@tangle-network/agent-eval 0.203.1 → 0.204.0

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 (246) hide show
  1. package/CHANGELOG.md +8 -0
  2. package/dist/adapters/http.d.ts +2 -2
  3. package/dist/analyst/index.d.ts +6 -6
  4. package/dist/analyst/index.js +5 -5
  5. package/dist/{benchmark-command-6OBcB0sG.js → benchmark-command-Jv_puUGM.js} +5 -5
  6. package/dist/{benchmark-command-6OBcB0sG.js.map → benchmark-command-Jv_puUGM.js.map} +1 -1
  7. package/dist/benchmarks/index.d.ts +4 -4
  8. package/dist/benchmarks/index.js +2 -2
  9. package/dist/campaign/index.d.ts +4 -4
  10. package/dist/campaign/index.js +3 -3
  11. package/dist/{campaign-BbObY2c6.js → campaign-DYhg4SjN.js} +4 -4
  12. package/dist/{campaign-BbObY2c6.js.map → campaign-DYhg4SjN.js.map} +1 -1
  13. package/dist/cli.js +1 -1
  14. package/dist/contract/index.d.ts +7 -7
  15. package/dist/contract/index.js +4 -4
  16. package/dist/{default-registry-CxEBVA6W.d.ts → default-registry-BHEfZHQd.d.ts} +3 -3
  17. package/dist/{default-registry-CxEBVA6W.d.ts.map → default-registry-BHEfZHQd.d.ts.map} +1 -1
  18. package/dist/{default-registry-wInsQWlQ.js → default-registry-BthquIoC.js} +2 -2
  19. package/dist/{default-registry-wInsQWlQ.js.map → default-registry-BthquIoC.js.map} +1 -1
  20. package/dist/{define-agent-eval-Da8mHyc7.js → define-agent-eval-B25Cc3ki.js} +2 -2
  21. package/dist/{define-agent-eval-Da8mHyc7.js.map → define-agent-eval-B25Cc3ki.js.map} +1 -1
  22. package/dist/{define-agent-eval-BbEXdM3H.d.ts → define-agent-eval-D8bPoBM5.d.ts} +5 -5
  23. package/dist/{define-agent-eval-BbEXdM3H.d.ts.map → define-agent-eval-D8bPoBM5.d.ts.map} +1 -1
  24. package/dist/{dspy-rlm-engine-D0h-P2Mz.js → dspy-rlm-engine-DCQKNQMn.js} +2 -2
  25. package/dist/{dspy-rlm-engine-D0h-P2Mz.js.map → dspy-rlm-engine-DCQKNQMn.js.map} +1 -1
  26. package/dist/{engine-CaSqEVlL.d.ts → engine-F_PvwesR.d.ts} +2 -2
  27. package/dist/{engine-CaSqEVlL.d.ts.map → engine-F_PvwesR.d.ts.map} +1 -1
  28. package/dist/evaluation.d.ts +1 -1
  29. package/dist/experiment/index.d.ts +1 -1
  30. package/dist/{index-Ce2HACcP.d.ts → index-B8n12TR2.d.ts} +9 -11
  31. package/dist/index-B8n12TR2.d.ts.map +1 -0
  32. package/dist/{index-CIkxRsLS.d.ts → index-C9l_-23M.d.ts} +3 -3
  33. package/dist/{index-CIkxRsLS.d.ts.map → index-C9l_-23M.d.ts.map} +1 -1
  34. package/dist/{index-CGAwP9l3.d.ts → index-CG7mtu_G.d.ts} +6 -6
  35. package/dist/{index-CGAwP9l3.d.ts.map → index-CG7mtu_G.d.ts.map} +1 -1
  36. package/dist/index.d.ts +10 -10
  37. package/dist/index.js +10 -10
  38. package/dist/jev.d.ts +1 -1
  39. package/dist/{kind-factory-CdPPHtGZ.js → kind-factory-Cn1lbsAd.js} +5 -5
  40. package/dist/kind-factory-Cn1lbsAd.js.map +1 -0
  41. package/dist/{llm-judge-EUMHOZOQ.js → llm-judge-DyZdrp6I.js} +54 -12
  42. package/dist/llm-judge-DyZdrp6I.js.map +1 -0
  43. package/dist/{matrix-BqHiMKIj.d.ts → matrix-W1lbSx72.d.ts} +2 -2
  44. package/dist/{matrix-BqHiMKIj.d.ts.map → matrix-W1lbSx72.d.ts.map} +1 -1
  45. package/dist/multishot/golden/index.d.ts +1 -1
  46. package/dist/multishot/index.d.ts +2 -2
  47. package/dist/openapi.json +1 -1
  48. package/dist/{produced-state-g9BPgu4D.js → produced-state-DLToUXrD.js} +2 -2
  49. package/dist/{produced-state-g9BPgu4D.js.map → produced-state-DLToUXrD.js.map} +1 -1
  50. package/dist/rl.d.ts +2 -2
  51. package/dist/search/index.d.ts +2 -2
  52. package/dist/{search-kernel-C5jua_UB.d.ts → search-kernel-DNpsHb8P.d.ts} +2 -2
  53. package/dist/{search-kernel-C5jua_UB.d.ts.map → search-kernel-DNpsHb8P.d.ts.map} +1 -1
  54. package/dist/{semantic-concept-judge-BgWxa4L_.js → semantic-concept-judge-BbiYuv7U.js} +3 -3
  55. package/dist/{semantic-concept-judge-BgWxa4L_.js.map → semantic-concept-judge-BbiYuv7U.js.map} +1 -1
  56. package/dist/{skillopt-optimization-method-DZrdQGZl.js → skillopt-optimization-method-Dz5CGZb8.js} +2 -2
  57. package/dist/{skillopt-optimization-method-DZrdQGZl.js.map → skillopt-optimization-method-Dz5CGZb8.js.map} +1 -1
  58. package/dist/{statistical-heldout-CJYAO-4a.d.ts → statistical-heldout-Dq3pnmxd.d.ts} +2 -2
  59. package/dist/{statistical-heldout-CJYAO-4a.d.ts.map → statistical-heldout-Dq3pnmxd.d.ts.map} +1 -1
  60. package/dist/{store-otlp-yd9lDziN.js → store-otlp-BA7G1Du-.js} +44 -11
  61. package/dist/store-otlp-BA7G1Du-.js.map +1 -0
  62. package/dist/{store-tool-spans-fIE31bU8.d.ts → store-tool-spans-BYIarVrG.d.ts} +19 -10
  63. package/dist/{store-tool-spans-fIE31bU8.d.ts.map → store-tool-spans-BYIarVrG.d.ts.map} +1 -1
  64. package/dist/{store-tool-spans-D5kOU0mr.js → store-tool-spans-DKoVbE-w.js} +3 -3
  65. package/dist/{store-tool-spans-D5kOU0mr.js.map → store-tool-spans-DKoVbE-w.js.map} +1 -1
  66. package/dist/{tool-groups-DP8b5kjF.d.ts → tool-groups-B7_jUPHp.d.ts} +2 -2
  67. package/dist/tool-groups-B7_jUPHp.d.ts.map +1 -0
  68. package/dist/{trace-questions-Cd0Fyhbv.d.ts → trace-questions-Dd3AMf-U.d.ts} +2 -2
  69. package/dist/{trace-questions-Cd0Fyhbv.d.ts.map → trace-questions-Dd3AMf-U.d.ts.map} +1 -1
  70. package/dist/trace-repair/index.d.ts +1 -1
  71. package/dist/traces.d.ts +3 -3
  72. package/dist/traces.js +4 -4
  73. package/dist/{types-hH9APahd.d.ts → types-DZQpUJse.d.ts} +2 -2
  74. package/dist/{types-hH9APahd.d.ts.map → types-DZQpUJse.d.ts.map} +1 -1
  75. package/dist/{types-Duc6xOoW.d.ts → types-ZUy5HGY5.d.ts} +9 -3
  76. package/dist/{types-Duc6xOoW.d.ts.map → types-ZUy5HGY5.d.ts.map} +1 -1
  77. package/docs/eval-surface-map.md +12 -2
  78. package/docs/proofs/campaign-attempt-retention-20261001/README.md +70 -0
  79. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/attempts/e7949644-3d60-4ab9-817e-f80d0a48f5d8/attempt-1/artifacts/capture.json +3 -0
  80. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/attempts/e7949644-3d60-4ab9-817e-f80d0a48f5d8/attempt-1/identity.json +10 -0
  81. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/attempts/e7949644-3d60-4ab9-817e-f80d0a48f5d8/attempt-1/result.json +1 -0
  82. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/attempts/e7949644-3d60-4ab9-817e-f80d0a48f5d8/attempt-1/trace/spans.jsonl +0 -0
  83. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/cached-result.json +1 -0
  84. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-judge-attribution/attribution_0/latest-attempt.json +1 -0
  85. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/attempts/7c7718f6-c85e-4867-9e3a-13ef9513de80/attempt-1/artifacts/captured-response.json +4 -0
  86. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/attempts/7c7718f6-c85e-4867-9e3a-13ef9513de80/attempt-1/identity.json +10 -0
  87. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/attempts/7c7718f6-c85e-4867-9e3a-13ef9513de80/attempt-1/result.json +1 -0
  88. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/attempts/7c7718f6-c85e-4867-9e3a-13ef9513de80/attempt-1/trace/spans.jsonl +1 -0
  89. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/cached-result.json +1 -0
  90. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/captured-response.json +6 -0
  91. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/failure-receipt.json +58 -0
  92. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/latest-attempt.json +1 -0
  93. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-legacy/capture-retention_0/spans.jsonl +1 -0
  94. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/attempts/3652118b-e1dc-4b17-bc15-6b802dc71eb1/attempt-1/artifacts/captured-response.json +4 -0
  95. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/attempts/3652118b-e1dc-4b17-bc15-6b802dc71eb1/attempt-1/identity.json +10 -0
  96. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/attempts/3652118b-e1dc-4b17-bc15-6b802dc71eb1/attempt-1/result.json +1 -0
  97. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/attempts/3652118b-e1dc-4b17-bc15-6b802dc71eb1/attempt-1/trace/spans.jsonl +1 -0
  98. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/cached-result.json +1 -0
  99. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/baseline/capture-retention_0/latest-attempt.json +1 -0
  100. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/attempts/40e411fe-ad12-4e6b-a279-2cf07c356a2f/attempt-1/artifacts/captured-response.json +4 -0
  101. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/attempts/40e411fe-ad12-4e6b-a279-2cf07c356a2f/attempt-1/failure-receipt.json +62 -0
  102. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/attempts/40e411fe-ad12-4e6b-a279-2cf07c356a2f/attempt-1/identity.json +10 -0
  103. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/attempts/40e411fe-ad12-4e6b-a279-2cf07c356a2f/attempt-1/result.json +1 -0
  104. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/attempts/40e411fe-ad12-4e6b-a279-2cf07c356a2f/attempt-1/trace/spans.jsonl +1 -0
  105. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/42b3fec6b7e55008/capture-retention_0/latest-attempt.json +1 -0
  106. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/attempts/9e7da63e-2d19-4ec1-a5ec-ec804e2b0f40/attempt-1/artifacts/captured-response.json +4 -0
  107. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/attempts/9e7da63e-2d19-4ec1-a5ec-ec804e2b0f40/attempt-1/identity.json +10 -0
  108. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/attempts/9e7da63e-2d19-4ec1-a5ec-ec804e2b0f40/attempt-1/result.json +1 -0
  109. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/attempts/9e7da63e-2d19-4ec1-a5ec-ec804e2b0f40/attempt-1/trace/spans.jsonl +1 -0
  110. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/cached-result.json +1 -0
  111. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/candidates/87740cc101602a08/capture-retention_0/latest-attempt.json +1 -0
  112. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/first-interruption.json +38 -0
  113. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/19147243d263e55b1eaf1c940691947382a7cce17c5f49e4a4da765a473d2a55.json +1 -0
  114. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/5dd0d871d559d1c55b2ac04b3fbd26d62f7c9e25d7632d4542502cd292818fc0.json +1 -0
  115. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/5de45d632937e75ff90fec9f1f2ec425904d3632da71f6b26f83b1e1eb8cde5a.json +1 -0
  116. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/6caba627806e499aa3d9e156aec9d5a731a3ef30e29756d343b85b7217c77c11.json +1 -0
  117. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/ce2770851b3b7651f27cedcbad0372b9191e9a9b016665987e3ddd2ad53012b5.json +1 -0
  118. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/d8ac4ba8d9844afb4defa5cd277fe3cb944d99e58cf41c07d3ae199b6f467966.json +1 -0
  119. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/blobs/f59a02035fe96ea2f5582dd715d2e3ad8c184ae959310978962f1ac0780753bd.json +1 -0
  120. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/ledger.jsonl +23 -0
  121. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-optimization/search/ledger.jsonl.head +1 -0
  122. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/attempts/3c8272d6-d35b-477e-9d08-085261b8dc64/attempt-1/artifacts/capture.json +3 -0
  123. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/attempts/3c8272d6-d35b-477e-9d08-085261b8dc64/attempt-1/identity.json +10 -0
  124. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/attempts/3c8272d6-d35b-477e-9d08-085261b8dc64/attempt-1/result.json +1 -0
  125. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/attempts/3c8272d6-d35b-477e-9d08-085261b8dc64/attempt-1/trace/spans.jsonl +0 -0
  126. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/cached-result.json +1 -0
  127. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-path-guard/capture-retention_0/latest-attempt.json +1 -0
  128. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/2b022ccb-adb3-4468-8c21-48ee82d0634a/attempt-1/artifacts/captured-response.json +6 -0
  129. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/2b022ccb-adb3-4468-8c21-48ee82d0634a/attempt-1/failure-receipt.json +81 -0
  130. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/2b022ccb-adb3-4468-8c21-48ee82d0634a/attempt-1/identity.json +10 -0
  131. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/2b022ccb-adb3-4468-8c21-48ee82d0634a/attempt-1/result.json +1 -0
  132. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/2b022ccb-adb3-4468-8c21-48ee82d0634a/attempt-1/trace/spans.jsonl +2 -0
  133. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/c4859951-1e7c-4d23-b89a-0ca8fa79e2cf/attempt-1/artifacts/captured-response.json +6 -0
  134. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/c4859951-1e7c-4d23-b89a-0ca8fa79e2cf/attempt-1/identity.json +10 -0
  135. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/c4859951-1e7c-4d23-b89a-0ca8fa79e2cf/attempt-1/result.json +1 -0
  136. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/attempts/c4859951-1e7c-4d23-b89a-0ca8fa79e2cf/attempt-1/trace/spans.jsonl +2 -0
  137. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/cached-result.json +1 -0
  138. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/capture-retention_0/latest-attempt.json +1 -0
  139. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-resume/observation.json +608 -0
  140. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/artifacts/captured-response.json +6 -0
  141. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/failure-receipt.json +81 -0
  142. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/identity.json +10 -0
  143. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/result.json +1 -0
  144. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/trace/spans.jsonl +2 -0
  145. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/artifacts/captured-response.json +6 -0
  146. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/identity.json +10 -0
  147. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/result.json +1 -0
  148. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/trace/spans.jsonl +2 -0
  149. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/cached-result.json +1 -0
  150. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/capture-retention_0/latest-attempt.json +1 -0
  151. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-retry/observation.json +514 -0
  152. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/after-summary.json +1853 -0
  153. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/judge-attribution.json +140 -0
  154. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/qualification.log +1 -0
  155. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/qualify.mjs +165 -0
  156. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduce.mjs +39 -0
  157. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-resume/capture-retention_0/cached-result.json +1 -0
  158. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-resume/capture-retention_0/captured-response.json +5 -0
  159. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-resume/capture-retention_0/failure-receipt.json +58 -0
  160. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-resume/capture-retention_0/spans.jsonl +1 -0
  161. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-resume/observer-retained-captures.json +20 -0
  162. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-retry/capture-retention_0/cached-result.json +1 -0
  163. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-retry/capture-retention_0/captured-response.json +5 -0
  164. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-retry/capture-retention_0/failure-receipt.attempt-1.json +58 -0
  165. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-retry/capture-retention_0/spans.jsonl +1 -0
  166. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-retry/observer-retained-captures.json +20 -0
  167. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction-summary.json +315 -0
  168. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/reproduction.log +1 -0
  169. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/source-and-build.json +2409 -0
  170. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/typecheck.log +1 -0
  171. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/current/verify-package.log +340 -0
  172. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/attempts/33d57ccb-d4d8-4c23-92e3-c1ddf7db6058/attempt-1/artifacts/captured-response.json +4 -0
  173. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/attempts/33d57ccb-d4d8-4c23-92e3-c1ddf7db6058/attempt-1/identity.json +10 -0
  174. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/attempts/33d57ccb-d4d8-4c23-92e3-c1ddf7db6058/attempt-1/result.json +1 -0
  175. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/attempts/33d57ccb-d4d8-4c23-92e3-c1ddf7db6058/attempt-1/trace/spans.jsonl +1 -0
  176. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/cached-result.json +1 -0
  177. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/captured-response.json +6 -0
  178. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/failure-receipt.json +58 -0
  179. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/latest-attempt.json +1 -0
  180. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-legacy/capture-retention_0/spans.jsonl +1 -0
  181. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/attempts/6b25a1d2-5324-445b-b59d-bdc7381cd4b4/attempt-1/artifacts/captured-response.json +4 -0
  182. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/attempts/6b25a1d2-5324-445b-b59d-bdc7381cd4b4/attempt-1/identity.json +10 -0
  183. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/attempts/6b25a1d2-5324-445b-b59d-bdc7381cd4b4/attempt-1/result.json +1 -0
  184. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/attempts/6b25a1d2-5324-445b-b59d-bdc7381cd4b4/attempt-1/trace/spans.jsonl +1 -0
  185. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/cached-result.json +1 -0
  186. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/baseline/capture-retention_0/latest-attempt.json +1 -0
  187. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/first-interruption.json +5 -0
  188. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/search/blobs/5de45d632937e75ff90fec9f1f2ec425904d3632da71f6b26f83b1e1eb8cde5a.json +1 -0
  189. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/search/ledger.jsonl +7 -0
  190. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-optimization/search/ledger.jsonl.head +1 -0
  191. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/attempts/723fa9c6-099e-437c-bddf-a4cbd28a8508/attempt-1/artifacts/capture.json +3 -0
  192. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/attempts/723fa9c6-099e-437c-bddf-a4cbd28a8508/attempt-1/identity.json +10 -0
  193. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/attempts/723fa9c6-099e-437c-bddf-a4cbd28a8508/attempt-1/result.json +1 -0
  194. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/attempts/723fa9c6-099e-437c-bddf-a4cbd28a8508/attempt-1/trace/spans.jsonl +0 -0
  195. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/cached-result.json +1 -0
  196. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-path-guard/capture-retention_0/latest-attempt.json +1 -0
  197. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/06db3429-c5c6-42c1-8623-08e2da1f563e/attempt-1/artifacts/captured-response.json +6 -0
  198. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/06db3429-c5c6-42c1-8623-08e2da1f563e/attempt-1/failure-receipt.json +62 -0
  199. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/06db3429-c5c6-42c1-8623-08e2da1f563e/attempt-1/identity.json +10 -0
  200. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/06db3429-c5c6-42c1-8623-08e2da1f563e/attempt-1/result.json +1 -0
  201. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/06db3429-c5c6-42c1-8623-08e2da1f563e/attempt-1/trace/spans.jsonl +1 -0
  202. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/d49f7e40-c4d2-4cb6-98d1-5e49f45c6e2c/attempt-1/artifacts/captured-response.json +6 -0
  203. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/d49f7e40-c4d2-4cb6-98d1-5e49f45c6e2c/attempt-1/identity.json +10 -0
  204. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/d49f7e40-c4d2-4cb6-98d1-5e49f45c6e2c/attempt-1/result.json +1 -0
  205. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/attempts/d49f7e40-c4d2-4cb6-98d1-5e49f45c6e2c/attempt-1/trace/spans.jsonl +1 -0
  206. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/cached-result.json +1 -0
  207. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/capture-retention_0/latest-attempt.json +1 -0
  208. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-resume/observation.json +458 -0
  209. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-1/artifacts/captured-response.json +6 -0
  210. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-1/failure-receipt.json +62 -0
  211. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-1/identity.json +10 -0
  212. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-1/result.json +1 -0
  213. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-1/trace/spans.jsonl +1 -0
  214. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-2/artifacts/captured-response.json +6 -0
  215. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-2/identity.json +10 -0
  216. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-2/result.json +1 -0
  217. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/attempts/cf34f352-67e5-4e8b-a510-0e896bcc96fc/attempt-2/trace/spans.jsonl +1 -0
  218. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/cached-result.json +1 -0
  219. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/capture-retention_0/latest-attempt.json +1 -0
  220. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/after-retry/observation.json +380 -0
  221. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-resume/capture-retention_0/cached-result.json +1 -0
  222. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-resume/capture-retention_0/captured-response.json +6 -0
  223. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-resume/capture-retention_0/failure-receipt.json +58 -0
  224. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-resume/capture-retention_0/spans.jsonl +1 -0
  225. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-resume/observation.json +198 -0
  226. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-retry/capture-retention_0/cached-result.json +1 -0
  227. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-retry/capture-retention_0/captured-response.json +6 -0
  228. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-retry/capture-retention_0/failure-receipt.attempt-1.json +58 -0
  229. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-retry/capture-retention_0/spans.jsonl +1 -0
  230. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-retry/observation.json +121 -0
  231. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/before-summary.json +321 -0
  232. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/build.attribution.log +491 -0
  233. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/build.final.log +491 -0
  234. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/build.log +495 -0
  235. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/qualification.log +18 -0
  236. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/qualification.wrong-root.log +9 -0
  237. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/qualify.mjs +148 -0
  238. package/docs/proofs/campaign-attempt-retention-20261001/artifacts/predecessor/qualify.wrong-root.mjs +148 -0
  239. package/docs/proofs/campaign-attempt-retention-20261001/manifest.json +808 -0
  240. package/docs/trace-analysis.md +10 -0
  241. package/package.json +1 -1
  242. package/dist/index-Ce2HACcP.d.ts.map +0 -1
  243. package/dist/kind-factory-CdPPHtGZ.js.map +0 -1
  244. package/dist/llm-judge-EUMHOZOQ.js.map +0 -1
  245. package/dist/store-otlp-yd9lDziN.js.map +0 -1
  246. package/dist/tool-groups-DP8b5kjF.d.ts.map +0 -1
@@ -0,0 +1,81 @@
1
+ {
2
+ "schemaVersion": 1,
3
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
4
+ "recordedAt": "2026-10-01T17:12:40.126Z",
5
+ "failure": {
6
+ "stage": "dispatch",
7
+ "error": {
8
+ "name": "Error",
9
+ "message": "Operational interruption after native artifact capture",
10
+ "stack": "Error: Operational interruption after native artifact capture\n at Object.dispatch (file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:36:33)\n at async executeCell (file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:491:19)\n at async file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:923:15\n at async Promise.allSettled (index 0)\n at async runCampaign (file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:970:22)\n at async qualify (file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:40:38)\n at async file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:85:96"
11
+ }
12
+ },
13
+ "cell": {
14
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
15
+ "attempt": {
16
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
17
+ "number": 1
18
+ },
19
+ "cellId": "capture-retention:0",
20
+ "scenarioId": "capture-retention",
21
+ "rep": 0,
22
+ "artifact": null,
23
+ "judgeScores": {},
24
+ "costUsd": 0,
25
+ "costProvenance": {
26
+ "kind": "estimated",
27
+ "usd": 0
28
+ },
29
+ "costCallIds": [
30
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
31
+ ],
32
+ "tokenUsage": {
33
+ "input": 0,
34
+ "output": 0
35
+ },
36
+ "resolvedModels": [
37
+ "not-an-inference"
38
+ ],
39
+ "resolvedModel": "not-an-inference",
40
+ "durationMs": 6,
41
+ "seed": 42,
42
+ "cached": false,
43
+ "errorStage": "dispatch",
44
+ "error": "Operational interruption after native artifact capture"
45
+ },
46
+ "cost": {
47
+ "totalCalls": 1,
48
+ "pendingCalls": 0,
49
+ "unresolvedCalls": 0,
50
+ "reservedCostUsd": 0,
51
+ "inputTokens": 0,
52
+ "outputTokens": 0,
53
+ "reasoningTokens": 0,
54
+ "cachedTokens": 0,
55
+ "cacheWriteTokens": 0,
56
+ "totalCostUsd": 0,
57
+ "costProvenance": {
58
+ "kind": "estimated",
59
+ "usd": 0
60
+ },
61
+ "byChannel": [
62
+ {
63
+ "channel": "agent",
64
+ "calls": 1,
65
+ "inputTokens": 0,
66
+ "outputTokens": 0,
67
+ "reasoningTokens": 0,
68
+ "cachedTokens": 0,
69
+ "cacheWriteTokens": 0,
70
+ "costUsd": 0,
71
+ "unpricedCalls": 0,
72
+ "unknownUsageCalls": 0
73
+ }
74
+ ],
75
+ "unpricedModels": [],
76
+ "fullyPriced": true,
77
+ "usageComplete": true,
78
+ "accountingComplete": true,
79
+ "incompleteReasons": []
80
+ }
81
+ }
@@ -0,0 +1,10 @@
1
+ {
2
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
3
+ "number": 1,
4
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
5
+ "cellId": "capture-retention:0",
6
+ "scenarioId": "capture-retention",
7
+ "rep": 0,
8
+ "seed": 42,
9
+ "startedAt": "2026-10-01T17:12:40.120Z"
10
+ }
@@ -0,0 +1 @@
1
+ {"manifestHash":"6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c","attempt":{"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","number":1},"cellId":"capture-retention:0","scenarioId":"capture-retention","rep":0,"artifact":null,"judgeScores":{},"costUsd":0,"costProvenance":{"kind":"estimated","usd":0},"costCallIds":["c26674f7-275c-4c6c-aceb-8516c1eff8e9"],"tokenUsage":{"input":0,"output":0},"resolvedModels":["not-an-inference"],"resolvedModel":"not-an-inference","durationMs":6,"seed":42,"cached":false,"errorStage":"dispatch","error":"Operational interruption after native artifact capture"}
@@ -0,0 +1,2 @@
1
+ {"name":"native.callback.capture","cellId":"capture-retention:0","startMs":1790874760121,"invocation":1,"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","durationMs":0}
2
+ {"name":"cost.operational-file-read","cellId":"capture-retention:0","startMs":1790874760125,"amountUsd":0,"durationMs":0}
@@ -0,0 +1,6 @@
1
+ {
2
+ "cellId": "capture-retention:0",
3
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
4
+ "invocation": 2,
5
+ "content": "Actual filesystem capture from invocation 2"
6
+ }
@@ -0,0 +1,10 @@
1
+ {
2
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
3
+ "number": 2,
4
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
5
+ "cellId": "capture-retention:0",
6
+ "scenarioId": "capture-retention",
7
+ "rep": 0,
8
+ "seed": 42,
9
+ "startedAt": "2026-10-01T17:12:40.126Z"
10
+ }
@@ -0,0 +1 @@
1
+ {"manifestHash":"6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c","attempt":{"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","number":2},"cellId":"capture-retention:0","scenarioId":"capture-retention","rep":0,"artifact":{"cellId":"capture-retention:0","runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","invocation":2,"content":"Actual filesystem capture from invocation 2"},"judgeScores":{},"costUsd":0,"costProvenance":{"kind":"estimated","usd":0},"costCallIds":["a409cf65-dba3-4dbd-b9b4-826f8ddd0c67","c26674f7-275c-4c6c-aceb-8516c1eff8e9"],"tokenUsage":{"input":0,"output":0},"resolvedModels":["not-an-inference"],"resolvedModel":"not-an-inference","durationMs":1,"seed":42,"cached":false,"retryAttempts":1}
@@ -0,0 +1,2 @@
1
+ {"name":"native.callback.capture","cellId":"capture-retention:0","startMs":1790874760126,"invocation":2,"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","durationMs":0}
2
+ {"name":"cost.operational-file-read","cellId":"capture-retention:0","startMs":1790874760127,"amountUsd":0,"durationMs":0}
@@ -0,0 +1 @@
1
+ {"manifestHash":"6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c","attempt":{"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","number":2},"cellId":"capture-retention:0","scenarioId":"capture-retention","rep":0,"artifact":{"cellId":"capture-retention:0","runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","invocation":2,"content":"Actual filesystem capture from invocation 2"},"judgeScores":{},"costUsd":0,"costProvenance":{"kind":"estimated","usd":0},"costCallIds":["a409cf65-dba3-4dbd-b9b4-826f8ddd0c67","c26674f7-275c-4c6c-aceb-8516c1eff8e9"],"tokenUsage":{"input":0,"output":0},"resolvedModels":["not-an-inference"],"resolvedModel":"not-an-inference","durationMs":1,"seed":42,"cached":false,"retryAttempts":1}
@@ -0,0 +1 @@
1
+ {"runAttemptId":"74c13884-d47e-4b6d-a823-5bd44ee0e6cb","number":2}
@@ -0,0 +1,514 @@
1
+ {
2
+ "invocations": 2,
3
+ "observed": [
4
+ {
5
+ "invocation": 1,
6
+ "path": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/artifacts/captured-response.json",
7
+ "sha256": "0b87b7e969e1233f4927f091369327480e998e28ac98bbe97a69423bd282fdac",
8
+ "value": {
9
+ "cellId": "capture-retention:0",
10
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
11
+ "invocation": 1,
12
+ "content": "Actual filesystem capture from invocation 1"
13
+ }
14
+ },
15
+ {
16
+ "invocation": 2,
17
+ "path": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/artifacts/captured-response.json",
18
+ "sha256": "74e46c137ea0c465395cd01a36465b52cc5658161344233e62d8bfbbc7b6ebd0",
19
+ "value": {
20
+ "cellId": "capture-retention:0",
21
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
22
+ "invocation": 2,
23
+ "content": "Actual filesystem capture from invocation 2"
24
+ }
25
+ }
26
+ ],
27
+ "attempts": [
28
+ {
29
+ "identity": {
30
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
31
+ "number": 1,
32
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
33
+ "cellId": "capture-retention:0",
34
+ "scenarioId": "capture-retention",
35
+ "rep": 0,
36
+ "seed": 42,
37
+ "startedAt": "2026-10-01T17:12:40.120Z"
38
+ },
39
+ "result": {
40
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
41
+ "attempt": {
42
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
43
+ "number": 1
44
+ },
45
+ "cellId": "capture-retention:0",
46
+ "scenarioId": "capture-retention",
47
+ "rep": 0,
48
+ "artifact": null,
49
+ "judgeScores": {},
50
+ "costUsd": 0,
51
+ "costProvenance": {
52
+ "kind": "estimated",
53
+ "usd": 0
54
+ },
55
+ "costCallIds": [
56
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
57
+ ],
58
+ "tokenUsage": {
59
+ "input": 0,
60
+ "output": 0
61
+ },
62
+ "resolvedModels": [
63
+ "not-an-inference"
64
+ ],
65
+ "resolvedModel": "not-an-inference",
66
+ "durationMs": 6,
67
+ "seed": 42,
68
+ "cached": false,
69
+ "errorStage": "dispatch",
70
+ "error": "Operational interruption after native artifact capture"
71
+ },
72
+ "trace": [
73
+ {
74
+ "name": "native.callback.capture",
75
+ "cellId": "capture-retention:0",
76
+ "startMs": 1790874760121,
77
+ "invocation": 1,
78
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
79
+ "durationMs": 0
80
+ },
81
+ {
82
+ "name": "cost.operational-file-read",
83
+ "cellId": "capture-retention:0",
84
+ "startMs": 1790874760125,
85
+ "amountUsd": 0,
86
+ "durationMs": 0
87
+ }
88
+ ],
89
+ "path": "after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1"
90
+ },
91
+ {
92
+ "identity": {
93
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
94
+ "number": 2,
95
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
96
+ "cellId": "capture-retention:0",
97
+ "scenarioId": "capture-retention",
98
+ "rep": 0,
99
+ "seed": 42,
100
+ "startedAt": "2026-10-01T17:12:40.126Z"
101
+ },
102
+ "result": {
103
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
104
+ "attempt": {
105
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
106
+ "number": 2
107
+ },
108
+ "cellId": "capture-retention:0",
109
+ "scenarioId": "capture-retention",
110
+ "rep": 0,
111
+ "artifact": {
112
+ "cellId": "capture-retention:0",
113
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
114
+ "invocation": 2,
115
+ "content": "Actual filesystem capture from invocation 2"
116
+ },
117
+ "judgeScores": {},
118
+ "costUsd": 0,
119
+ "costProvenance": {
120
+ "kind": "estimated",
121
+ "usd": 0
122
+ },
123
+ "costCallIds": [
124
+ "a409cf65-dba3-4dbd-b9b4-826f8ddd0c67",
125
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
126
+ ],
127
+ "tokenUsage": {
128
+ "input": 0,
129
+ "output": 0
130
+ },
131
+ "resolvedModels": [
132
+ "not-an-inference"
133
+ ],
134
+ "resolvedModel": "not-an-inference",
135
+ "durationMs": 1,
136
+ "seed": 42,
137
+ "cached": false,
138
+ "retryAttempts": 1
139
+ },
140
+ "trace": [
141
+ {
142
+ "name": "native.callback.capture",
143
+ "cellId": "capture-retention:0",
144
+ "startMs": 1790874760126,
145
+ "invocation": 2,
146
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
147
+ "durationMs": 0
148
+ },
149
+ {
150
+ "name": "cost.operational-file-read",
151
+ "cellId": "capture-retention:0",
152
+ "startMs": 1790874760127,
153
+ "amountUsd": 0,
154
+ "durationMs": 0
155
+ }
156
+ ],
157
+ "path": "after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2"
158
+ }
159
+ ],
160
+ "receipt": {
161
+ "schemaVersion": 1,
162
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
163
+ "recordedAt": "2026-10-01T17:12:40.126Z",
164
+ "failure": {
165
+ "stage": "dispatch",
166
+ "error": {
167
+ "name": "Error",
168
+ "message": "Operational interruption after native artifact capture",
169
+ "stack": "Error: Operational interruption after native artifact capture\n at Object.dispatch (file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:36:33)\n at async executeCell (file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:491:19)\n at async file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:923:15\n at async Promise.allSettled (index 0)\n at async runCampaign (file:///home/drew/code/_wt/eval-router-cache-usage-20261001/dist/llm-judge-DyZdrp6I.js:970:22)\n at async qualify (file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:40:38)\n at async file:///mnt/traces/eval-attempt-artifacts-20261001-r2/qualify.mjs:85:96"
170
+ }
171
+ },
172
+ "cell": {
173
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
174
+ "attempt": {
175
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
176
+ "number": 1
177
+ },
178
+ "cellId": "capture-retention:0",
179
+ "scenarioId": "capture-retention",
180
+ "rep": 0,
181
+ "artifact": null,
182
+ "judgeScores": {},
183
+ "costUsd": 0,
184
+ "costProvenance": {
185
+ "kind": "estimated",
186
+ "usd": 0
187
+ },
188
+ "costCallIds": [
189
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
190
+ ],
191
+ "tokenUsage": {
192
+ "input": 0,
193
+ "output": 0
194
+ },
195
+ "resolvedModels": [
196
+ "not-an-inference"
197
+ ],
198
+ "resolvedModel": "not-an-inference",
199
+ "durationMs": 6,
200
+ "seed": 42,
201
+ "cached": false,
202
+ "errorStage": "dispatch",
203
+ "error": "Operational interruption after native artifact capture"
204
+ },
205
+ "cost": {
206
+ "totalCalls": 1,
207
+ "pendingCalls": 0,
208
+ "unresolvedCalls": 0,
209
+ "reservedCostUsd": 0,
210
+ "inputTokens": 0,
211
+ "outputTokens": 0,
212
+ "reasoningTokens": 0,
213
+ "cachedTokens": 0,
214
+ "cacheWriteTokens": 0,
215
+ "totalCostUsd": 0,
216
+ "costProvenance": {
217
+ "kind": "estimated",
218
+ "usd": 0
219
+ },
220
+ "byChannel": [
221
+ {
222
+ "channel": "agent",
223
+ "calls": 1,
224
+ "inputTokens": 0,
225
+ "outputTokens": 0,
226
+ "reasoningTokens": 0,
227
+ "cachedTokens": 0,
228
+ "cacheWriteTokens": 0,
229
+ "costUsd": 0,
230
+ "unpricedCalls": 0,
231
+ "unknownUsageCalls": 0
232
+ }
233
+ ],
234
+ "unpricedModels": [],
235
+ "fullyPriced": true,
236
+ "usageComplete": true,
237
+ "accountingComplete": true,
238
+ "incompleteReasons": []
239
+ }
240
+ },
241
+ "pointer": {
242
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
243
+ "number": 2
244
+ },
245
+ "reports": [
246
+ {
247
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
248
+ "splitDigest": "sha256:23ea149e81be5435966bd1d185776c7d2e119739830f98a47d4738bdbd46ade6",
249
+ "seed": 42,
250
+ "reps": 1,
251
+ "startedAt": "2026-10-01T17:12:40.119Z",
252
+ "endedAt": "2026-10-01T17:12:40.127Z",
253
+ "durationMs": 8,
254
+ "cells": [
255
+ {
256
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
257
+ "attempt": {
258
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
259
+ "number": 2
260
+ },
261
+ "cellId": "capture-retention:0",
262
+ "scenarioId": "capture-retention",
263
+ "rep": 0,
264
+ "artifact": {
265
+ "cellId": "capture-retention:0",
266
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
267
+ "invocation": 2,
268
+ "content": "Actual filesystem capture from invocation 2"
269
+ },
270
+ "judgeScores": {},
271
+ "costUsd": 0,
272
+ "costProvenance": {
273
+ "kind": "estimated",
274
+ "usd": 0
275
+ },
276
+ "costCallIds": [
277
+ "a409cf65-dba3-4dbd-b9b4-826f8ddd0c67",
278
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
279
+ ],
280
+ "tokenUsage": {
281
+ "input": 0,
282
+ "output": 0
283
+ },
284
+ "resolvedModels": [
285
+ "not-an-inference"
286
+ ],
287
+ "resolvedModel": "not-an-inference",
288
+ "durationMs": 1,
289
+ "seed": 42,
290
+ "cached": false,
291
+ "retryAttempts": 1
292
+ }
293
+ ],
294
+ "aggregates": {
295
+ "byJudge": {},
296
+ "byScenario": {},
297
+ "cost": {
298
+ "totalCalls": 2,
299
+ "pendingCalls": 0,
300
+ "unresolvedCalls": 0,
301
+ "reservedCostUsd": 0,
302
+ "inputTokens": 0,
303
+ "outputTokens": 0,
304
+ "reasoningTokens": 0,
305
+ "cachedTokens": 0,
306
+ "cacheWriteTokens": 0,
307
+ "totalCostUsd": 0,
308
+ "costProvenance": {
309
+ "kind": "estimated",
310
+ "usd": 0
311
+ },
312
+ "byChannel": [
313
+ {
314
+ "channel": "agent",
315
+ "calls": 2,
316
+ "inputTokens": 0,
317
+ "outputTokens": 0,
318
+ "reasoningTokens": 0,
319
+ "cachedTokens": 0,
320
+ "cacheWriteTokens": 0,
321
+ "costUsd": 0,
322
+ "unpricedCalls": 0,
323
+ "unknownUsageCalls": 0
324
+ }
325
+ ],
326
+ "unpricedModels": [],
327
+ "fullyPriced": true,
328
+ "usageComplete": true,
329
+ "accountingComplete": true,
330
+ "incompleteReasons": []
331
+ },
332
+ "cellsExecuted": 1,
333
+ "cellsSkipped": 0,
334
+ "cellsCached": 0,
335
+ "cellsFailed": 0,
336
+ "cellsDispatchFailed": 0,
337
+ "cellsJudgeFailed": 0,
338
+ "cellsUnclassifiedFailed": 0
339
+ },
340
+ "runDir": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry",
341
+ "artifactsByPath": {
342
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/identity.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/identity.json",
343
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/artifacts/captured-response.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/artifacts/captured-response.json",
344
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/failure-receipt.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/failure-receipt.json",
345
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/result.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-1/result.json",
346
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/identity.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/identity.json",
347
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/artifacts/captured-response.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/artifacts/captured-response.json",
348
+ "capture-retention:0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/result.json": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry/capture-retention_0/attempts/74c13884-d47e-4b6d-a823-5bd44ee0e6cb/attempt-2/result.json"
349
+ },
350
+ "scenarios": [
351
+ {
352
+ "id": "capture-retention",
353
+ "kind": "operational",
354
+ "scenarioDigest": "sha256:70a52f79806918ce9e4833c4ed81360245988fced43a4605ab703d7afacb20a3"
355
+ }
356
+ ]
357
+ }
358
+ ],
359
+ "ledgerReceipts": [
360
+ {
361
+ "status": "settled",
362
+ "callId": "c26674f7-275c-4c6c-aceb-8516c1eff8e9",
363
+ "channel": "agent",
364
+ "phase": "campaign",
365
+ "actor": "operational-file-read",
366
+ "model": "not-an-inference",
367
+ "timestamp": 1790874760122,
368
+ "inputTokens": 0,
369
+ "outputTokens": 0,
370
+ "costUsd": 0,
371
+ "costUnknown": false,
372
+ "usageUnknown": false,
373
+ "estimatedCostUsd": 0,
374
+ "tags": {
375
+ "runDir": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry",
376
+ "cellId": "capture-retention:0",
377
+ "scenarioId": "capture-retention",
378
+ "rep": "0",
379
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
380
+ "attemptNumber": "1"
381
+ }
382
+ },
383
+ {
384
+ "status": "settled",
385
+ "callId": "a409cf65-dba3-4dbd-b9b4-826f8ddd0c67",
386
+ "channel": "agent",
387
+ "phase": "campaign",
388
+ "actor": "operational-file-read",
389
+ "model": "not-an-inference",
390
+ "timestamp": 1790874760126,
391
+ "inputTokens": 0,
392
+ "outputTokens": 0,
393
+ "costUsd": 0,
394
+ "costUnknown": false,
395
+ "usageUnknown": false,
396
+ "estimatedCostUsd": 0,
397
+ "tags": {
398
+ "runDir": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry",
399
+ "cellId": "capture-retention:0",
400
+ "scenarioId": "capture-retention",
401
+ "rep": "0",
402
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
403
+ "attemptNumber": "2"
404
+ }
405
+ }
406
+ ],
407
+ "cacheReuse": {
408
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
409
+ "splitDigest": "sha256:23ea149e81be5435966bd1d185776c7d2e119739830f98a47d4738bdbd46ade6",
410
+ "seed": 42,
411
+ "reps": 1,
412
+ "startedAt": "2026-10-01T17:12:40.128Z",
413
+ "endedAt": "2026-10-01T17:12:40.128Z",
414
+ "durationMs": 0,
415
+ "cells": [
416
+ {
417
+ "manifestHash": "6d4d16fdd6cbaa1119fc301e329b128e6485461e9bc013bdbeaea9ec0270b19c",
418
+ "attempt": {
419
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
420
+ "number": 2
421
+ },
422
+ "cellId": "capture-retention:0",
423
+ "scenarioId": "capture-retention",
424
+ "rep": 0,
425
+ "artifact": {
426
+ "cellId": "capture-retention:0",
427
+ "runAttemptId": "74c13884-d47e-4b6d-a823-5bd44ee0e6cb",
428
+ "invocation": 2,
429
+ "content": "Actual filesystem capture from invocation 2"
430
+ },
431
+ "judgeScores": {},
432
+ "costUsd": 0,
433
+ "costProvenance": {
434
+ "kind": "estimated",
435
+ "usd": 0
436
+ },
437
+ "costCallIds": [
438
+ "a409cf65-dba3-4dbd-b9b4-826f8ddd0c67",
439
+ "c26674f7-275c-4c6c-aceb-8516c1eff8e9"
440
+ ],
441
+ "tokenUsage": {
442
+ "input": 0,
443
+ "output": 0
444
+ },
445
+ "durationMs": 1,
446
+ "seed": 42,
447
+ "cached": true,
448
+ "retryAttempts": 1,
449
+ "resolvedModels": [
450
+ "not-an-inference"
451
+ ],
452
+ "resolvedModel": "not-an-inference"
453
+ }
454
+ ],
455
+ "aggregates": {
456
+ "byJudge": {},
457
+ "byScenario": {},
458
+ "cost": {
459
+ "totalCalls": 2,
460
+ "pendingCalls": 0,
461
+ "unresolvedCalls": 0,
462
+ "reservedCostUsd": 0,
463
+ "inputTokens": 0,
464
+ "outputTokens": 0,
465
+ "reasoningTokens": 0,
466
+ "cachedTokens": 0,
467
+ "cacheWriteTokens": 0,
468
+ "totalCostUsd": 0,
469
+ "costProvenance": {
470
+ "kind": "estimated",
471
+ "usd": 0
472
+ },
473
+ "byChannel": [
474
+ {
475
+ "channel": "agent",
476
+ "calls": 2,
477
+ "inputTokens": 0,
478
+ "outputTokens": 0,
479
+ "reasoningTokens": 0,
480
+ "cachedTokens": 0,
481
+ "cacheWriteTokens": 0,
482
+ "costUsd": 0,
483
+ "unpricedCalls": 0,
484
+ "unknownUsageCalls": 0
485
+ }
486
+ ],
487
+ "unpricedModels": [],
488
+ "fullyPriced": true,
489
+ "usageComplete": true,
490
+ "accountingComplete": true,
491
+ "incompleteReasons": []
492
+ },
493
+ "cellsExecuted": 1,
494
+ "cellsSkipped": 0,
495
+ "cellsCached": 1,
496
+ "cellsFailed": 0,
497
+ "cellsDispatchFailed": 0,
498
+ "cellsJudgeFailed": 0,
499
+ "cellsUnclassifiedFailed": 0
500
+ },
501
+ "runDir": "/mnt/traces/eval-attempt-artifacts-20261001-r2/after-retry",
502
+ "artifactsByPath": {},
503
+ "scenarios": [
504
+ {
505
+ "id": "capture-retention",
506
+ "kind": "operational",
507
+ "scenarioDigest": "sha256:70a52f79806918ce9e4833c4ed81360245988fced43a4605ab703d7afacb20a3"
508
+ }
509
+ ]
510
+ },
511
+ "firstContentStillPresent": true,
512
+ "modelInference": false,
513
+ "researchUnits": 0
514
+ }