@zhivex-ai/harness 1.0.0 → 1.1.0-rc.2

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 (181) hide show
  1. package/CHANGELOG.md +97 -1
  2. package/README.md +52 -306
  3. package/SECURITY.md +1 -1
  4. package/SUPPORT.md +7 -3
  5. package/benchmarks/README.md +3 -3
  6. package/contracts/client-protocol-v1.schema.json +341 -0
  7. package/contracts/public-api.json +100 -1
  8. package/contracts/stable-api-signatures.json +14 -14
  9. package/dist/approval-diff.d.ts +45 -0
  10. package/dist/approval-history.d.ts +38 -0
  11. package/dist/approval-preview.d.ts +26 -0
  12. package/dist/cli-credentials.d.ts +38 -0
  13. package/dist/cli-help.d.ts +3 -0
  14. package/dist/cli-options.d.ts +1 -1
  15. package/dist/cli-stream.d.ts +3 -3
  16. package/dist/cli.d.ts +15 -69
  17. package/dist/cli.js +13796 -10032
  18. package/dist/client-contract.d.ts +208 -0
  19. package/dist/console-commands.d.ts +6 -0
  20. package/dist/console-context.d.ts +5 -1
  21. package/dist/console-input.d.ts +44 -4
  22. package/dist/console-model-catalog.d.ts +25 -0
  23. package/dist/console-navigation.d.ts +27 -0
  24. package/dist/console-presentation.d.ts +18 -0
  25. package/dist/console-welcome.d.ts +13 -0
  26. package/dist/execution-environment.d.ts +4 -0
  27. package/dist/harness.d.ts +3 -0
  28. package/dist/hostile-repository-demo.js +1913 -798
  29. package/dist/index.d.ts +7 -0
  30. package/dist/index.js +7908 -6330
  31. package/dist/local-service.d.ts +34 -0
  32. package/dist/operations.d.ts +23 -1
  33. package/dist/orchestration.d.ts +3 -0
  34. package/dist/persistence-secrets.d.ts +2 -0
  35. package/dist/recorded-workspace.d.ts +4 -0
  36. package/dist/run-document.d.ts +91 -0
  37. package/dist/runtime-policy.d.ts +4 -4
  38. package/dist/service-cli.d.ts +2 -0
  39. package/dist/service-cli.js +10858 -0
  40. package/dist/service-client-cli.d.ts +2 -0
  41. package/dist/service-events.d.ts +50 -0
  42. package/dist/sessions.d.ts +3 -0
  43. package/dist/sqlite-access.d.ts +26 -0
  44. package/dist/sqlite-database.d.ts +2 -0
  45. package/dist/state-backup.d.ts +8 -0
  46. package/dist/task-memory.d.ts +2 -2
  47. package/dist/terminal-continuation.d.ts +3 -0
  48. package/dist/terminal-markdown.d.ts +1 -0
  49. package/dist/terminal-ui.d.ts +5 -0
  50. package/dist/usage-ledger.d.ts +84 -0
  51. package/dist/verify-historical-migrations.js +5699 -5191
  52. package/dist/workspace.d.ts +22 -0
  53. package/dist/zhx.js +13796 -10032
  54. package/docs/CLI.md +136 -30
  55. package/docs/CLIENT_PROTOCOL.md +212 -0
  56. package/docs/CREDENTIALS.md +59 -0
  57. package/docs/DURABLE_OPERATIONS.md +2 -2
  58. package/docs/FIRST_USE.md +82 -0
  59. package/docs/FIRST_USE_EXAMPLE.md +40 -0
  60. package/docs/HOSTILE_REPOSITORY_DEMO.md +1 -1
  61. package/docs/LOCAL_SERVICE.md +142 -0
  62. package/docs/README.md +24 -23
  63. package/docs/THREAT_MODEL.md +1 -1
  64. package/docs/TIME_TO_SAFE_FIX.md +2 -2
  65. package/docs/USAGE.md +181 -0
  66. package/evaluations/daily-cli-cohort.json +33 -0
  67. package/evaluations/representative-assembly-matrix.json +19 -1
  68. package/examples/README.md +26 -0
  69. package/examples/first-use/greeting.test.ts +3 -0
  70. package/examples/first-use/greeting.ts +1 -0
  71. package/examples/first-use/package.json +7 -0
  72. package/package.json +34 -9
  73. package/ROADMAP.md +0 -414
  74. package/benchmarks/baselines/context-navigation-2026-09-09.json +0 -94
  75. package/benchmarks/baselines/efficiency-2026-09-08.json +0 -233
  76. package/benchmarks/baselines/external-final-2026-09-08.json +0 -588
  77. package/benchmarks/baselines/external-intermediate-2026-09-08.json +0 -116
  78. package/benchmarks/baselines/external-matched-openai-2026-09-08.json +0 -1922
  79. package/benchmarks/baselines/external-matched-qwen-2026-09-08.json +0 -1706
  80. package/benchmarks/baselines/external-pilot-2026-09-08.json +0 -439
  81. package/benchmarks/baselines/external-qwen-flash-2026-09-08.json +0 -1570
  82. package/benchmarks/baselines/external-qwen-remediation-2026-09-09.json +0 -1509
  83. package/benchmarks/baselines/external-qwen-sdk-114-progress-2026-09-09.json +0 -2509
  84. package/benchmarks/baselines/external-sdk-upgrade-2026-09-08.json +0 -564
  85. package/benchmarks/baselines/external-verifier-recovery-2026-09-09.json +0 -794
  86. package/benchmarks/baselines/full-harness-review-2026-09-10.json +0 -134
  87. package/benchmarks/baselines/harness-audit-2026-09-09.json +0 -136
  88. package/benchmarks/baselines/harness-remediation-2026-09-09.json +0 -160
  89. package/benchmarks/baselines/oci-remediation-2026-09-10.json +0 -535
  90. package/benchmarks/baselines/rc14-holdout-2026-09-10.json +0 -854
  91. package/benchmarks/baselines/repair-architecture-audit-2026-09-09.json +0 -75
  92. package/benchmarks/baselines/time-to-safe-fix-live-expanded-2026-08-21.json +0 -186
  93. package/benchmarks/baselines/time-to-safe-fix-live-smoke-2026-08-21.json +0 -186
  94. package/docs/GA_READINESS.md +0 -29
  95. package/docs/GEMINI_1_0_DECISION.md +0 -9
  96. package/docs/LIVE_CERTIFICATION.md +0 -107
  97. package/docs/RC13_SECURITY_REVIEW.md +0 -28
  98. package/docs/RC14_DELEGATED_SECURITY_DECISION.md +0 -80
  99. package/docs/RC14_SECURITY_REVIEW.md +0 -40
  100. package/docs/RELEASE.md +0 -96
  101. package/docs/ROLLBACK.md +0 -9
  102. package/docs/SECURITY_REVIEW_EVIDENCE.md +0 -17
  103. package/docs/ga-readiness.json +0 -366
  104. package/docs/reports/EFFICIENCY_REVIEW.md +0 -156
  105. package/docs/reports/EXTERNAL_COMPARISON.md +0 -434
  106. package/docs/reports/FULL_HARNESS_REVIEW_2026-09-10.md +0 -200
  107. package/docs/reports/GA_BUDGET_REMEDIATION_2026-09-19.md +0 -38
  108. package/docs/reports/GA_REPAIR_RECOVERY_2026-09-19.md +0 -700
  109. package/docs/reports/GA_VALIDATION_2026-09-19.md +0 -111
  110. package/docs/reports/HARNESS_ARCHITECTURE_REVIEW_2026-09-09.md +0 -202
  111. package/docs/reports/HARNESS_AUDIT_2026-09-09.md +0 -306
  112. package/docs/reports/HARNESS_REMEDIATION_2026-09-09.md +0 -74
  113. package/docs/reports/HARNESS_REMEDIATION_2026-09-10.md +0 -90
  114. package/docs/reports/MODEL_COMPARISON_2026-09-09.md +0 -184
  115. package/docs/reports/QWEN_REMEDIATION_COMPARISON_2026-09-09.md +0 -142
  116. package/docs/reports/RC14_VALIDATION_2026-09-10.md +0 -45
  117. package/docs/reports/README.md +0 -26
  118. package/docs/reports/evidence/ga-budget-remediation-2026-09-19.json +0 -133
  119. package/docs/reports/evidence/ga-catalog-ablation-2026-09-19.json +0 -699
  120. package/docs/reports/evidence/ga-catalog-ablation-plan-2026-09-19.json +0 -40
  121. package/docs/reports/evidence/ga-catalog-observed-2026-09-19.json +0 -985
  122. package/docs/reports/evidence/ga-catalog-observed-plan-2026-09-19.json +0 -28
  123. package/docs/reports/evidence/ga-catalog-observed-status-2026-09-19.json +0 -15
  124. package/docs/reports/evidence/ga-complete-local-check-2026-09-19.json +0 -47
  125. package/docs/reports/evidence/ga-completion-reminder-2026-09-19.json +0 -22
  126. package/docs/reports/evidence/ga-fresh-cohort-plan-2026-09-19.json +0 -54
  127. package/docs/reports/evidence/ga-fresh-cohort-preparation-2026-09-19.json +0 -242
  128. package/docs/reports/evidence/ga-fresh-cohort-result-2026-09-19.json +0 -4318
  129. package/docs/reports/evidence/ga-local-validation-followup-2026-09-19.json +0 -66
  130. package/docs/reports/evidence/ga-node-consumers-2026-09-19.json +0 -31
  131. package/docs/reports/evidence/ga-package-content-validation-2026-09-19.json +0 -22
  132. package/docs/reports/evidence/ga-plan-completion-2026-09-19.json +0 -39
  133. package/docs/reports/evidence/ga-plan-reserve-2026-09-19.json +0 -573
  134. package/docs/reports/evidence/ga-plan-reserve-plan-2026-09-19.json +0 -24
  135. package/docs/reports/evidence/ga-plan-transition-2026-09-19.json +0 -19
  136. package/docs/reports/evidence/ga-private-verifier-plan-2026-09-19.json +0 -24
  137. package/docs/reports/evidence/ga-private-verifier-root-cause-2026-09-19.json +0 -41
  138. package/docs/reports/evidence/ga-published-sdk-2026-09-19.json +0 -35
  139. package/docs/reports/evidence/ga-python-checkout-imports-2026-09-19.json +0 -95
  140. package/docs/reports/evidence/ga-python-origin-delivery-2026-09-19.json +0 -211
  141. package/docs/reports/evidence/ga-python-origin-delivery-plan-2026-09-19.json +0 -27
  142. package/docs/reports/evidence/ga-qwen-patched-package-2026-09-19.json +0 -1092
  143. package/docs/reports/evidence/ga-qwen-patched-package-plan-2026-09-19.json +0 -18
  144. package/docs/reports/evidence/ga-qwen-published-2026-09-19.json +0 -571
  145. package/docs/reports/evidence/ga-qwen-published-plan-2026-09-19.json +0 -18
  146. package/docs/reports/evidence/ga-qwen-tool-choice-2026-09-19.json +0 -85
  147. package/docs/reports/evidence/ga-qwen-tool-choice-patched-2026-09-19.json +0 -85
  148. package/docs/reports/evidence/ga-recovery-development-2026-09-19.json +0 -1066
  149. package/docs/reports/evidence/ga-recovery-graded-development-2026-09-19.json +0 -1313
  150. package/docs/reports/evidence/ga-recovery-graded-v2-2026-09-19.json +0 -1151
  151. package/docs/reports/evidence/ga-recovery-v3-2026-09-19.json +0 -3943
  152. package/docs/reports/evidence/ga-recovery-v4-2026-09-19.json +0 -1055
  153. package/docs/reports/evidence/ga-recovery-v5-2026-09-19.json +0 -1245
  154. package/docs/reports/evidence/ga-recovery-v6-2026-09-19.json +0 -699
  155. package/docs/reports/evidence/ga-recovery-v7-2026-09-19.json +0 -854
  156. package/docs/reports/evidence/ga-recovery-v8-2026-09-19.json +0 -596
  157. package/docs/reports/evidence/ga-required-delivery-2026-09-19.json +0 -25
  158. package/docs/reports/evidence/ga-required-delivery-live-2026-09-19.json +0 -1337
  159. package/docs/reports/evidence/ga-required-delivery-live-plan-2026-09-19.json +0 -25
  160. package/docs/reports/evidence/ga-retained-candidate-grading-2026-09-19.json +0 -30
  161. package/docs/reports/evidence/ga-tool-selection-live-2026-09-19.json +0 -559
  162. package/docs/reports/evidence/ga-tool-selection-live-plan-2026-09-19.json +0 -28
  163. package/docs/reports/evidence/ga-tool-selection-recovery-2026-09-19.json +0 -27
  164. package/docs/reports/evidence/ga-validation-2026-09-19.json +0 -236
  165. package/docs/reports/evidence/ga-verifier-discovery-live-2026-09-19.json +0 -1369
  166. package/docs/reports/evidence/ga-verifier-discovery-plan-2026-09-19.json +0 -29
  167. package/docs/reports/evidence/ga-verifier-failure-identification-2026-09-19.json +0 -30
  168. package/docs/reports/evidence/ga-verifier-feedback-2026-09-19.json +0 -65
  169. package/docs/reports/evidence/ga-verifier-hints-live-2026-09-19.json +0 -1297
  170. package/docs/reports/evidence/ga-verifier-hints-plan-2026-09-19.json +0 -27
  171. package/docs/reports/evidence/ga-work-boundary-context-analysis-2026-09-19.json +0 -113
  172. package/docs/reports/evidence/ga-work-boundary-live-2026-09-19.json +0 -301
  173. package/docs/reports/evidence/ga-work-boundary-live-plan-2026-09-19.json +0 -84
  174. package/docs/reports/evidence/ga-work-budget-planning-2026-09-19.json +0 -32
  175. package/docs/reports/evidence/ga-working-context-2026-09-19.json +0 -22
  176. package/docs/reports/evidence/rc14-delegated-security-review-2026-09-20.json +0 -81
  177. package/docs/reports/evidence/rc14-ga-gate-2026-09-20.json +0 -18
  178. package/docs/reports/evidence/rc14-representative-2026-09-20.json +0 -509
  179. package/docs/reports/evidence/rc14-security-regressions-2026-09-20.json +0 -64
  180. package/docs/reports/evidence/security-current-2026-09-19.json +0 -91
  181. package/docs/reports/evidence/security-remediation-2026-09-19.json +0 -97
@@ -1,1055 +0,0 @@
1
- {
2
- "schemaVersion": 1,
3
- "kind": "development-comparison",
4
- "classification": "Known development task, one repetition per provider and agent; not holdout",
5
- "providers": {
6
- "openai": {
7
- "model": "gpt-5.6-luna",
8
- "limits": {
9
- "steps": 24,
10
- "outputPerTurn": 2048,
11
- "inputTokens": 100000,
12
- "outputTokens": 16000,
13
- "timeoutSeconds": 300,
14
- "memoryMb": 2048,
15
- "cpus": 2
16
- },
17
- "implementation": {
18
- "gitCommit": "41bd1cb4ab8481a7413f096a0817f612fe9806fe",
19
- "sources": {
20
- "bun.lock": "e370459197ce52c3961770f3c5ccaab7ff35ddbe641ef72e8bd61f87a8b8e3fe",
21
- "evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
22
- "package.json": "a4a2507e3ad876d72cad49a16e149192d33521fb4175bb40a3eb077ccfdf43a1",
23
- "scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
24
- "scripts/swebench/candidate.py": "f5e3410b91b950acfbe50fc7511891816a8fb1c223082397e2c6300cf1e866cf",
25
- "scripts/swebench/candidate.ts": "bf102becfcce0c9a3f0df0a8f1ce0e5aaea7db2d252f6d9454a8370c24e53982",
26
- "scripts/swebench/common.py": "ad790227dac3bec1f592c458c993efff4f9d3de8f0df39b33e3238b58b3f0b3b",
27
- "scripts/swebench/context-metrics.ts": "7b84da6789413aa634d997c5963ee6796584e06c3eafe5f68f424fb7e27815e9",
28
- "scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
29
- "scripts/swebench/model-budget.ts": "e46dd406ee5d58a051a15c7d7133513666292303d8c6534f861a6db0e711f47c",
30
- "scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
31
- "scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
32
- "scripts/swebench/report.py": "d0b8d45f32c48e0ba2cb977f50838065a497a0ecaf3b4243ed94139393ce4d6b",
33
- "scripts/swebench/run.py": "94dccd55dfe6f70d8f8f2da70331700ec85eddd958547e0c3afe3dc416c9802c",
34
- "scripts/swebench/telemetry.ts": "743c9baf273e9240d704d098b6bd5eea9e76ef3e49ae636aaafa9bfd7a85b45b",
35
- "scripts/swebench/test_candidate.py": "e77d16379b748baa2daf184a45ea62587d938b6a7d48dcf9f6061f8af21fafd9",
36
- "scripts/swebench/test_contracts.py": "902e966d0fbf8501dbb15e011849f160331c033cd2a5480f9920d0f365a08d03",
37
- "scripts/swebench/zhivex-driver.ts": "b17e937484a28cf7fc7728595b51900efa5015dfa37cf6c1d4f9322051fe1ece",
38
- "scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
39
- "src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
40
- "src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
41
- "src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
42
- "src/cli-options.ts": "f48d9be132fed38f2ddbcfecbd8222c664a0df2f20edc28ce8d1b9b1a1900863",
43
- "src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
44
- "src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
45
- "src/cli.ts": "5a4acb4695a968a58fbeac05c4bb89848a9653d6189590f0ff90e430ffa98d14",
46
- "src/compaction.ts": "9e06a0abeeef155377f40c160b21c56256bca95ec18c5ee897681a5fc73db169",
47
- "src/config.ts": "b1d3991e84add58603fb85b2a604b9b3eb56ae879d0fa30b0cb40545fa44edc1",
48
- "src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
49
- "src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
50
- "src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
51
- "src/context-metrics.ts": "3eb30f9db808f918ed8d73d4f8ac19a13acb09ed5c1ee70fdae05201714b9198",
52
- "src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
53
- "src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
54
- "src/execution-environment.ts": "9ce6c49afab26527d5296f0c11f106063590f7333300d58dd49d5a1ec1b4b6e9",
55
- "src/file-security.ts": "b8c9f4988c4339c73a60ab4fa8d394b3eecc19117a0c61a8d19ab6483ca253df",
56
- "src/harness.ts": "420694110f2e205322bdcf06f3571e289809c96df92ad8bbda042d218b0692e1",
57
- "src/index.ts": "4bbfcd11376d48c4f1c1ec17d6b5c9f51baa9331fa999432be178bae94530a5e",
58
- "src/json-contracts.ts": "0780145636b47829a1aa6087c574d5654eb31b7534ea7d9c38c188d648c90272",
59
- "src/mcp.ts": "7722de031921c620ec03b9e42a6596ef8415aa8e4bb5f93ea84d2427c466e554",
60
- "src/model-budget.ts": "e25eca12f998998acb2e03f808ed0d84ba69b97711fc577716f230244c62aa80",
61
- "src/operations.ts": "bcccaf4d1d0ec117c2ed35df92185564d33399e5fdc5c12225ac61f14fd0b4b1",
62
- "src/orchestration.ts": "b56c8c7d526a99b745cc17769e0a1b7b500b25920e77f34aa0dd01c9758de319",
63
- "src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
64
- "src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
65
- "src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
66
- "src/repair-controller.ts": "91150ee614ab1444b6a89342f4b327fe2d626879467f879c6cc3313b8173f69a",
67
- "src/repair-progress.ts": "3550509a2a6c1e4a2e091c1012d5fb63ad8512a06f09b41e6b3a915a9a2b31a3",
68
- "src/repair-verifier.ts": "b70fb11feee8cdc7933947ccac0b78f8bb5c0eaccc3b29ad527197d77f40ffc9",
69
- "src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
70
- "src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
71
- "src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
72
- "src/runtime-checkpoints.ts": "a9b555d4c6b5f369f86a3d79ec1a3aab45886282cc7d0ef211aa084ee17e3897",
73
- "src/runtime-diagnostics.ts": "e1e5a3a558fa2aaeb477335dc336551f022c1234fe5e742eefc6599c8803e364",
74
- "src/runtime-policy.ts": "8356acf0102f0e73839b81ed628a5c0d9a89d9c21b0ca416efedde8f1b3414e2",
75
- "src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
76
- "src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
77
- "src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
78
- "src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
79
- "src/task-memory.ts": "5b62c863be5a23c33608633cd7aac0464f5536d01cba930aca58b11870170731",
80
- "src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
81
- "src/terminal-ui.ts": "4da1878a76ba608a2517bf5d5c992a9a88702aba3fd800a9980129ba93e157bc",
82
- "src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
83
- "src/tool-registry.ts": "7e90fff11940e525883c9ec0d2adbea2e6616f8bd9bc4ec77e1b7277e4217f2c",
84
- "src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
85
- "src/workspace-mutation-lock.ts": "ca201242db16173bd3a4d86c5d8bb0c2b6f3c43bfed3e4fa4ff5f06a75b19e80",
86
- "src/workspace.ts": "809f8bbfcf72fed10d860c48585089ca65bb3ef7c672ee6df3c918907a968302"
87
- }
88
- },
89
- "report": {
90
- "schemaVersion": 1,
91
- "kind": "swebench-comparison",
92
- "manifestSha256": "9d2ef95e826442c8960bb7b285b12bd33f00b43f34d69e82f50cbbcc7e886cb5",
93
- "complete": true,
94
- "candidates": {
95
- "zhivex": {
96
- "planned": 1,
97
- "recorded": 1,
98
- "resolved": 0,
99
- "candidateCaptured": 1,
100
- "candidateResolved": 0,
101
- "candidateGraded": 1,
102
- "resolutionRate": 0.0,
103
- "missing": 0,
104
- "gradingFailures": 0,
105
- "usageComplete": false,
106
- "totalCostUsd": null,
107
- "costPerResolvedUsd": null,
108
- "resolvedP50Ms": null
109
- },
110
- "mini-swe-agent": {
111
- "planned": 1,
112
- "recorded": 1,
113
- "resolved": 1,
114
- "candidateCaptured": 1,
115
- "candidateResolved": 1,
116
- "candidateGraded": 1,
117
- "resolutionRate": 1.0,
118
- "missing": 0,
119
- "gradingFailures": 0,
120
- "usageComplete": true,
121
- "totalCostUsd": null,
122
- "costPerResolvedUsd": null,
123
- "resolvedP50Ms": 52952.04270799877
124
- }
125
- },
126
- "pairedTaskResolutionDifference": -1.0,
127
- "pairedTaskBootstrap95": null,
128
- "evidenceBoundary": "Local matched-model comparison. Missing and failed runs remain in the denominator. Small-sample intervals are exploratory. No safety superiority is inferred from SWE-bench correctness."
129
- },
130
- "samples": [
131
- {
132
- "candidate": "zhivex",
133
- "instanceId": "django__django-15851",
134
- "repetition": 0,
135
- "resolved": false,
136
- "gradingStatus": "completed",
137
- "usageComplete": false,
138
- "costUsd": null,
139
- "candidateCaptureFailure": null,
140
- "modelTimings": [
141
- {
142
- "durationMs": 2428.9315,
143
- "firstTokenMs": 2391.3381249999998,
144
- "completed": true
145
- },
146
- {
147
- "durationMs": 1581.7063750000007,
148
- "firstTokenMs": 1581.1455830000004,
149
- "completed": true
150
- },
151
- {
152
- "durationMs": 2339.406166999999,
153
- "firstTokenMs": 2338.6784580000003,
154
- "completed": true
155
- },
156
- {
157
- "durationMs": 3309.444125,
158
- "firstTokenMs": 3224.615791,
159
- "completed": true
160
- },
161
- {
162
- "durationMs": 11227.875250000001,
163
- "firstTokenMs": null,
164
- "completed": false
165
- }
166
- ],
167
- "stageTimings": [],
168
- "approvalTimings": [],
169
- "toolTimings": [
170
- {
171
- "name": "read_task",
172
- "calls": 1,
173
- "errors": 0,
174
- "totalMs": 0.14816700000028504,
175
- "maxMs": 0.14816700000028504
176
- },
177
- {
178
- "name": "list_files",
179
- "calls": 1,
180
- "errors": 0,
181
- "totalMs": 360.7721250000004,
182
- "maxMs": 360.7721250000004
183
- },
184
- {
185
- "name": "search_files",
186
- "calls": 3,
187
- "errors": 0,
188
- "totalMs": 465.0301239999999,
189
- "maxMs": 436.10212500000034
190
- },
191
- {
192
- "name": "read_files",
193
- "calls": 1,
194
- "errors": 0,
195
- "totalMs": 2.266250000000582,
196
- "maxMs": 2.266250000000582
197
- },
198
- {
199
- "name": "repair_plan",
200
- "calls": 1,
201
- "errors": 0,
202
- "totalMs": 0.10829199999898265,
203
- "maxMs": 0.10829199999898265
204
- }
205
- ],
206
- "schemaVersion": 1,
207
- "status": "failed",
208
- "failure": "EXECUTION_FAILED",
209
- "phase": "agent",
210
- "durationMs": 29061.879583,
211
- "approvals": 0,
212
- "turns": [
213
- {
214
- "index": 1,
215
- "status": "completed",
216
- "finishReason": "tool-calls",
217
- "requestedTools": [
218
- "read_task"
219
- ],
220
- "inputTokens": 3015,
221
- "outputTokens": 87,
222
- "cachedInputTokens": 3012,
223
- "requestCharacters": 1439,
224
- "tools": [
225
- {
226
- "name": "read_task",
227
- "isError": false,
228
- "characters": 1147
229
- }
230
- ]
231
- },
232
- {
233
- "index": 2,
234
- "status": "completed",
235
- "finishReason": "tool-calls",
236
- "requestedTools": [
237
- "list_files",
238
- "search_files"
239
- ],
240
- "inputTokens": 3436,
241
- "outputTokens": 80,
242
- "cachedInputTokens": 2476,
243
- "requestCharacters": 12188,
244
- "tools": [
245
- {
246
- "name": "list_files",
247
- "isError": false,
248
- "characters": 9769
249
- },
250
- {
251
- "name": "search_files",
252
- "isError": false,
253
- "characters": 4422
254
- }
255
- ]
256
- },
257
- {
258
- "index": 3,
259
- "status": "completed",
260
- "finishReason": "tool-calls",
261
- "requestedTools": [
262
- "read_files",
263
- "search_files",
264
- "search_files"
265
- ],
266
- "inputTokens": 7990,
267
- "outputTokens": 168,
268
- "cachedInputTokens": 2476,
269
- "requestCharacters": 27907,
270
- "tools": [
271
- {
272
- "name": "read_files",
273
- "isError": false,
274
- "characters": 10466
275
- },
276
- {
277
- "name": "search_files",
278
- "isError": false,
279
- "characters": 347
280
- },
281
- {
282
- "name": "search_files",
283
- "isError": false,
284
- "characters": 806
285
- }
286
- ]
287
- },
288
- {
289
- "index": 4,
290
- "status": "completed",
291
- "finishReason": "tool-calls",
292
- "requestedTools": [
293
- "repair_plan"
294
- ],
295
- "inputTokens": 7098,
296
- "outputTokens": 225,
297
- "cachedInputTokens": 2476,
298
- "requestCharacters": 1439,
299
- "tools": [
300
- {
301
- "name": "repair_plan",
302
- "isError": false,
303
- "characters": 965
304
- }
305
- ]
306
- }
307
- ],
308
- "inputTokens": 21539,
309
- "outputTokens": 560,
310
- "cachedInputTokens": 10440,
311
- "modelCalls": 5,
312
- "compactions": [
313
- {
314
- "beforeStep": 4,
315
- "estimatedTokensBefore": 10913,
316
- "estimatedTokensAfter": 6167
317
- }
318
- ],
319
- "terminalDiagnostic": {
320
- "code": "EXECUTION_FAILED",
321
- "category": "execution",
322
- "failureType": "parsing",
323
- "retryable": false,
324
- "chain": [
325
- {
326
- "validationCode": null,
327
- "providerReason": "incomplete_arguments",
328
- "validationIssues": [],
329
- "status": null,
330
- "errorClass": "Object",
331
- "providerHints": [],
332
- "fingerprint": "83aa171c6634719d8782be5fb7aa7816bb4f67d6611ebe92242c6ecb68472a43",
333
- "reason": "tool call"
334
- }
335
- ]
336
- },
337
- "thrownDiagnostic": {
338
- "code": "EXECUTION_FAILED",
339
- "category": "execution",
340
- "failureType": "parsing",
341
- "retryable": false,
342
- "chain": [
343
- {
344
- "validationCode": null,
345
- "providerReason": null,
346
- "validationIssues": [],
347
- "status": null,
348
- "errorClass": "HarnessExecutionError",
349
- "providerHints": [],
350
- "fingerprint": "83aa171c6634719d8782be5fb7aa7816bb4f67d6611ebe92242c6ecb68472a43",
351
- "reason": "tool call"
352
- },
353
- {
354
- "validationCode": null,
355
- "providerReason": "incomplete_arguments",
356
- "validationIssues": [],
357
- "status": null,
358
- "errorClass": "ProviderToolCallError",
359
- "providerHints": [],
360
- "fingerprint": "83aa171c6634719d8782be5fb7aa7816bb4f67d6611ebe92242c6ecb68472a43",
361
- "reason": "tool call"
362
- },
363
- {
364
- "validationCode": null,
365
- "providerReason": "incomplete_arguments",
366
- "validationIssues": [],
367
- "status": null,
368
- "errorClass": "ProviderToolCallError",
369
- "providerHints": [],
370
- "fingerprint": "83aa171c6634719d8782be5fb7aa7816bb4f67d6611ebe92242c6ecb68472a43",
371
- "reason": "tool call"
372
- }
373
- ]
374
- },
375
- "guardrails": [],
376
- "repairProgress": {
377
- "repeatedResults": 0,
378
- "suppressedResults": 0,
379
- "blockedBroadCalls": 0,
380
- "enteredClosure": false,
381
- "phase": "explore",
382
- "blockedPhaseCalls": 0,
383
- "planRecorded": true
384
- },
385
- "contextMetrics": [
386
- {
387
- "systemCharacters": 4724,
388
- "userCharacters": 2403,
389
- "assistantCharacters": 0,
390
- "toolResultCharacters": 0,
391
- "otherMessageCharacters": 0,
392
- "toolDefinitionCharacters": 9375,
393
- "unmeasuredToolDefinitions": 0
394
- },
395
- {
396
- "systemCharacters": 4724,
397
- "userCharacters": 2403,
398
- "assistantCharacters": 3751,
399
- "toolResultCharacters": 1304,
400
- "otherMessageCharacters": 0,
401
- "toolDefinitionCharacters": 9375,
402
- "unmeasuredToolDefinitions": 0
403
- },
404
- {
405
- "systemCharacters": 4724,
406
- "userCharacters": 2403,
407
- "assistantCharacters": 4958,
408
- "toolResultCharacters": 15813,
409
- "otherMessageCharacters": 0,
410
- "toolDefinitionCharacters": 9375,
411
- "unmeasuredToolDefinitions": 0
412
- },
413
- {
414
- "systemCharacters": 4724,
415
- "userCharacters": 1437,
416
- "assistantCharacters": 7838,
417
- "toolResultCharacters": 12097,
418
- "otherMessageCharacters": 0,
419
- "toolDefinitionCharacters": 9375,
420
- "unmeasuredToolDefinitions": 0
421
- },
422
- {
423
- "systemCharacters": 4724,
424
- "userCharacters": 1874,
425
- "assistantCharacters": 13247,
426
- "toolResultCharacters": 13221,
427
- "otherMessageCharacters": 0,
428
- "toolDefinitionCharacters": 9375,
429
- "unmeasuredToolDefinitions": 0
430
- }
431
- ],
432
- "omittedContextMeasurements": 0,
433
- "toolCalls": 7,
434
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
435
- "protectedFilesChanged": [],
436
- "gradingMethod": "empty-patch-no-solution",
437
- "officialResolved": false,
438
- "candidateCaptured": true,
439
- "candidatePatchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
440
- "candidateGradingStatus": "completed",
441
- "candidateResolved": false,
442
- "repairStages": {
443
- "candidatePresent": false,
444
- "importedPatchPresent": false,
445
- "verificationAttempted": false
446
- },
447
- "gradingDurationMs": 0.1657090033404529,
448
- "totalDurationMs": 39817.722584004514
449
- },
450
- {
451
- "candidate": "mini-swe-agent",
452
- "instanceId": "django__django-15851",
453
- "repetition": 0,
454
- "resolved": true,
455
- "gradingStatus": "completed",
456
- "usageComplete": true,
457
- "costUsd": null,
458
- "schemaVersion": 1,
459
- "status": "Submitted",
460
- "failure": null,
461
- "phase": "agent",
462
- "durationMs": 26246.54299998656,
463
- "approvals": 0,
464
- "toolCalls": 11,
465
- "inputTokens": 26794,
466
- "outputTokens": 1440,
467
- "cachedInputTokens": 23173,
468
- "modelCalls": 8,
469
- "patchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
470
- "protectedFilesChanged": [],
471
- "officialResolved": true,
472
- "candidateCaptured": true,
473
- "candidatePatchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
474
- "candidateGradingStatus": "completed",
475
- "candidateResolved": true,
476
- "repairStages": {
477
- "candidatePresent": true,
478
- "importedPatchPresent": null,
479
- "verificationAttempted": null
480
- },
481
- "gradingDurationMs": 24538.846707961056,
482
- "totalDurationMs": 52952.04270799877
483
- }
484
- ]
485
- },
486
- "qwen": {
487
- "model": "qwen3.8-flash",
488
- "limits": {
489
- "steps": 24,
490
- "outputPerTurn": 2048,
491
- "inputTokens": 100000,
492
- "outputTokens": 16000,
493
- "timeoutSeconds": 300,
494
- "memoryMb": 2048,
495
- "cpus": 2
496
- },
497
- "implementation": {
498
- "gitCommit": "41bd1cb4ab8481a7413f096a0817f612fe9806fe",
499
- "sources": {
500
- "bun.lock": "e370459197ce52c3961770f3c5ccaab7ff35ddbe641ef72e8bd61f87a8b8e3fe",
501
- "evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
502
- "package.json": "a4a2507e3ad876d72cad49a16e149192d33521fb4175bb40a3eb077ccfdf43a1",
503
- "scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
504
- "scripts/swebench/candidate.py": "f5e3410b91b950acfbe50fc7511891816a8fb1c223082397e2c6300cf1e866cf",
505
- "scripts/swebench/candidate.ts": "bf102becfcce0c9a3f0df0a8f1ce0e5aaea7db2d252f6d9454a8370c24e53982",
506
- "scripts/swebench/common.py": "ad790227dac3bec1f592c458c993efff4f9d3de8f0df39b33e3238b58b3f0b3b",
507
- "scripts/swebench/context-metrics.ts": "7b84da6789413aa634d997c5963ee6796584e06c3eafe5f68f424fb7e27815e9",
508
- "scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
509
- "scripts/swebench/model-budget.ts": "e46dd406ee5d58a051a15c7d7133513666292303d8c6534f861a6db0e711f47c",
510
- "scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
511
- "scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
512
- "scripts/swebench/report.py": "d0b8d45f32c48e0ba2cb977f50838065a497a0ecaf3b4243ed94139393ce4d6b",
513
- "scripts/swebench/run.py": "94dccd55dfe6f70d8f8f2da70331700ec85eddd958547e0c3afe3dc416c9802c",
514
- "scripts/swebench/telemetry.ts": "743c9baf273e9240d704d098b6bd5eea9e76ef3e49ae636aaafa9bfd7a85b45b",
515
- "scripts/swebench/test_candidate.py": "e77d16379b748baa2daf184a45ea62587d938b6a7d48dcf9f6061f8af21fafd9",
516
- "scripts/swebench/test_contracts.py": "902e966d0fbf8501dbb15e011849f160331c033cd2a5480f9920d0f365a08d03",
517
- "scripts/swebench/zhivex-driver.ts": "b17e937484a28cf7fc7728595b51900efa5015dfa37cf6c1d4f9322051fe1ece",
518
- "scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
519
- "src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
520
- "src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
521
- "src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
522
- "src/cli-options.ts": "f48d9be132fed38f2ddbcfecbd8222c664a0df2f20edc28ce8d1b9b1a1900863",
523
- "src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
524
- "src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
525
- "src/cli.ts": "5a4acb4695a968a58fbeac05c4bb89848a9653d6189590f0ff90e430ffa98d14",
526
- "src/compaction.ts": "9e06a0abeeef155377f40c160b21c56256bca95ec18c5ee897681a5fc73db169",
527
- "src/config.ts": "b1d3991e84add58603fb85b2a604b9b3eb56ae879d0fa30b0cb40545fa44edc1",
528
- "src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
529
- "src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
530
- "src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
531
- "src/context-metrics.ts": "3eb30f9db808f918ed8d73d4f8ac19a13acb09ed5c1ee70fdae05201714b9198",
532
- "src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
533
- "src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
534
- "src/execution-environment.ts": "9ce6c49afab26527d5296f0c11f106063590f7333300d58dd49d5a1ec1b4b6e9",
535
- "src/file-security.ts": "b8c9f4988c4339c73a60ab4fa8d394b3eecc19117a0c61a8d19ab6483ca253df",
536
- "src/harness.ts": "420694110f2e205322bdcf06f3571e289809c96df92ad8bbda042d218b0692e1",
537
- "src/index.ts": "4bbfcd11376d48c4f1c1ec17d6b5c9f51baa9331fa999432be178bae94530a5e",
538
- "src/json-contracts.ts": "0780145636b47829a1aa6087c574d5654eb31b7534ea7d9c38c188d648c90272",
539
- "src/mcp.ts": "7722de031921c620ec03b9e42a6596ef8415aa8e4bb5f93ea84d2427c466e554",
540
- "src/model-budget.ts": "e25eca12f998998acb2e03f808ed0d84ba69b97711fc577716f230244c62aa80",
541
- "src/operations.ts": "bcccaf4d1d0ec117c2ed35df92185564d33399e5fdc5c12225ac61f14fd0b4b1",
542
- "src/orchestration.ts": "b56c8c7d526a99b745cc17769e0a1b7b500b25920e77f34aa0dd01c9758de319",
543
- "src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
544
- "src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
545
- "src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
546
- "src/repair-controller.ts": "91150ee614ab1444b6a89342f4b327fe2d626879467f879c6cc3313b8173f69a",
547
- "src/repair-progress.ts": "3550509a2a6c1e4a2e091c1012d5fb63ad8512a06f09b41e6b3a915a9a2b31a3",
548
- "src/repair-verifier.ts": "b70fb11feee8cdc7933947ccac0b78f8bb5c0eaccc3b29ad527197d77f40ffc9",
549
- "src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
550
- "src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
551
- "src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
552
- "src/runtime-checkpoints.ts": "a9b555d4c6b5f369f86a3d79ec1a3aab45886282cc7d0ef211aa084ee17e3897",
553
- "src/runtime-diagnostics.ts": "e1e5a3a558fa2aaeb477335dc336551f022c1234fe5e742eefc6599c8803e364",
554
- "src/runtime-policy.ts": "8356acf0102f0e73839b81ed628a5c0d9a89d9c21b0ca416efedde8f1b3414e2",
555
- "src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
556
- "src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
557
- "src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
558
- "src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
559
- "src/task-memory.ts": "5b62c863be5a23c33608633cd7aac0464f5536d01cba930aca58b11870170731",
560
- "src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
561
- "src/terminal-ui.ts": "4da1878a76ba608a2517bf5d5c992a9a88702aba3fd800a9980129ba93e157bc",
562
- "src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
563
- "src/tool-registry.ts": "7e90fff11940e525883c9ec0d2adbea2e6616f8bd9bc4ec77e1b7277e4217f2c",
564
- "src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
565
- "src/workspace-mutation-lock.ts": "ca201242db16173bd3a4d86c5d8bb0c2b6f3c43bfed3e4fa4ff5f06a75b19e80",
566
- "src/workspace.ts": "809f8bbfcf72fed10d860c48585089ca65bb3ef7c672ee6df3c918907a968302"
567
- }
568
- },
569
- "report": {
570
- "schemaVersion": 1,
571
- "kind": "swebench-comparison",
572
- "manifestSha256": "90a4670519c4d4c92e6705cc963e56319249a125b71127d855859d470c485626",
573
- "complete": true,
574
- "candidates": {
575
- "zhivex": {
576
- "planned": 1,
577
- "recorded": 1,
578
- "resolved": 0,
579
- "candidateCaptured": 1,
580
- "candidateResolved": 0,
581
- "candidateGraded": 1,
582
- "resolutionRate": 0.0,
583
- "missing": 0,
584
- "gradingFailures": 0,
585
- "usageComplete": true,
586
- "totalCostUsd": null,
587
- "costPerResolvedUsd": null,
588
- "resolvedP50Ms": null
589
- },
590
- "mini-swe-agent": {
591
- "planned": 1,
592
- "recorded": 1,
593
- "resolved": 1,
594
- "candidateCaptured": 1,
595
- "candidateResolved": 1,
596
- "candidateGraded": 1,
597
- "resolutionRate": 1.0,
598
- "missing": 0,
599
- "gradingFailures": 0,
600
- "usageComplete": true,
601
- "totalCostUsd": null,
602
- "costPerResolvedUsd": null,
603
- "resolvedP50Ms": 63600.59749998618
604
- }
605
- },
606
- "pairedTaskResolutionDifference": -1.0,
607
- "pairedTaskBootstrap95": null,
608
- "evidenceBoundary": "Local matched-model comparison. Missing and failed runs remain in the denominator. Small-sample intervals are exploratory. No safety superiority is inferred from SWE-bench correctness."
609
- },
610
- "samples": [
611
- {
612
- "candidate": "zhivex",
613
- "instanceId": "django__django-15851",
614
- "repetition": 0,
615
- "resolved": false,
616
- "gradingStatus": "completed",
617
- "usageComplete": true,
618
- "costUsd": null,
619
- "candidateCaptureFailure": null,
620
- "modelTimings": [
621
- {
622
- "durationMs": 3411.033667,
623
- "firstTokenMs": 3410.5682920000004,
624
- "completed": true
625
- },
626
- {
627
- "durationMs": 1952.283875,
628
- "firstTokenMs": 1952.1599579999993,
629
- "completed": true
630
- },
631
- {
632
- "durationMs": 4325.639832999999,
633
- "firstTokenMs": 4325.256958,
634
- "completed": true
635
- },
636
- {
637
- "durationMs": 1925.3485,
638
- "firstTokenMs": 1925.0503329999992,
639
- "completed": true
640
- },
641
- {
642
- "durationMs": 1737.0970420000012,
643
- "firstTokenMs": 1736.953125,
644
- "completed": true
645
- },
646
- {
647
- "durationMs": 2770.967667000001,
648
- "firstTokenMs": 2770.771917000002,
649
- "completed": true
650
- },
651
- {
652
- "durationMs": 2163.2452090000006,
653
- "firstTokenMs": 2163.001124999999,
654
- "completed": true
655
- },
656
- {
657
- "durationMs": 1771.3451659999992,
658
- "firstTokenMs": 1771.2574999999997,
659
- "completed": true
660
- },
661
- {
662
- "durationMs": 1638.9994159999987,
663
- "firstTokenMs": 1383.3052080000016,
664
- "completed": true
665
- }
666
- ],
667
- "stageTimings": [],
668
- "approvalTimings": [],
669
- "toolTimings": [
670
- {
671
- "name": "search_files",
672
- "calls": 3,
673
- "errors": 1,
674
- "totalMs": 942.8500009999998,
675
- "maxMs": 560.5733340000006
676
- },
677
- {
678
- "name": "list_files",
679
- "calls": 1,
680
- "errors": 0,
681
- "totalMs": 8.758624999999483,
682
- "maxMs": 8.758624999999483
683
- },
684
- {
685
- "name": "read_files",
686
- "calls": 4,
687
- "errors": 0,
688
- "totalMs": 2.655624999993961,
689
- "maxMs": 0.980874999999287
690
- },
691
- {
692
- "name": "repair_plan",
693
- "calls": 1,
694
- "errors": 0,
695
- "totalMs": 0.09883300000001327,
696
- "maxMs": 0.09883300000001327
697
- }
698
- ],
699
- "schemaVersion": 1,
700
- "status": "completed",
701
- "failure": null,
702
- "phase": "agent",
703
- "durationMs": 30056.868917,
704
- "approvals": 0,
705
- "turns": [
706
- {
707
- "index": 1,
708
- "status": "completed",
709
- "finishReason": "tool-calls",
710
- "requestedTools": [
711
- "search_files",
712
- "list_files"
713
- ],
714
- "inputTokens": 4558,
715
- "outputTokens": 123,
716
- "cachedInputTokens": 0,
717
- "requestCharacters": 7131,
718
- "tools": [
719
- {
720
- "name": "search_files",
721
- "isError": false,
722
- "characters": 714
723
- },
724
- {
725
- "name": "list_files",
726
- "isError": false,
727
- "characters": 452
728
- }
729
- ]
730
- },
731
- {
732
- "index": 2,
733
- "status": "completed",
734
- "finishReason": "tool-calls",
735
- "requestedTools": [
736
- "read_files"
737
- ],
738
- "inputTokens": 5127,
739
- "outputTokens": 64,
740
- "cachedInputTokens": 4096,
741
- "requestCharacters": 9009,
742
- "tools": [
743
- {
744
- "name": "read_files",
745
- "isError": false,
746
- "characters": 3979
747
- }
748
- ]
749
- },
750
- {
751
- "index": 3,
752
- "status": "completed",
753
- "finishReason": "tool-calls",
754
- "requestedTools": [
755
- "repair_plan"
756
- ],
757
- "inputTokens": 6484,
758
- "outputTokens": 257,
759
- "cachedInputTokens": 4096,
760
- "requestCharacters": 13413,
761
- "tools": [
762
- {
763
- "name": "repair_plan",
764
- "isError": false,
765
- "characters": 901
766
- }
767
- ]
768
- },
769
- {
770
- "index": 4,
771
- "status": "completed",
772
- "finishReason": "tool-calls",
773
- "requestedTools": [
774
- "search_files"
775
- ],
776
- "inputTokens": 7019,
777
- "outputTokens": 69,
778
- "cachedInputTokens": 5120,
779
- "requestCharacters": 15784,
780
- "tools": [
781
- {
782
- "name": "search_files",
783
- "isError": true,
784
- "characters": 4,
785
- "diagnostic": {
786
- "code": "EXECUTION_FAILED",
787
- "category": "execution",
788
- "failureType": "execution",
789
- "retryable": false,
790
- "chain": [
791
- {
792
- "validationCode": null,
793
- "providerReason": null,
794
- "validationIssues": [],
795
- "status": null,
796
- "errorClass": "Object",
797
- "providerHints": [],
798
- "fingerprint": "c6395153a3284cde8b6adc873ed023b81569a76e6fd3dde15f1a78f68fd78247",
799
- "reason": "enoent"
800
- }
801
- ]
802
- }
803
- }
804
- ]
805
- },
806
- {
807
- "index": 5,
808
- "status": "completed",
809
- "finishReason": "tool-calls",
810
- "requestedTools": [
811
- "search_files"
812
- ],
813
- "inputTokens": 7200,
814
- "outputTokens": 65,
815
- "cachedInputTokens": 6144,
816
- "requestCharacters": 16390,
817
- "tools": [
818
- {
819
- "name": "search_files",
820
- "isError": false,
821
- "characters": 6651
822
- }
823
- ]
824
- },
825
- {
826
- "index": 6,
827
- "status": "completed",
828
- "finishReason": "tool-calls",
829
- "requestedTools": [
830
- "read_files"
831
- ],
832
- "inputTokens": 10376,
833
- "outputTokens": 44,
834
- "cachedInputTokens": 6144,
835
- "requestCharacters": 23414,
836
- "tools": [
837
- {
838
- "name": "read_files",
839
- "isError": false,
840
- "characters": 4790
841
- }
842
- ]
843
- },
844
- {
845
- "index": 7,
846
- "status": "completed",
847
- "finishReason": "tool-calls",
848
- "requestedTools": [
849
- "read_files"
850
- ],
851
- "inputTokens": 11869,
852
- "outputTokens": 55,
853
- "cachedInputTokens": 9216,
854
- "requestCharacters": 28563,
855
- "tools": [
856
- {
857
- "name": "read_files",
858
- "isError": false,
859
- "characters": 3399
860
- }
861
- ]
862
- },
863
- {
864
- "index": 8,
865
- "status": "completed",
866
- "finishReason": "tool-calls",
867
- "requestedTools": [
868
- "read_files"
869
- ],
870
- "inputTokens": 6762,
871
- "outputTokens": 80,
872
- "cachedInputTokens": 3328,
873
- "requestCharacters": 14188,
874
- "tools": [
875
- {
876
- "name": "read_files",
877
- "isError": false,
878
- "characters": 5074
879
- }
880
- ]
881
- },
882
- {
883
- "index": 9,
884
- "status": "completed",
885
- "finishReason": "stop",
886
- "requestedTools": [],
887
- "inputTokens": 8446,
888
- "outputTokens": 12,
889
- "cachedInputTokens": 6144,
890
- "requestCharacters": 19712,
891
- "tools": []
892
- }
893
- ],
894
- "inputTokens": 67841,
895
- "outputTokens": 769,
896
- "cachedInputTokens": 44288,
897
- "modelCalls": 9,
898
- "compactions": [
899
- {
900
- "beforeStep": 8,
901
- "estimatedTokensBefore": 7646,
902
- "estimatedTokensAfter": 3109
903
- }
904
- ],
905
- "terminalDiagnostic": null,
906
- "thrownDiagnostic": null,
907
- "guardrails": [],
908
- "repairProgress": {
909
- "repeatedResults": 0,
910
- "suppressedResults": 0,
911
- "blockedBroadCalls": 0,
912
- "enteredClosure": false,
913
- "phase": "explore",
914
- "blockedPhaseCalls": 0,
915
- "planRecorded": true
916
- },
917
- "contextMetrics": [
918
- {
919
- "systemCharacters": 4724,
920
- "userCharacters": 2403,
921
- "assistantCharacters": 0,
922
- "toolResultCharacters": 0,
923
- "otherMessageCharacters": 0,
924
- "toolDefinitionCharacters": 9375,
925
- "unmeasuredToolDefinitions": 0
926
- },
927
- {
928
- "systemCharacters": 4724,
929
- "userCharacters": 2403,
930
- "assistantCharacters": 391,
931
- "toolResultCharacters": 1484,
932
- "otherMessageCharacters": 0,
933
- "toolDefinitionCharacters": 9375,
934
- "unmeasuredToolDefinitions": 0
935
- },
936
- {
937
- "systemCharacters": 4724,
938
- "userCharacters": 2403,
939
- "assistantCharacters": 656,
940
- "toolResultCharacters": 5621,
941
- "otherMessageCharacters": 0,
942
- "toolDefinitionCharacters": 9375,
943
- "unmeasuredToolDefinitions": 0
944
- },
945
- {
946
- "systemCharacters": 4724,
947
- "userCharacters": 2718,
948
- "assistantCharacters": 1650,
949
- "toolResultCharacters": 6681,
950
- "otherMessageCharacters": 0,
951
- "toolDefinitionCharacters": 9375,
952
- "unmeasuredToolDefinitions": 0
953
- },
954
- {
955
- "systemCharacters": 4724,
956
- "userCharacters": 2718,
957
- "assistantCharacters": 1883,
958
- "toolResultCharacters": 7052,
959
- "otherMessageCharacters": 0,
960
- "toolDefinitionCharacters": 9375,
961
- "unmeasuredToolDefinitions": 0
962
- },
963
- {
964
- "systemCharacters": 4724,
965
- "userCharacters": 2718,
966
- "assistantCharacters": 2094,
967
- "toolResultCharacters": 13863,
968
- "otherMessageCharacters": 0,
969
- "toolDefinitionCharacters": 9375,
970
- "unmeasuredToolDefinitions": 0
971
- },
972
- {
973
- "systemCharacters": 4724,
974
- "userCharacters": 2718,
975
- "assistantCharacters": 2293,
976
- "toolResultCharacters": 18811,
977
- "otherMessageCharacters": 0,
978
- "toolDefinitionCharacters": 9375,
979
- "unmeasuredToolDefinitions": 0
980
- },
981
- {
982
- "systemCharacters": 4724,
983
- "userCharacters": 1752,
984
- "assistantCharacters": 4149,
985
- "toolResultCharacters": 3557,
986
- "otherMessageCharacters": 0,
987
- "toolDefinitionCharacters": 9375,
988
- "unmeasuredToolDefinitions": 0
989
- },
990
- {
991
- "systemCharacters": 4724,
992
- "userCharacters": 1752,
993
- "assistantCharacters": 4439,
994
- "toolResultCharacters": 8789,
995
- "otherMessageCharacters": 0,
996
- "toolDefinitionCharacters": 9375,
997
- "unmeasuredToolDefinitions": 0
998
- }
999
- ],
1000
- "omittedContextMeasurements": 0,
1001
- "toolCalls": 9,
1002
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1003
- "protectedFilesChanged": [],
1004
- "gradingMethod": "empty-patch-no-solution",
1005
- "officialResolved": false,
1006
- "candidateCaptured": true,
1007
- "candidatePatchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1008
- "candidateGradingStatus": "completed",
1009
- "candidateResolved": false,
1010
- "repairStages": {
1011
- "candidatePresent": false,
1012
- "importedPatchPresent": false,
1013
- "verificationAttempted": false
1014
- },
1015
- "gradingDurationMs": 0.16479199985042214,
1016
- "totalDurationMs": 40870.69120799424
1017
- },
1018
- {
1019
- "candidate": "mini-swe-agent",
1020
- "instanceId": "django__django-15851",
1021
- "repetition": 0,
1022
- "resolved": true,
1023
- "gradingStatus": "completed",
1024
- "usageComplete": true,
1025
- "costUsd": null,
1026
- "schemaVersion": 1,
1027
- "status": "Submitted",
1028
- "failure": null,
1029
- "phase": "agent",
1030
- "durationMs": 37058.857541996986,
1031
- "approvals": 0,
1032
- "toolCalls": 14,
1033
- "inputTokens": 43085,
1034
- "outputTokens": 913,
1035
- "cachedInputTokens": 33024,
1036
- "modelCalls": 14,
1037
- "patchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
1038
- "protectedFilesChanged": [],
1039
- "officialResolved": true,
1040
- "candidateCaptured": true,
1041
- "candidatePatchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
1042
- "candidateGradingStatus": "completed",
1043
- "candidateResolved": true,
1044
- "repairStages": {
1045
- "candidatePresent": true,
1046
- "importedPatchPresent": null,
1047
- "verificationAttempted": null
1048
- },
1049
- "gradingDurationMs": 24455.263457959518,
1050
- "totalDurationMs": 63600.59749998618
1051
- }
1052
- ]
1053
- }
1054
- }
1055
- }