-ZAF.d.ts.map +0 -1
@@ -4,7 +4,7 @@ import { _ as ProposalFinding } from "./types-KEkKkETv.js";
4
4
  import { M as LlmCallMetadata } from "./types-Lub7o6bN.js";
5
5
  import { t as SeriesDistribution } from "./descriptive-B2iPaT9J.js";
6
6
  import { A as SearchLedgerHash, G as SearchSplit, I as SearchNodeStatus, L as SearchOpenedEvent, N as SearchNodeDecision, O as SearchLedgerEntry, P as SearchNodeRef, Q as SearchTaskOutcome, R as SearchOperationKind, U as SearchReservation, Z as SearchTask, b as SearchClosedEvent, d as SearchCellCancelledEvent, i as SearchArtifactRef, l as SearchCandidateSurface, o as SearchAudit, p as SearchCellStage, t as NodeEstimate, w as SearchEdgeRecordedEvent, z as SearchOperationRecordedEvent } from "./search-ledger-types-p2xKfiDJ.js";
7
- import { p as LedgerProjector } from "./index-Bj96W0jZ.js";
7
+ import { w as LedgerProjector } from "./index-DHKEAv3s.js";
8
8
  import { i as RolloutSearchLineage } from "./mint-CDy-brUR.js";
9
9
  //#region src/campaign/search-state.d.ts
10
10
  /** Deterministic node id: identical content in one search is one node. */
@@ -1015,4 +1015,4 @@ interface CampaignResult<TArtifact = unknown, TScenario extends Scenario = Scena
1015
1015
  }
1016
1016
  //#endregion
1017
1017
  export { SearchNode as $, LabeledScenarioWrite as A, ScoredSurfaceOutcome as B, JudgeDimension as C, LabeledScenarioSampleArgs as D, LabeledScenarioRecord as E, ProposeContext as F, labelTrustRank as G, SurfaceProposer as H, ProposedCandidate as I, renderSearchSummary as J, SearchProposerView as K, RedactionStatus as L, OptimizerConfig as M, ParetoParent as N, LabeledScenarioSource as O, ProposalTrackContext as P, SearchDecisionRecord as Q, Scenario as R, JudgeConfig as S, LabelTrust as T, TraceSpan as U, SessionScript as V, isProposedCandidate as W, SearchCell as X, searchProposerView as Y, SearchCompletion as Z, GateDecision as _, CampaignResult as a, SearchUnitScore as at, GenerationRecord as b, CampaignTraceWriter as c, searchTaskSetDigest as ct, DispatchContext as d, SearchOperation as et, DispatchFn as f, GateContribution as g, GateContext as h, CampaignCostMeter as i, SearchStateView as it, MutableSurface as j, LabeledScenarioStore as k, CodeSurface as l, searchUnitScores as lt, GateCheckStatus as m, CampaignArtifactWriter as n, SearchSpend as nt, CampaignScenarioIdentity as o, searchCellId as ot, Gate as p, SearchSummaryOptions as q, CampaignCellResult as r, SearchState as rt, CampaignTokenUsage as s, searchNodeId as st, CampaignAggregates as t, SearchScoredCell as tt, ComponentSurface as u, GateResult as v, JudgeScore as w, JudgeAggregate as x, GenerationCandidate as y, ScenarioAggregate as z };
1018
- //# sourceMappingURL=types-Dv9sYwBB.d.ts.map
1018
+ //# sourceMappingURL=types-CMnPcbW5.d.ts.map
@@ -1 +1 @@
1
- {"version":3,"file":"types-Dv9sYwBB.d.ts","names":[],"sources":["../src/campaign/search-state.ts","../src/campaign/search-summary.ts","../src/campaign/types.ts"],"mappings":";;;;;;;;;;iBAwDgB,aAAa,kBAAkB;;iBAK/B,aACd,kBACA,gBACA,gBACA,OAAO,aACP;;iBAMc,oBAAoB,gBAAgB,eAAe;UAkBlD;EACf;EACA,UAAU;EACV,OAAO;EACP;EACA;;UAGe;EACf;EACA;EACA;EACA;;;;UAKe;EACf;EACA;EACA;EACA;;;UAIe;EACf;EACA;;EAEA;EACA;;;;;;;;iBASc,iBAAiB,gBAAgB,qBAAqB;UAoBrD;EACf;;EAEA;EACA,gBAAgB;EAChB,UAAU;EACV,mBAAmB;;;EAGnB;;EAEA;;;;EAIA,kBAAkB;EAClB;EACA;EACA,QAAQ;EACR;;EAEA;EACA,oBAAoB;EACpB;EACA,OAAO;EACP;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA,OAAO;EACP;EACA,OAAO;EACP;EACA,aAAa;EACb;EACA;EACA,SAAS;;EAET;;EAEA;EACA,WAAW;;EAEX;;EAEA;EACA;EACA;EACA;;UAGe;EACf;EACA,eAAe;EACf,aAAa;EACb;;EAEA,SAAS;;;EAGT,WAAW;;;EAGX,oBAAoB;;EAEpB;;EAEA;;UAGe;EACf;EACA;;;;;;;cA6BW,uBAAuB,gBAAgB,mBAAmB;WAC5D;UACD;UACA;UACA;UACA;UACA;UACA;mBACS;mBACA;mBACA;mBACA;mBACA;mBACA;mBACA;mBACA;UACT;UACA;;mBAES;;UAET;mBACS;EAEjB,YAAY;EAgCZ,MAAM,OAAO,mBAAmB;EA6BhC,YAAY;;EAKZ,KAAK;UAIG;UAgCA;UA6CA;UAsBA;UA8BA;UA8CA;UAsHA;UAmEA;UA4EA;UAoBA;UAmDA;UAaA;;;UA2EA;;EASR;;;UAmBQ;;;UAkBA;;;;;;;;;;;UAoBA;;;EAsBR;IAAgB;IAA4B;;;EAa5C,cAAc;;EAKd,cAAc;;EAKd;IAAc;IAAkB,WAAW;;;EAK3C;;EAKA,aAAa;;EAcb,SAAS,iBAAiB;;EAM1B;;EAKA,YAAY;;EAKZ,oBAAoB;;EAKpB,SAAS,iBAAiB;;EAK1B;;EAKA,SAAS,iBAAiB;;EAM1B,YAAY;;EAMZ,cAAc,sBAAsB;;EAMpC,gBAAgB,gBAAgB,OAAO,cAAc;;;;;;;cAiB1C;WACF;WACA,QAAQ;WACR,QAAQ;WACR;IAAQ;IAAkB,WAAW;;;WAErC;WACA,OAAO;WACP,YAAY;;;;;;;WAOZ;IAAU;IAA4B;;mBAC9B;mBACA;EAEjB,YAAY,OAAO,aAAa;;MAgB5B;EAIJ,KAAK,iBAAiB;EAItB,QAAQ;;EAKR;;EAKA,SAAS;EAKT,gBAAgB;EAIhB,KAAK,iBAAiB;;EAKtB,SAAS;EAKT,KAAK,iBAAiB;;EAKtB,MAAM;IAAU;MAAyB;EAKzC,UAAU,sBAAsB;;;EAMhC,YAAY,gBAAgB,OAAO,cAAc;;EAKjD,WAAW,gBAAgB,OAAO,cAAc;;EAKhD,QAAQ,aAAa,uBAAuB;UAmCpC;;;;;;UCluCO;WACN;WACA;;;EAGT,YAAY,0BAA0B;;EAEtC,WAAW,0BAA0B;;;iBAIvB,mBAAmB,OAAO,kBAAkB;UAW3C;;;;;;;;EAQf,OAAO;;EAEP;;;;;;;;;;;iBAYc,oBAAoB,OAAO,iBAAiB,SAAS;;;;;UCtCpD;EACf;EACA;EACA;;;;;EAKA;;;UAIe,iCAAiC,KAAK;EACrD;;;;;UAMe;EACf;;EAEA;EACA;EACA;EACA;EACA,QAAQ;EACR,OAAO;EACP,WAAW;EACX,MAAM;;EAEN;;EAEA;;;;;;;;EAQA;;;;KAKU,WAAW,kBAAkB,UAAU,cACjD,UAAU,WACV,KAAK,oBACF,QAAQ;;;;UAOI,cAAc,WAAW;EACxC;EACA;EACA;;EAEA;;;EAGA,sBAAsB,UAAU,WAAW,sBAAsB,UAAU,cAAc;;UAK1E;;EAEf;;EAEA;;;;;;;;;UAUe,YAAY,WAAW,kBAAkB,WAAW;EACnE;EACA,YAAY;;;;EAIZ;;;EAGA,MAAM;IACJ,UAAU;IACV,UAAU;IACV,QAAQ;;IAER,aAAa;IACb;IACA,WAAW;MACT,aAAa,QAAQ;EACzB,aAAa,UAAU;;;;;;;;;;;UAYR;EACf,YAAY;EACZ;EACA;;EAEA,UAAU;;;;;;;EAOV;;;;EAIA,eAAe,eAAe;IAAQ;IAAe;;;;;EAIrD;;;EAGA;;EAEA;;EAEA,WAAW,eAAe;;;;;;;UAUX;WACN;;;WAGA;;WAEA;;WAEA;;WAEA;;WAEA;;WAEA;;;WAGA;aACE;aACA;aACA;;;WAGF;;;UAIM;WACN;WACA,YAAY,SAAS;;;;;;;;;;;KAYpB,0BAA0B,mBAAmB;;;;;;;UAQxC;EACf,SAAS;;EAET;;;;EAIA;;;;;;EAMA,cAAc,SAAS;;;;iBAKT,oBACd,OAAO,iBAAiB,oBACvB,SAAS;;;;;;;;;UAkBK;EACf,SAAS;EACT;;;EAGA,YAAY;;;EAGZ;;EAEA;EACA;EACA;;;UAIe;EACf;EACA;EACA;EACA;EACA;;;;;UAMe;;;EAGf;;EAEA;EACA;EACA;EACA,YAAY;;;;;EAKZ,WAAW;IAAQ;IAAoB;IAAmB;IAAgB;;EAC1E;IACE;IACA;;;;;UAMa,eAAe,YAAY;;;;WAIjC,gBAAgB;;;;;WAKhB;;;;WAIA,UAAU;IAAgB;IAAgB,UAAU;;;;WAGpD,QAAQ;;;;WAIR;;;WAGA,SAAS,cAAc;WACvB,UAAU,cAAc;;WAExB;WACA;WACA,QAAQ;;WAER,QAAQ;;;WAGR,kBAAkB;;;WAGlB,mBAAmB;;;;WAInB,gBAAgB;;;;WAIhB;;;;;;;WAOA,gBAAgB,cAAc;;WAE9B,aAAa;WACb;;;;;;;;;;;;;;UAeM,gBAAgB,YAAY;EAC3C;;;;;EAKA,QAAQ,KAAK,eAAe,aAAa,QAAQ,MAAM,iBAAiB;;;EAGxE,QAAQ;IAAQ,SAAS,cAAc;;IAAwB;IAAe;;;UAG/D;EACf;EACA;EACA,mBAAmB,qBAAqB;;UAGzB,wBAAwB;EACvC,UAAU;;;KAMA;;KAGA;UAEK;EACf;EACA,QAAQ;EACR;;UAGe,YAAY,WAAW,kBAAkB;EACxD,oBAAoB,YAAY;EAChC,oBAAoB,YAAY;;EAEhC,aAAa,YAAY,eAAe;;;;;EAKxC,sBAAsB,YAAY,eAAe;;;;;;;EAOjD,yBAAyB,YAAY,eAAe;;;EAGpD,uBAAuB,YAAY;EACnC,WAAW;EACX;IAAQ;IAAmB;;;EAE3B,aAAa;EACb;EACA,QAAQ;;UAGO;EACf,UAAU;EACV;EACA,mBAAmB;EACnB;;;UAIe,KAAK,qBAAqB,kBAAkB,WAAW;EACtE;EACA,OAAO,KAAK,YAAY,WAAW,aAAa,QAAQ;;;;UAOzC;EACf,KAAK,cAAc,aAAa,0BAA0B;EAC1D,SAAS;;UAGM;EACf,IAAI,aAAa;EACjB,aAAa,aAAa;;;;;UAMX;EACf,MAAM,cAAc,kBAAkB,aAAa;EACnD,UAAU,cAAc,iBAAiB;;;;;;KAO/B,qBAAqB;;;;;UAMhB;;EAEf,YAAY,GACV,OAAO,KAAK,iBAAiB;IAC3B,UAAU;MAEX,QAAQ,eAAe;;;;;KAQhB;KAOA;;;;;;;;;;;;;;;KAgBA;;iBASI,eAAe,OAAO;;;;UAOrB,qBAAqB,kBAAkB,WAAW,UAAU;EAC3E,UAAU;EACV,UAAU;EACV,aAAa,eAAe;EAC5B,QAAQ;EACR;EACA;EACA,iBAAiB;;;;;EAKjB,aAAa;;EAEb;;UAGe,sBAAsB,kBAAkB,WAAW,UAAU,6BACpE,qBAAqB,WAAW;;EAExC;;;EAGA;;UAGe;EACf;;EAEA;;;;EAIA;EACA;IACE;IACA,SAAS,wBAAwB;IACjC;IACA;;;;;IAKA,WAAW;;;UAIE;EACf,QAAQ,OAAO,uBAAuB;EACtC,OAAO,MAAM,4BAA4B,QAAQ;EACjD,QAAQ;IACN;IACA;IACA,UAAU;;;IAGV,SAAS,OAAO;;;UAMH,mBAAmB;;;EAGlC;EACA;;EAEA;IAAY;IAAsB;;EAClC;EACA;EACA;EACA,UAAU;EACV,aAAa,eAAe;;;EAG5B;;EAEA,gBAAgB;;EAEhB;;;EAGA,YAAY;;;EAGZ;;;EAGA;EACA;EACA;EACA;;;;EAIA;;EAEA;;EAEA;EACA;;UAGe;EACf;EACA;EACA;EACA;;;;;;EAMA,cAAc;;UAGC;EACf;EACA;EACA;;;EAGA,cAAc;;UAGC;EACf;EACA,YAAY;EACZ;;;;;;UAOe;EACf;;;;EAIA;;;;EAIA;;;;EAIA;IACE;IACA;IACA,iBAAiB;MAAQ;MAAgB;;;;;EAI3C,YAAY;;;;;;;;;;EAUZ,WAAW;IAAQ;IAAoB;IAAmB;IAAgB;;;;EAG1E;;;;EAIA;;;;EAIA,cAAc,SAAS;;UAGR;EACf,SAAS,eAAe;EACxB,YAAY,eAAe;;EAE3B,MAAM;;EAEN;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;;UAGe,eAAe,qBAAqB,kBAAkB,WAAW;;EAEhF;;EAEA;EACA;;EAEA;EACA;EACA;EACA;EACA,OAAO,MAAM,mBAAmB;EAChC,YAAY;EACZ;IACE,aAAa;IACb;;EAEF,OAAO;EACP;EACA;EACA,iBAAiB;;;EAGjB,WAAW,MAAM,2BAA2B,KAAK"}
1
+ {"version":3,"file":"types-CMnPcbW5.d.ts","names":[],"sources":["../src/campaign/search-state.ts","../src/campaign/search-summary.ts","../src/campaign/types.ts"],"mappings":";;;;;;;;;;iBAwDgB,aAAa,kBAAkB;;iBAK/B,aACd,kBACA,gBACA,gBACA,OAAO,aACP;;iBAMc,oBAAoB,gBAAgB,eAAe;UAkBlD;EACf;EACA,UAAU;EACV,OAAO;EACP;EACA;;UAGe;EACf;EACA;EACA;EACA;;;;UAKe;EACf;EACA;EACA;EACA;;;UAIe;EACf;EACA;;EAEA;EACA;;;;;;;;iBASc,iBAAiB,gBAAgB,qBAAqB;UAoBrD;EACf;;EAEA;EACA,gBAAgB;EAChB,UAAU;EACV,mBAAmB;;;EAGnB;;EAEA;;;;EAIA,kBAAkB;EAClB;EACA;EACA,QAAQ;EACR;;EAEA;EACA,oBAAoB;EACpB;EACA,OAAO;EACP;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA,OAAO;EACP;EACA,OAAO;EACP;EACA,aAAa;EACb;EACA;EACA,SAAS;;EAET;;EAEA;EACA,WAAW;;EAEX;;EAEA;EACA;EACA;EACA;;UAGe;EACf;EACA,eAAe;EACf,aAAa;EACb;;EAEA,SAAS;;;EAGT,WAAW;;;EAGX,oBAAoB;;EAEpB;;EAEA;;UAGe;EACf;EACA;;;;;;;cA6BW,uBAAuB,gBAAgB,mBAAmB;WAC5D;UACD;UACA;UACA;UACA;UACA;UACA;mBACS;mBACA;mBACA;mBACA;mBACA;mBACA;mBACA;mBACA;UACT;UACA;;mBAES;;UAET;mBACS;EAEjB,YAAY;EAgCZ,MAAM,OAAO,mBAAmB;EA6BhC,YAAY;;EAKZ,KAAK;UAIG;UAgCA;UA6CA;UAsBA;UA8BA;UA8CA;UAsHA;UAmEA;UA4EA;UAoBA;UAmDA;UAaA;;;UA2EA;;EASR;;;UAmBQ;;;UAkBA;;;;;;;;;;;UAoBA;;;EAsBR;IAAgB;IAA4B;;;EAa5C,cAAc;;EAKd,cAAc;;EAKd;IAAc;IAAkB,WAAW;;;EAK3C;;EAKA,aAAa;;EAcb,SAAS,iBAAiB;;EAM1B;;EAKA,YAAY;;EAKZ,oBAAoB;;EAKpB,SAAS,iBAAiB;;EAK1B;;EAKA,SAAS,iBAAiB;;EAM1B,YAAY;;EAMZ,cAAc,sBAAsB;;EAMpC,gBAAgB,gBAAgB,OAAO,cAAc;;;;;;;cAiB1C;WACF;WACA,QAAQ;WACR,QAAQ;WACR;IAAQ;IAAkB,WAAW;;;WAErC;WACA,OAAO;WACP,YAAY;;;;;;;WAOZ;IAAU;IAA4B;;mBAC9B;mBACA;EAEjB,YAAY,OAAO,aAAa;;MAgB5B;EAIJ,KAAK,iBAAiB;EAItB,QAAQ;;EAKR;;EAKA,SAAS;EAKT,gBAAgB;EAIhB,KAAK,iBAAiB;;EAKtB,SAAS;EAKT,KAAK,iBAAiB;;EAKtB,MAAM;IAAU;MAAyB;EAKzC,UAAU,sBAAsB;;;EAMhC,YAAY,gBAAgB,OAAO,cAAc;;EAKjD,WAAW,gBAAgB,OAAO,cAAc;;EAKhD,QAAQ,aAAa,uBAAuB;UAmCpC;;;;;;UCluCO;WACN;WACA;;;EAGT,YAAY,0BAA0B;;EAEtC,WAAW,0BAA0B;;;iBAIvB,mBAAmB,OAAO,kBAAkB;UAW3C;;;;;;;;EAQf,OAAO;;EAEP;;;;;;;;;;;iBAYc,oBAAoB,OAAO,iBAAiB,SAAS;;;;;UCtCpD;EACf;EACA;EACA;;;;;EAKA;;;UAIe,iCAAiC,KAAK;EACrD;;;;;UAMe;EACf;;EAEA;EACA;EACA;EACA;EACA,QAAQ;EACR,OAAO;EACP,WAAW;EACX,MAAM;;EAEN;;EAEA;;;;;;;;EAQA;;;;KAKU,WAAW,kBAAkB,UAAU,cACjD,UAAU,WACV,KAAK,oBACF,QAAQ;;;;UAOI,cAAc,WAAW;EACxC;EACA;EACA;;EAEA;;;EAGA,sBAAsB,UAAU,WAAW,sBAAsB,UAAU,cAAc;;UAK1E;;EAEf;;EAEA;;;;;;;;;UAUe,YAAY,WAAW,kBAAkB,WAAW;EACnE;EACA,YAAY;;;;EAIZ;;;EAGA,MAAM;IACJ,UAAU;IACV,UAAU;IACV,QAAQ;;IAER,aAAa;IACb;IACA,WAAW;MACT,aAAa,QAAQ;EACzB,aAAa,UAAU;;;;;;;;;;;UAYR;EACf,YAAY;EACZ;EACA;;EAEA,UAAU;;;;;;;EAOV;;;;EAIA,eAAe,eAAe;IAAQ;IAAe;;;;;EAIrD;;;EAGA;;EAEA;;EAEA,WAAW,eAAe;;;;;;;UAUX;WACN;;;WAGA;;WAEA;;WAEA;;WAEA;;WAEA;;WAEA;;;WAGA;aACE;aACA;aACA;;;WAGF;;;UAIM;WACN;WACA,YAAY,SAAS;;;;;;;;;;;KAYpB,0BAA0B,mBAAmB;;;;;;;UAQxC;EACf,SAAS;;EAET;;;;EAIA;;;;;;EAMA,cAAc,SAAS;;;;iBAKT,oBACd,OAAO,iBAAiB,oBACvB,SAAS;;;;;;;;;UAkBK;EACf,SAAS;EACT;;;EAGA,YAAY;;;EAGZ;;EAEA;EACA;EACA;;;UAIe;EACf;EACA;EACA;EACA;EACA;;;;;UAMe;;;EAGf;;EAEA;EACA;EACA;EACA,YAAY;;;;;EAKZ,WAAW;IAAQ;IAAoB;IAAmB;IAAgB;;EAC1E;IACE;IACA;;;;;UAMa,eAAe,YAAY;;;;WAIjC,gBAAgB;;;;;WAKhB;;;;WAIA,UAAU;IAAgB;IAAgB,UAAU;;;;WAGpD,QAAQ;;;;WAIR;;;WAGA,SAAS,cAAc;WACvB,UAAU,cAAc;;WAExB;WACA;WACA,QAAQ;;WAER,QAAQ;;;WAGR,kBAAkB;;;WAGlB,mBAAmB;;;;WAInB,gBAAgB;;;;WAIhB;;;;;;;WAOA,gBAAgB,cAAc;;WAE9B,aAAa;WACb;;;;;;;;;;;;;;UAeM,gBAAgB,YAAY;EAC3C;;;;;EAKA,QAAQ,KAAK,eAAe,aAAa,QAAQ,MAAM,iBAAiB;;;EAGxE,QAAQ;IAAQ,SAAS,cAAc;;IAAwB;IAAe;;;UAG/D;EACf;EACA;EACA,mBAAmB,qBAAqB;;UAGzB,wBAAwB;EACvC,UAAU;;;KAMA;;KAGA;UAEK;EACf;EACA,QAAQ;EACR;;UAGe,YAAY,WAAW,kBAAkB;EACxD,oBAAoB,YAAY;EAChC,oBAAoB,YAAY;;EAEhC,aAAa,YAAY,eAAe;;;;;EAKxC,sBAAsB,YAAY,eAAe;;;;;;;EAOjD,yBAAyB,YAAY,eAAe;;;EAGpD,uBAAuB,YAAY;EACnC,WAAW;EACX;IAAQ;IAAmB;;;EAE3B,aAAa;EACb;EACA,QAAQ;;UAGO;EACf,UAAU;EACV;EACA,mBAAmB;EACnB;;;UAIe,KAAK,qBAAqB,kBAAkB,WAAW;EACtE;EACA,OAAO,KAAK,YAAY,WAAW,aAAa,QAAQ;;;;UAOzC;EACf,KAAK,cAAc,aAAa,0BAA0B;EAC1D,SAAS;;UAGM;EACf,IAAI,aAAa;EACjB,aAAa,aAAa;;;;;UAMX;EACf,MAAM,cAAc,kBAAkB,aAAa;EACnD,UAAU,cAAc,iBAAiB;;;;;;KAO/B,qBAAqB;;;;;UAMhB;;EAEf,YAAY,GACV,OAAO,KAAK,iBAAiB;IAC3B,UAAU;MAEX,QAAQ,eAAe;;;;;KAQhB;KAOA;;;;;;;;;;;;;;;KAgBA;;iBASI,eAAe,OAAO;;;;UAOrB,qBAAqB,kBAAkB,WAAW,UAAU;EAC3E,UAAU;EACV,UAAU;EACV,aAAa,eAAe;EAC5B,QAAQ;EACR;EACA;EACA,iBAAiB;;;;;EAKjB,aAAa;;EAEb;;UAGe,sBAAsB,kBAAkB,WAAW,UAAU,6BACpE,qBAAqB,WAAW;;EAExC;;;EAGA;;UAGe;EACf;;EAEA;;;;EAIA;EACA;IACE;IACA,SAAS,wBAAwB;IACjC;IACA;;;;;IAKA,WAAW;;;UAIE;EACf,QAAQ,OAAO,uBAAuB;EACtC,OAAO,MAAM,4BAA4B,QAAQ;EACjD,QAAQ;IACN;IACA;IACA,UAAU;;;IAGV,SAAS,OAAO;;;UAMH,mBAAmB;;;EAGlC;EACA;;EAEA;IAAY;IAAsB;;EAClC;EACA;EACA;EACA,UAAU;EACV,aAAa,eAAe;;;EAG5B;;EAEA,gBAAgB;;EAEhB;;;EAGA,YAAY;;;EAGZ;;;EAGA;EACA;EACA;EACA;;;;EAIA;;EAEA;;EAEA;EACA;;UAGe;EACf;EACA;EACA;EACA;;;;;;EAMA,cAAc;;UAGC;EACf;EACA;EACA;;;EAGA,cAAc;;UAGC;EACf;EACA,YAAY;EACZ;;;;;;UAOe;EACf;;;;EAIA;;;;EAIA;;;;EAIA;IACE;IACA;IACA,iBAAiB;MAAQ;MAAgB;;;;;EAI3C,YAAY;;;;;;;;;;EAUZ,WAAW;IAAQ;IAAoB;IAAmB;IAAgB;;;;EAG1E;;;;EAIA;;;;EAIA,cAAc,SAAS;;UAGR;EACf,SAAS,eAAe;EACxB,YAAY,eAAe;;EAE3B,MAAM;;EAEN;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;;UAGe,eAAe,qBAAqB,kBAAkB,WAAW;;EAEhF;;EAEA;EACA;;EAEA;EACA;EACA;EACA;EACA,OAAO,MAAM,mBAAmB;EAChC,YAAY;EACZ;IACE,aAAa;IACb;;EAEF,OAAO;EACP;EACA;EACA,iBAAiB;;;EAGjB,WAAW,MAAM,2BAA2B,KAAK"}
@@ -1,5 +1,5 @@
1
1
  import { c as RunSplitTag } from "./run-record-p2AYTrCn.js";
2
- import { R as Scenario, d as DispatchContext } from "./types-Dv9sYwBB.js";
2
+ import { R as Scenario, d as DispatchContext } from "./types-CMnPcbW5.js";
3
3
  //#region src/benchmarks/types.d.ts
4
4
  type BenchmarkTaskKind = 'retrieval' | 'rag-answer' | 'hallucination' | 'kb-improvement' | 'routing' | 'custom';
5
5
  type BenchmarkFamily = 'beir' | 'mteb-retrieval' | 'msmarco' | 'trec-dl' | 'miracl' | 'lotte' | 'bright' | 'crag' | 'hotpotqa' | 'kilt' | 'ragtruth' | 'faithbench' | 'first-party' | 'custom';
@@ -99,4 +99,4 @@ declare const BENCHMARK_SPLIT_SEED = "agent-eval-v1";
99
99
  declare function deterministicSplit(itemId: string, seed?: string): RunSplitTag;
100
100
  //#endregion
101
101
  export { BenchmarkFamily as a, BenchmarkSource as c, BenchmarkEvaluation as i, BenchmarkTaskKind as l, BenchmarkAdapter as n, BenchmarkResponder as o, BenchmarkDatasetItem as r, BenchmarkScenario as s, BENCHMARK_SPLIT_SEED as t, deterministicSplit as u };
102
- //# sourceMappingURL=types-tWFY6bON.d.ts.map
102
+ //# sourceMappingURL=types-Dj0Qv2qi.d.ts.map
@@ -1 +1 @@
1
- {"version":3,"file":"types-tWFY6bON.d.ts","names":[],"sources":["../src/benchmarks/types.ts"],"mappings":";;;KASY;KAQA;UAgBK,qBAAqB;;;EAGpC;;EAEA,SAAS;;EAET,QAAQ;;EAER,SAAS;;EAET,WAAW;;EAEX;;EAEA,SAAS;EACT,WAAW;;UAGI;;;;EAIf;;EAEA;;EAEA,aAAa;;;EAGb,KAAK;EACL;;UAGe;EACf;EACA;EACA;;EAEA;EACA;EACA;;EAEA;;;;;EAKA;;;UAQe,iBAAiB,kBAAkB,oBAAoB;;EAEtE;EACA,SAAS;EACT,WAAW;EACX;EACA,SAAS;EACT;;;;;EAKA,YAAY,OAAO,cAAc,QAAQ,qBAAqB;;EAE9D,SAAS,MAAM,qBAAqB,WAAW,UAAU,YAAY,QAAQ;;;;EAI7E,YAAY,iBAAiB;;UAGd,kBAAkB,4BAA4B;EAC7D;EACA;EACA,QAAQ;EACR,UAAU;EACV,UAAU;EACV,MAAM,qBAAqB;;KAGjB,mBAAmB,oBAAoB,uBAAuB;EACxE,UAAU,kBAAkB;EAC5B,MAAM,qBAAqB;EAC3B,SAAS;MACL,QAAQ,aAAa;;;cAoBd;;;;;;;;;iBAUG,mBACd,gBACA,gBACC"}
1
+ {"version":3,"file":"types-Dj0Qv2qi.d.ts","names":[],"sources":["../src/benchmarks/types.ts"],"mappings":";;;KASY;KAQA;UAgBK,qBAAqB;;;EAGpC;;EAEA,SAAS;;EAET,QAAQ;;EAER,SAAS;;EAET,WAAW;;EAEX;;EAEA,SAAS;EACT,WAAW;;UAGI;;;;EAIf;;EAEA;;EAEA,aAAa;;;EAGb,KAAK;EACL;;UAGe;EACf;EACA;EACA;;EAEA;EACA;EACA;;EAEA;;;;;EAKA;;;UAQe,iBAAiB,kBAAkB,oBAAoB;;EAEtE;EACA,SAAS;EACT,WAAW;EACX;EACA,SAAS;EACT;;;;;EAKA,YAAY,OAAO,cAAc,QAAQ,qBAAqB;;EAE9D,SAAS,MAAM,qBAAqB,WAAW,UAAU,YAAY,QAAQ;;;;EAI7E,YAAY,iBAAiB;;UAGd,kBAAkB,4BAA4B;EAC7D;EACA;EACA,QAAQ;EACR,UAAU;EACV,UAAU;EACV,MAAM,qBAAqB;;KAGjB,mBAAmB,oBAAoB,uBAAuB;EACxE,UAAU,kBAAkB;EAC5B,MAAM,qBAAqB;EAC3B,SAAS;MACL,QAAQ,aAAa;;;cAoBd;;;;;;;;;iBAUG,mBACd,gBACA,gBACC"}
@@ -22,7 +22,7 @@ Derived from what the run2 replay batch actually consumed (its enumeration exclu
22
22
  | Claude Code transcripts `~/.claude/projects` | 474 projects, 7,579 session JSONLs, 4.6 GB | no image; `cwd` + `gitBranch` + harness `version` per message | yes — full tool calls + results | **no** — Bash `toolUseResult` records `stdout`/`stderr`/`interrupted` only, no exit code field | no (host env unpinned, no rc) |
23
23
  | Codex sessions `~/.codex/sessions` | 4,332 rollout files, 113 GB | no image; `session_meta` has `cwd` + git `commit_hash`/`branch`/`repository_url` + `cli_version` | yes | shell events carry exit codes in payloads (format varies by version) | no (host env unpinned) |
24
24
  | traces CLI normalized envelope (`~/code/traces`, 18 harness adapters: claude, codex, amp, gemini, opencode, pi, copilot, factory, forge, qwen, …) | imports the two stores above | `cwd` filter exists; **no image/sandbox field in the envelope** | yes | adapter-dependent | no — the schema itself cannot express environment identity |
25
- | Sandbox sessions (agent-dev-container `PersistedSession`) | remote fleet; none stored on this host | runtime knows the image — `runtime.ready` event carries `image?` + `sandboxId` + backend — but `PersistedSession` persists only `workspaceRoot` + free-form `metadata`; `image?` is optional even on the event | yes (message store) | via tool parts, not normalized | **almost** — the image is in hand at runtime and dropped at persistence time |
25
+ | Sandbox sessions (agent-container `PersistedSession`) | remote fleet; none stored on this host | runtime knows the image — `runtime.ready` event carries `image?` + `sandboxId` + backend — but `PersistedSession` persists only `workspaceRoot` + free-form `metadata`; `image?` is optional even on the event | yes (message store) | via tool parts, not normalized | **almost** — the image is in hand at runtime and dropped at persistence time |
26
26
  | mini-SWE / CodeTraceBench benchmark trajectories (`~/bench-cache/ctb-20260801`) | 133 labeled, 22 replayable | yes — `mswebench/*` images + cwd | yes | yes — `<returncode>N</returncode>` in every observation | **yes — the only eligible class; run2 proved 16/22 reproduce** |
27
27
 
28
28
  Conclusion: today only benchmark-imported trajectories are replay-eligible.
@@ -31,7 +31,7 @@ The sandbox class is one persistence field away from eligibility — the runtime
31
31
 
32
32
  ## Phase-2 capture changes (ranked by unlock per line of code)
33
33
 
34
- 1. **Persist the sandbox image at session start** (agent-dev-container): copy `runtime.ready`'s `image` (as a digest, not a tag) + `sandboxId` into `PersistedSession` as first-class fields, and make `image` required on the event.
34
+ 1. **Persist the sandbox image at session start** (agent-container): copy `runtime.ready`'s `image` (as a digest, not a tag) + `sandboxId` into `PersistedSession` as first-class fields, and make `image` required on the event.
35
35
  This single change makes every future fleet sandbox session satisfy requirements 1–2 — the highest-leverage line in the flywheel.
36
36
  2. **Add environment identity to the traces envelope** (traces repo): an optional `environment: { image?, imageDigest?, cwd, gitCommit? }` block on the normalized session, populated by adapters where known.
37
37
  Without it, eligible sandbox sessions lose their eligibility at import time.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@tangle-network/agent-eval",
3
- "version": "0.213.0",
3
+ "version": "0.214.0",
4
4
  "description": "Evaluate and improve AI agents from runs, traces, judges, and feedback. Compare candidates, cluster failures, measure lift, and gate releases.",
5
5
  "homepage": "https://github.com/tangle-network/agent-eval#readme",
6
6
  "repository": {