@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
@@ -1,1313 +0,0 @@
1
- {
2
- "schemaVersion": 1,
3
- "kind": "development-comparison",
4
- "date": "2026-09-19",
5
- "classification": "known development task; not holdout",
6
- "providers": {
7
- "openai": {
8
- "model": "gpt-5.6-luna",
9
- "limits": {
10
- "steps": 24,
11
- "outputPerTurn": 2048,
12
- "inputTokens": 100000,
13
- "outputTokens": 16000,
14
- "timeoutSeconds": 300,
15
- "memoryMb": 2048,
16
- "cpus": 2
17
- },
18
- "implementation": {
19
- "gitCommit": "28723ff3350b7572d5bd68f2d40b6feef1e0e6b4",
20
- "sources": {
21
- "bun.lock": "e370459197ce52c3961770f3c5ccaab7ff35ddbe641ef72e8bd61f87a8b8e3fe",
22
- "evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
23
- "package.json": "a4a2507e3ad876d72cad49a16e149192d33521fb4175bb40a3eb077ccfdf43a1",
24
- "scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
25
- "scripts/swebench/candidate.py": "f5e3410b91b950acfbe50fc7511891816a8fb1c223082397e2c6300cf1e866cf",
26
- "scripts/swebench/candidate.ts": "bf102becfcce0c9a3f0df0a8f1ce0e5aaea7db2d252f6d9454a8370c24e53982",
27
- "scripts/swebench/common.py": "ad790227dac3bec1f592c458c993efff4f9d3de8f0df39b33e3238b58b3f0b3b",
28
- "scripts/swebench/context-metrics.ts": "7b84da6789413aa634d997c5963ee6796584e06c3eafe5f68f424fb7e27815e9",
29
- "scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
30
- "scripts/swebench/model-budget.ts": "e46dd406ee5d58a051a15c7d7133513666292303d8c6534f861a6db0e711f47c",
31
- "scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
32
- "scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
33
- "scripts/swebench/report.py": "d0b8d45f32c48e0ba2cb977f50838065a497a0ecaf3b4243ed94139393ce4d6b",
34
- "scripts/swebench/run.py": "94dccd55dfe6f70d8f8f2da70331700ec85eddd958547e0c3afe3dc416c9802c",
35
- "scripts/swebench/telemetry.ts": "743c9baf273e9240d704d098b6bd5eea9e76ef3e49ae636aaafa9bfd7a85b45b",
36
- "scripts/swebench/test_candidate.py": "e77d16379b748baa2daf184a45ea62587d938b6a7d48dcf9f6061f8af21fafd9",
37
- "scripts/swebench/test_contracts.py": "902e966d0fbf8501dbb15e011849f160331c033cd2a5480f9920d0f365a08d03",
38
- "scripts/swebench/zhivex-driver.ts": "b17e937484a28cf7fc7728595b51900efa5015dfa37cf6c1d4f9322051fe1ece",
39
- "scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
40
- "src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
41
- "src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
42
- "src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
43
- "src/cli-options.ts": "f48d9be132fed38f2ddbcfecbd8222c664a0df2f20edc28ce8d1b9b1a1900863",
44
- "src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
45
- "src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
46
- "src/cli.ts": "5a4acb4695a968a58fbeac05c4bb89848a9653d6189590f0ff90e430ffa98d14",
47
- "src/compaction.ts": "9e06a0abeeef155377f40c160b21c56256bca95ec18c5ee897681a5fc73db169",
48
- "src/config.ts": "b1d3991e84add58603fb85b2a604b9b3eb56ae879d0fa30b0cb40545fa44edc1",
49
- "src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
50
- "src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
51
- "src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
52
- "src/context-metrics.ts": "3eb30f9db808f918ed8d73d4f8ac19a13acb09ed5c1ee70fdae05201714b9198",
53
- "src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
54
- "src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
55
- "src/execution-environment.ts": "9ce6c49afab26527d5296f0c11f106063590f7333300d58dd49d5a1ec1b4b6e9",
56
- "src/file-security.ts": "b8c9f4988c4339c73a60ab4fa8d394b3eecc19117a0c61a8d19ab6483ca253df",
57
- "src/harness.ts": "6168cfd53557b8fc07261fef1a4f5a6d5e689a169256c0e03c5abce0dfc228d0",
58
- "src/index.ts": "4bbfcd11376d48c4f1c1ec17d6b5c9f51baa9331fa999432be178bae94530a5e",
59
- "src/json-contracts.ts": "0780145636b47829a1aa6087c574d5654eb31b7534ea7d9c38c188d648c90272",
60
- "src/mcp.ts": "7722de031921c620ec03b9e42a6596ef8415aa8e4bb5f93ea84d2427c466e554",
61
- "src/model-budget.ts": "e5a5d3dd8c97df7f18f36ce0a82ed250e4a0c7d49b12580a7393db6ff2183a55",
62
- "src/operations.ts": "bcccaf4d1d0ec117c2ed35df92185564d33399e5fdc5c12225ac61f14fd0b4b1",
63
- "src/orchestration.ts": "b56c8c7d526a99b745cc17769e0a1b7b500b25920e77f34aa0dd01c9758de319",
64
- "src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
65
- "src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
66
- "src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
67
- "src/repair-controller.ts": "6eb48cf207cc56b18809cd762872253c95faeb6f47bc09c7fc2f1debd9c70b76",
68
- "src/repair-progress.ts": "38270f07c4e1d81d745aa3a01837cea4420b7caf310df59c7447e56b1a3e0440",
69
- "src/repair-verifier.ts": "b70fb11feee8cdc7933947ccac0b78f8bb5c0eaccc3b29ad527197d77f40ffc9",
70
- "src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
71
- "src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
72
- "src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
73
- "src/runtime-checkpoints.ts": "a9b555d4c6b5f369f86a3d79ec1a3aab45886282cc7d0ef211aa084ee17e3897",
74
- "src/runtime-diagnostics.ts": "e1e5a3a558fa2aaeb477335dc336551f022c1234fe5e742eefc6599c8803e364",
75
- "src/runtime-policy.ts": "8356acf0102f0e73839b81ed628a5c0d9a89d9c21b0ca416efedde8f1b3414e2",
76
- "src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
77
- "src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
78
- "src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
79
- "src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
80
- "src/task-memory.ts": "5b62c863be5a23c33608633cd7aac0464f5536d01cba930aca58b11870170731",
81
- "src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
82
- "src/terminal-ui.ts": "4da1878a76ba608a2517bf5d5c992a9a88702aba3fd800a9980129ba93e157bc",
83
- "src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
84
- "src/tool-registry.ts": "7e90fff11940e525883c9ec0d2adbea2e6616f8bd9bc4ec77e1b7277e4217f2c",
85
- "src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
86
- "src/workspace-mutation-lock.ts": "ca201242db16173bd3a4d86c5d8bb0c2b6f3c43bfed3e4fa4ff5f06a75b19e80",
87
- "src/workspace.ts": "809f8bbfcf72fed10d860c48585089ca65bb3ef7c672ee6df3c918907a968302"
88
- }
89
- },
90
- "report": {
91
- "schemaVersion": 1,
92
- "kind": "swebench-comparison",
93
- "manifestSha256": "c1bc64a18ece966055ee31aeeea678dd64bdae80e8634a00c669ab1258be1666",
94
- "complete": true,
95
- "candidates": {
96
- "zhivex": {
97
- "planned": 1,
98
- "recorded": 1,
99
- "resolved": 0,
100
- "candidateCaptured": 1,
101
- "candidateResolved": 0,
102
- "candidateGraded": 1,
103
- "resolutionRate": 0.0,
104
- "missing": 0,
105
- "gradingFailures": 0,
106
- "usageComplete": true,
107
- "totalCostUsd": null,
108
- "costPerResolvedUsd": null,
109
- "resolvedP50Ms": null
110
- },
111
- "mini-swe-agent": {
112
- "planned": 1,
113
- "recorded": 1,
114
- "resolved": 1,
115
- "candidateCaptured": 1,
116
- "candidateResolved": 1,
117
- "candidateGraded": 1,
118
- "resolutionRate": 1.0,
119
- "missing": 0,
120
- "gradingFailures": 0,
121
- "usageComplete": true,
122
- "totalCostUsd": null,
123
- "costPerResolvedUsd": null,
124
- "resolvedP50Ms": 69569.50954097556
125
- }
126
- },
127
- "pairedTaskResolutionDifference": -1.0,
128
- "pairedTaskBootstrap95": null,
129
- "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."
130
- },
131
- "samples": [
132
- {
133
- "candidate": "zhivex",
134
- "instanceId": "django__django-15851",
135
- "repetition": 0,
136
- "resolved": false,
137
- "gradingStatus": "completed",
138
- "usageComplete": true,
139
- "costUsd": null,
140
- "candidateCaptureFailure": null,
141
- "modelTimings": [
142
- {
143
- "durationMs": 1788.7389169999997,
144
- "firstTokenMs": 1770.7552500000002,
145
- "completed": true
146
- },
147
- {
148
- "durationMs": 1350.9064159999998,
149
- "firstTokenMs": 1290.8159999999998,
150
- "completed": true
151
- },
152
- {
153
- "durationMs": 2235.265375,
154
- "firstTokenMs": 2174.7994580000004,
155
- "completed": true
156
- },
157
- {
158
- "durationMs": 1176.5293329999986,
159
- "firstTokenMs": 1111.1294579999994,
160
- "completed": true
161
- },
162
- {
163
- "durationMs": 1413.2802499999998,
164
- "firstTokenMs": 1339.3893750000007,
165
- "completed": true
166
- },
167
- {
168
- "durationMs": 1362.5271659999999,
169
- "firstTokenMs": 1294.7329580000005,
170
- "completed": true
171
- },
172
- {
173
- "durationMs": 1386.9577090000002,
174
- "firstTokenMs": 1346.4139169999999,
175
- "completed": true
176
- },
177
- {
178
- "durationMs": 1477.5612919999985,
179
- "firstTokenMs": 1409.4364999999998,
180
- "completed": true
181
- },
182
- {
183
- "durationMs": 4537.724417000001,
184
- "firstTokenMs": 4494.494042000002,
185
- "completed": true
186
- }
187
- ],
188
- "stageTimings": [],
189
- "approvalTimings": [],
190
- "toolTimings": [
191
- {
192
- "name": "list_files",
193
- "calls": 1,
194
- "errors": 0,
195
- "totalMs": 428.98029099999985,
196
- "maxMs": 428.98029099999985
197
- },
198
- {
199
- "name": "search_files",
200
- "calls": 3,
201
- "errors": 0,
202
- "totalMs": 466.7124989999993,
203
- "maxMs": 382.14387499999975
204
- },
205
- {
206
- "name": "read_task",
207
- "calls": 2,
208
- "errors": 0,
209
- "totalMs": 0.26041600000098697,
210
- "maxMs": 0.24233300000014424
211
- },
212
- {
213
- "name": "read_files",
214
- "calls": 2,
215
- "errors": 0,
216
- "totalMs": 2.3135010000005423,
217
- "maxMs": 1.5475839999999152
218
- },
219
- {
220
- "name": "repair_plan",
221
- "calls": 1,
222
- "errors": 0,
223
- "totalMs": 0.07899999999790452,
224
- "maxMs": 0.07899999999790452
225
- }
226
- ],
227
- "schemaVersion": 1,
228
- "status": "failed",
229
- "failure": "WORK_TOKEN_BUDGET",
230
- "phase": "agent",
231
- "durationMs": 25648.538833000002,
232
- "approvals": 0,
233
- "turns": [
234
- {
235
- "index": 1,
236
- "status": "completed",
237
- "finishReason": "tool-calls",
238
- "requestedTools": [
239
- "list_files"
240
- ],
241
- "inputTokens": 3015,
242
- "outputTokens": 42,
243
- "cachedInputTokens": 3012,
244
- "requestCharacters": 1439,
245
- "tools": [
246
- {
247
- "name": "list_files",
248
- "isError": false,
249
- "characters": 9769
250
- }
251
- ]
252
- },
253
- {
254
- "index": 2,
255
- "status": "completed",
256
- "finishReason": "tool-calls",
257
- "requestedTools": [
258
- "search_files"
259
- ],
260
- "inputTokens": 5756,
261
- "outputTokens": 45,
262
- "cachedInputTokens": 2476,
263
- "requestCharacters": 20541,
264
- "tools": [
265
- {
266
- "name": "search_files",
267
- "isError": false,
268
- "characters": 166
269
- }
270
- ]
271
- },
272
- {
273
- "index": 3,
274
- "status": "completed",
275
- "finishReason": "tool-calls",
276
- "requestedTools": [
277
- "search_files"
278
- ],
279
- "inputTokens": 5863,
280
- "outputTokens": 66,
281
- "cachedInputTokens": 2476,
282
- "requestCharacters": 24400,
283
- "tools": [
284
- {
285
- "name": "search_files",
286
- "isError": false,
287
- "characters": 1110
288
- }
289
- ]
290
- },
291
- {
292
- "index": 4,
293
- "status": "completed",
294
- "finishReason": "tool-calls",
295
- "requestedTools": [
296
- "read_task"
297
- ],
298
- "inputTokens": 6328,
299
- "outputTokens": 75,
300
- "cachedInputTokens": 2476,
301
- "requestCharacters": 29507,
302
- "tools": [
303
- {
304
- "name": "read_task",
305
- "isError": false,
306
- "characters": 1147
307
- }
308
- ]
309
- },
310
- {
311
- "index": 5,
312
- "status": "completed",
313
- "finishReason": "tool-calls",
314
- "requestedTools": [
315
- "read_files"
316
- ],
317
- "inputTokens": 6737,
318
- "outputTokens": 57,
319
- "cachedInputTokens": 2476,
320
- "requestCharacters": 34436,
321
- "tools": [
322
- {
323
- "name": "read_files",
324
- "isError": false,
325
- "characters": 3583
326
- }
327
- ]
328
- },
329
- {
330
- "index": 6,
331
- "status": "completed",
332
- "finishReason": "tool-calls",
333
- "requestedTools": [
334
- "search_files"
335
- ],
336
- "inputTokens": 7852,
337
- "outputTokens": 64,
338
- "cachedInputTokens": 2476,
339
- "requestCharacters": 39030,
340
- "tools": [
341
- {
342
- "name": "search_files",
343
- "isError": false,
344
- "characters": 4509
345
- }
346
- ]
347
- },
348
- {
349
- "index": 7,
350
- "status": "completed",
351
- "finishReason": "tool-calls",
352
- "requestedTools": [
353
- "read_task"
354
- ],
355
- "inputTokens": 5390,
356
- "outputTokens": 61,
357
- "cachedInputTokens": 2476,
358
- "requestCharacters": 1439,
359
- "tools": [
360
- {
361
- "name": "read_task",
362
- "isError": false,
363
- "characters": 1147
364
- }
365
- ]
366
- },
367
- {
368
- "index": 8,
369
- "status": "completed",
370
- "finishReason": "tool-calls",
371
- "requestedTools": [
372
- "read_files"
373
- ],
374
- "inputTokens": 5785,
375
- "outputTokens": 59,
376
- "cachedInputTokens": 2476,
377
- "requestCharacters": 19610,
378
- "tools": [
379
- {
380
- "name": "read_files",
381
- "isError": false,
382
- "characters": 10073
383
- }
384
- ]
385
- },
386
- {
387
- "index": 9,
388
- "status": "completed",
389
- "finishReason": "tool-calls",
390
- "requestedTools": [
391
- "repair_plan"
392
- ],
393
- "inputTokens": 8613,
394
- "outputTokens": 480,
395
- "cachedInputTokens": 2476,
396
- "requestCharacters": 30714,
397
- "tools": [
398
- {
399
- "name": "repair_plan",
400
- "isError": false,
401
- "characters": 1584
402
- }
403
- ]
404
- }
405
- ],
406
- "inputTokens": 55339,
407
- "outputTokens": 949,
408
- "cachedInputTokens": 22820,
409
- "modelCalls": 9,
410
- "compactions": [
411
- {
412
- "beforeStep": 7,
413
- "estimatedTokensBefore": 11520,
414
- "estimatedTokensAfter": 4038
415
- }
416
- ],
417
- "terminalDiagnostic": {
418
- "code": "EXECUTION_FAILED",
419
- "category": "execution",
420
- "failureType": "budget",
421
- "retryable": false,
422
- "chain": [
423
- {
424
- "validationCode": null,
425
- "providerReason": null,
426
- "validationIssues": [],
427
- "status": null,
428
- "errorClass": "Object",
429
- "providerHints": [],
430
- "fingerprint": "e28d443c2f098dca74c2ebb674bfac969cc7e5a8c1621de8718feec56bdb994a",
431
- "reason": "budget"
432
- }
433
- ]
434
- },
435
- "thrownDiagnostic": {
436
- "code": "EXECUTION_FAILED",
437
- "category": "execution",
438
- "failureType": "budget",
439
- "retryable": false,
440
- "chain": [
441
- {
442
- "validationCode": null,
443
- "providerReason": null,
444
- "validationIssues": [],
445
- "status": null,
446
- "errorClass": "HarnessExecutionError",
447
- "providerHints": [],
448
- "fingerprint": "e28d443c2f098dca74c2ebb674bfac969cc7e5a8c1621de8718feec56bdb994a",
449
- "reason": "budget"
450
- },
451
- {
452
- "validationCode": null,
453
- "providerReason": null,
454
- "validationIssues": [],
455
- "status": null,
456
- "errorClass": "Error",
457
- "providerHints": [],
458
- "fingerprint": "e28d443c2f098dca74c2ebb674bfac969cc7e5a8c1621de8718feec56bdb994a",
459
- "reason": "budget"
460
- }
461
- ]
462
- },
463
- "guardrails": [],
464
- "repairProgress": {
465
- "repeatedResults": 0,
466
- "suppressedResults": 0,
467
- "blockedBroadCalls": 0,
468
- "enteredClosure": false,
469
- "phase": "explore",
470
- "blockedPhaseCalls": 0,
471
- "planRecorded": true
472
- },
473
- "contextMetrics": [
474
- {
475
- "systemCharacters": 4724,
476
- "userCharacters": 2403,
477
- "assistantCharacters": 0,
478
- "toolResultCharacters": 0,
479
- "otherMessageCharacters": 0,
480
- "toolDefinitionCharacters": 9375,
481
- "unmeasuredToolDefinitions": 0
482
- },
483
- {
484
- "systemCharacters": 4724,
485
- "userCharacters": 2403,
486
- "assistantCharacters": 3481,
487
- "toolResultCharacters": 9927,
488
- "otherMessageCharacters": 0,
489
- "toolDefinitionCharacters": 9375,
490
- "unmeasuredToolDefinitions": 0
491
- },
492
- {
493
- "systemCharacters": 4724,
494
- "userCharacters": 2403,
495
- "assistantCharacters": 7012,
496
- "toolResultCharacters": 10253,
497
- "otherMessageCharacters": 0,
498
- "toolDefinitionCharacters": 9375,
499
- "unmeasuredToolDefinitions": 0
500
- },
501
- {
502
- "systemCharacters": 4724,
503
- "userCharacters": 2403,
504
- "assistantCharacters": 10847,
505
- "toolResultCharacters": 11523,
506
- "otherMessageCharacters": 0,
507
- "toolDefinitionCharacters": 9375,
508
- "unmeasuredToolDefinitions": 0
509
- },
510
- {
511
- "systemCharacters": 4724,
512
- "userCharacters": 2403,
513
- "assistantCharacters": 14470,
514
- "toolResultCharacters": 12827,
515
- "otherMessageCharacters": 0,
516
- "toolDefinitionCharacters": 9375,
517
- "unmeasuredToolDefinitions": 0
518
- },
519
- {
520
- "systemCharacters": 4724,
521
- "userCharacters": 2403,
522
- "assistantCharacters": 15321,
523
- "toolResultCharacters": 16568,
524
- "otherMessageCharacters": 0,
525
- "toolDefinitionCharacters": 9375,
526
- "unmeasuredToolDefinitions": 0
527
- },
528
- {
529
- "systemCharacters": 4724,
530
- "userCharacters": 1437,
531
- "assistantCharacters": 6753,
532
- "toolResultCharacters": 4669,
533
- "otherMessageCharacters": 0,
534
- "toolDefinitionCharacters": 9375,
535
- "unmeasuredToolDefinitions": 0
536
- },
537
- {
538
- "systemCharacters": 4724,
539
- "userCharacters": 1437,
540
- "assistantCharacters": 7468,
541
- "toolResultCharacters": 5973,
542
- "otherMessageCharacters": 0,
543
- "toolDefinitionCharacters": 9375,
544
- "unmeasuredToolDefinitions": 0
545
- },
546
- {
547
- "systemCharacters": 4724,
548
- "userCharacters": 1437,
549
- "assistantCharacters": 8339,
550
- "toolResultCharacters": 16204,
551
- "otherMessageCharacters": 0,
552
- "toolDefinitionCharacters": 9375,
553
- "unmeasuredToolDefinitions": 0
554
- }
555
- ],
556
- "omittedContextMeasurements": 0,
557
- "toolCalls": 9,
558
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
559
- "protectedFilesChanged": [],
560
- "gradingMethod": "empty-patch-no-solution",
561
- "officialResolved": false,
562
- "candidateCaptured": true,
563
- "candidatePatchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
564
- "candidateGradingStatus": "completed",
565
- "candidateResolved": false,
566
- "repairStages": {
567
- "candidatePresent": false,
568
- "importedPatchPresent": false,
569
- "verificationAttempted": false
570
- },
571
- "gradingDurationMs": 0.16658299136906862,
572
- "totalDurationMs": 37302.090500015765
573
- },
574
- {
575
- "candidate": "mini-swe-agent",
576
- "instanceId": "django__django-15851",
577
- "repetition": 0,
578
- "resolved": true,
579
- "gradingStatus": "completed",
580
- "usageComplete": true,
581
- "costUsd": null,
582
- "schemaVersion": 1,
583
- "status": "Submitted",
584
- "failure": null,
585
- "phase": "agent",
586
- "durationMs": 21712.25812495686,
587
- "approvals": 0,
588
- "toolCalls": 8,
589
- "inputTokens": 31269,
590
- "outputTokens": 1381,
591
- "cachedInputTokens": 24949,
592
- "modelCalls": 7,
593
- "patchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
594
- "protectedFilesChanged": [],
595
- "officialResolved": true,
596
- "candidateCaptured": true,
597
- "candidatePatchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
598
- "candidateGradingStatus": "completed",
599
- "candidateResolved": true,
600
- "repairStages": {
601
- "candidatePresent": true,
602
- "importedPatchPresent": null,
603
- "verificationAttempted": null
604
- },
605
- "gradingDurationMs": 40513.59941699775,
606
- "totalDurationMs": 69569.50954097556
607
- }
608
- ]
609
- },
610
- "qwen": {
611
- "model": "qwen3.8-flash",
612
- "limits": {
613
- "steps": 24,
614
- "outputPerTurn": 2048,
615
- "inputTokens": 100000,
616
- "outputTokens": 16000,
617
- "timeoutSeconds": 300,
618
- "memoryMb": 2048,
619
- "cpus": 2
620
- },
621
- "implementation": {
622
- "gitCommit": "28723ff3350b7572d5bd68f2d40b6feef1e0e6b4",
623
- "sources": {
624
- "bun.lock": "e370459197ce52c3961770f3c5ccaab7ff35ddbe641ef72e8bd61f87a8b8e3fe",
625
- "evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
626
- "package.json": "a4a2507e3ad876d72cad49a16e149192d33521fb4175bb40a3eb077ccfdf43a1",
627
- "scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
628
- "scripts/swebench/candidate.py": "f5e3410b91b950acfbe50fc7511891816a8fb1c223082397e2c6300cf1e866cf",
629
- "scripts/swebench/candidate.ts": "bf102becfcce0c9a3f0df0a8f1ce0e5aaea7db2d252f6d9454a8370c24e53982",
630
- "scripts/swebench/common.py": "ad790227dac3bec1f592c458c993efff4f9d3de8f0df39b33e3238b58b3f0b3b",
631
- "scripts/swebench/context-metrics.ts": "7b84da6789413aa634d997c5963ee6796584e06c3eafe5f68f424fb7e27815e9",
632
- "scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
633
- "scripts/swebench/model-budget.ts": "e46dd406ee5d58a051a15c7d7133513666292303d8c6534f861a6db0e711f47c",
634
- "scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
635
- "scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
636
- "scripts/swebench/report.py": "d0b8d45f32c48e0ba2cb977f50838065a497a0ecaf3b4243ed94139393ce4d6b",
637
- "scripts/swebench/run.py": "94dccd55dfe6f70d8f8f2da70331700ec85eddd958547e0c3afe3dc416c9802c",
638
- "scripts/swebench/telemetry.ts": "743c9baf273e9240d704d098b6bd5eea9e76ef3e49ae636aaafa9bfd7a85b45b",
639
- "scripts/swebench/test_candidate.py": "e77d16379b748baa2daf184a45ea62587d938b6a7d48dcf9f6061f8af21fafd9",
640
- "scripts/swebench/test_contracts.py": "902e966d0fbf8501dbb15e011849f160331c033cd2a5480f9920d0f365a08d03",
641
- "scripts/swebench/zhivex-driver.ts": "b17e937484a28cf7fc7728595b51900efa5015dfa37cf6c1d4f9322051fe1ece",
642
- "scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
643
- "src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
644
- "src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
645
- "src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
646
- "src/cli-options.ts": "f48d9be132fed38f2ddbcfecbd8222c664a0df2f20edc28ce8d1b9b1a1900863",
647
- "src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
648
- "src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
649
- "src/cli.ts": "5a4acb4695a968a58fbeac05c4bb89848a9653d6189590f0ff90e430ffa98d14",
650
- "src/compaction.ts": "9e06a0abeeef155377f40c160b21c56256bca95ec18c5ee897681a5fc73db169",
651
- "src/config.ts": "b1d3991e84add58603fb85b2a604b9b3eb56ae879d0fa30b0cb40545fa44edc1",
652
- "src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
653
- "src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
654
- "src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
655
- "src/context-metrics.ts": "3eb30f9db808f918ed8d73d4f8ac19a13acb09ed5c1ee70fdae05201714b9198",
656
- "src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
657
- "src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
658
- "src/execution-environment.ts": "9ce6c49afab26527d5296f0c11f106063590f7333300d58dd49d5a1ec1b4b6e9",
659
- "src/file-security.ts": "b8c9f4988c4339c73a60ab4fa8d394b3eecc19117a0c61a8d19ab6483ca253df",
660
- "src/harness.ts": "6168cfd53557b8fc07261fef1a4f5a6d5e689a169256c0e03c5abce0dfc228d0",
661
- "src/index.ts": "4bbfcd11376d48c4f1c1ec17d6b5c9f51baa9331fa999432be178bae94530a5e",
662
- "src/json-contracts.ts": "0780145636b47829a1aa6087c574d5654eb31b7534ea7d9c38c188d648c90272",
663
- "src/mcp.ts": "7722de031921c620ec03b9e42a6596ef8415aa8e4bb5f93ea84d2427c466e554",
664
- "src/model-budget.ts": "e5a5d3dd8c97df7f18f36ce0a82ed250e4a0c7d49b12580a7393db6ff2183a55",
665
- "src/operations.ts": "bcccaf4d1d0ec117c2ed35df92185564d33399e5fdc5c12225ac61f14fd0b4b1",
666
- "src/orchestration.ts": "b56c8c7d526a99b745cc17769e0a1b7b500b25920e77f34aa0dd01c9758de319",
667
- "src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
668
- "src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
669
- "src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
670
- "src/repair-controller.ts": "6eb48cf207cc56b18809cd762872253c95faeb6f47bc09c7fc2f1debd9c70b76",
671
- "src/repair-progress.ts": "38270f07c4e1d81d745aa3a01837cea4420b7caf310df59c7447e56b1a3e0440",
672
- "src/repair-verifier.ts": "b70fb11feee8cdc7933947ccac0b78f8bb5c0eaccc3b29ad527197d77f40ffc9",
673
- "src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
674
- "src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
675
- "src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
676
- "src/runtime-checkpoints.ts": "a9b555d4c6b5f369f86a3d79ec1a3aab45886282cc7d0ef211aa084ee17e3897",
677
- "src/runtime-diagnostics.ts": "e1e5a3a558fa2aaeb477335dc336551f022c1234fe5e742eefc6599c8803e364",
678
- "src/runtime-policy.ts": "8356acf0102f0e73839b81ed628a5c0d9a89d9c21b0ca416efedde8f1b3414e2",
679
- "src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
680
- "src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
681
- "src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
682
- "src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
683
- "src/task-memory.ts": "5b62c863be5a23c33608633cd7aac0464f5536d01cba930aca58b11870170731",
684
- "src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
685
- "src/terminal-ui.ts": "4da1878a76ba608a2517bf5d5c992a9a88702aba3fd800a9980129ba93e157bc",
686
- "src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
687
- "src/tool-registry.ts": "7e90fff11940e525883c9ec0d2adbea2e6616f8bd9bc4ec77e1b7277e4217f2c",
688
- "src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
689
- "src/workspace-mutation-lock.ts": "ca201242db16173bd3a4d86c5d8bb0c2b6f3c43bfed3e4fa4ff5f06a75b19e80",
690
- "src/workspace.ts": "809f8bbfcf72fed10d860c48585089ca65bb3ef7c672ee6df3c918907a968302"
691
- }
692
- },
693
- "report": {
694
- "schemaVersion": 1,
695
- "kind": "swebench-comparison",
696
- "manifestSha256": "234e5c319772f0bc67d7ce8bccd9caae512177a89524c3cc47e4c38ff4715df9",
697
- "complete": true,
698
- "candidates": {
699
- "zhivex": {
700
- "planned": 1,
701
- "recorded": 1,
702
- "resolved": 0,
703
- "candidateCaptured": 1,
704
- "candidateResolved": 0,
705
- "candidateGraded": 1,
706
- "resolutionRate": 0.0,
707
- "missing": 0,
708
- "gradingFailures": 0,
709
- "usageComplete": true,
710
- "totalCostUsd": null,
711
- "costPerResolvedUsd": null,
712
- "resolvedP50Ms": null
713
- },
714
- "mini-swe-agent": {
715
- "planned": 1,
716
- "recorded": 1,
717
- "resolved": 1,
718
- "candidateCaptured": 1,
719
- "candidateResolved": 1,
720
- "candidateGraded": 1,
721
- "resolutionRate": 1.0,
722
- "missing": 0,
723
- "gradingFailures": 0,
724
- "usageComplete": true,
725
- "totalCostUsd": null,
726
- "costPerResolvedUsd": null,
727
- "resolvedP50Ms": 49566.27295794897
728
- }
729
- },
730
- "pairedTaskResolutionDifference": -1.0,
731
- "pairedTaskBootstrap95": null,
732
- "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."
733
- },
734
- "samples": [
735
- {
736
- "candidate": "zhivex",
737
- "instanceId": "django__django-15851",
738
- "repetition": 0,
739
- "resolved": false,
740
- "gradingStatus": "completed",
741
- "usageComplete": true,
742
- "costUsd": null,
743
- "candidateCaptureFailure": null,
744
- "modelTimings": [
745
- {
746
- "durationMs": 13494.942957999998,
747
- "firstTokenMs": 11987.065499999999,
748
- "completed": true
749
- },
750
- {
751
- "durationMs": 1606.3987080000006,
752
- "firstTokenMs": 1606.1942500000005,
753
- "completed": true
754
- },
755
- {
756
- "durationMs": 2560.6657909999994,
757
- "firstTokenMs": 2560.4924999999967,
758
- "completed": true
759
- },
760
- {
761
- "durationMs": 2110.684583000002,
762
- "firstTokenMs": 2110.529958000003,
763
- "completed": true
764
- },
765
- {
766
- "durationMs": 3735.8417919999993,
767
- "firstTokenMs": 3735.701541999999,
768
- "completed": true
769
- },
770
- {
771
- "durationMs": 1547.485166999999,
772
- "firstTokenMs": 1547.2844579999983,
773
- "completed": true
774
- },
775
- {
776
- "durationMs": 2811.7602910000023,
777
- "firstTokenMs": 1323.8450410000005,
778
- "completed": true
779
- },
780
- {
781
- "durationMs": 1899.3515419999967,
782
- "firstTokenMs": 1899.1887919999936,
783
- "completed": true
784
- },
785
- {
786
- "durationMs": 2462.1707499999975,
787
- "firstTokenMs": 1522.6253339999967,
788
- "completed": true
789
- },
790
- {
791
- "durationMs": 6644.867083000005,
792
- "firstTokenMs": null,
793
- "completed": true
794
- }
795
- ],
796
- "stageTimings": [
797
- {
798
- "event": "edit-requested",
799
- "elapsedMs": 31548.041666
800
- },
801
- {
802
- "event": "verification-requested",
803
- "elapsedMs": 33600.492375
804
- },
805
- {
806
- "event": "governed-edit-completed",
807
- "elapsedMs": 33608.793041
808
- },
809
- {
810
- "event": "command-requested",
811
- "elapsedMs": 40905.695
812
- }
813
- ],
814
- "approvalTimings": [
815
- {
816
- "durationMs": 0.1386659999989206,
817
- "resolved": true
818
- },
819
- {
820
- "durationMs": 0.02570900000137044,
821
- "resolved": true
822
- },
823
- {
824
- "durationMs": 0.027290999998513144,
825
- "resolved": true
826
- },
827
- {
828
- "durationMs": 0.023874999998952262,
829
- "resolved": true
830
- }
831
- ],
832
- "toolTimings": [
833
- {
834
- "name": "search_files",
835
- "calls": 2,
836
- "errors": 0,
837
- "totalMs": 439.46991699999853,
838
- "maxMs": 378.27487499999916
839
- },
840
- {
841
- "name": "list_files",
842
- "calls": 1,
843
- "errors": 0,
844
- "totalMs": 16.088958000000275,
845
- "maxMs": 16.088958000000275
846
- },
847
- {
848
- "name": "read_files",
849
- "calls": 3,
850
- "errors": 0,
851
- "totalMs": 2.675290999995923,
852
- "maxMs": 1.3670829999973648
853
- },
854
- {
855
- "name": "repair_plan",
856
- "calls": 1,
857
- "errors": 0,
858
- "totalMs": 0.14245799999844166,
859
- "maxMs": 0.14245799999844166
860
- },
861
- {
862
- "name": "apply_reviewed_replacement",
863
- "calls": 1,
864
- "errors": 0,
865
- "totalMs": 6.89862500000163,
866
- "maxMs": 6.89862500000163
867
- },
868
- {
869
- "name": "verify_and_apply_environment_patch",
870
- "calls": 1,
871
- "errors": 1,
872
- "totalMs": 4237.259832999996,
873
- "maxMs": 4237.259832999996
874
- },
875
- {
876
- "name": "run_environment_command",
877
- "calls": 1,
878
- "errors": 0,
879
- "totalMs": 2762.7135409999973,
880
- "maxMs": 2762.7135409999973
881
- },
882
- {
883
- "name": "run_environment_shell",
884
- "calls": 1,
885
- "errors": 0,
886
- "totalMs": 11506.245166999994,
887
- "maxMs": 11506.245166999994
888
- }
889
- ],
890
- "schemaVersion": 1,
891
- "status": "failed",
892
- "failure": "AGENT_FAILED",
893
- "phase": "agent",
894
- "durationMs": 72587.34108299999,
895
- "approvals": 4,
896
- "turns": [
897
- {
898
- "index": 1,
899
- "status": "completed",
900
- "finishReason": "tool-calls",
901
- "requestedTools": [
902
- "search_files",
903
- "list_files"
904
- ],
905
- "inputTokens": 4558,
906
- "outputTokens": 124,
907
- "cachedInputTokens": 0,
908
- "requestCharacters": 7131,
909
- "tools": [
910
- {
911
- "name": "search_files",
912
- "isError": false,
913
- "characters": 158
914
- },
915
- {
916
- "name": "list_files",
917
- "isError": false,
918
- "characters": 452
919
- }
920
- ]
921
- },
922
- {
923
- "index": 2,
924
- "status": "completed",
925
- "finishReason": "tool-calls",
926
- "requestedTools": [
927
- "read_files"
928
- ],
929
- "inputTokens": 4860,
930
- "outputTokens": 39,
931
- "cachedInputTokens": 4096,
932
- "requestCharacters": 8541,
933
- "tools": [
934
- {
935
- "name": "read_files",
936
- "isError": false,
937
- "characters": 2673
938
- }
939
- ]
940
- },
941
- {
942
- "index": 3,
943
- "status": "completed",
944
- "finishReason": "tool-calls",
945
- "requestedTools": [
946
- "search_files"
947
- ],
948
- "inputTokens": 5781,
949
- "outputTokens": 65,
950
- "cachedInputTokens": 4096,
951
- "requestCharacters": 11566,
952
- "tools": [
953
- {
954
- "name": "search_files",
955
- "isError": false,
956
- "characters": 6524
957
- }
958
- ]
959
- },
960
- {
961
- "index": 4,
962
- "status": "completed",
963
- "finishReason": "tool-calls",
964
- "requestedTools": [
965
- "read_files"
966
- ],
967
- "inputTokens": 8958,
968
- "outputTokens": 53,
969
- "cachedInputTokens": 5120,
970
- "requestCharacters": 18467,
971
- "tools": [
972
- {
973
- "name": "read_files",
974
- "isError": false,
975
- "characters": 6380
976
- }
977
- ]
978
- },
979
- {
980
- "index": 5,
981
- "status": "completed",
982
- "finishReason": "tool-calls",
983
- "requestedTools": [
984
- "repair_plan"
985
- ],
986
- "inputTokens": 10965,
987
- "outputTokens": 261,
988
- "cachedInputTokens": 8320,
989
- "requestCharacters": 25220,
990
- "tools": [
991
- {
992
- "name": "repair_plan",
993
- "isError": false,
994
- "characters": 969
995
- }
996
- ]
997
- },
998
- {
999
- "index": 6,
1000
- "status": "completed",
1001
- "finishReason": "tool-calls",
1002
- "requestedTools": [
1003
- "read_files"
1004
- ],
1005
- "inputTokens": 11526,
1006
- "outputTokens": 55,
1007
- "cachedInputTokens": 10240,
1008
- "requestCharacters": 27775,
1009
- "tools": [
1010
- {
1011
- "name": "read_files",
1012
- "isError": false,
1013
- "characters": 2033
1014
- }
1015
- ]
1016
- },
1017
- {
1018
- "index": 7,
1019
- "status": "waiting_approval",
1020
- "finishReason": "tool-calls",
1021
- "requestedTools": [
1022
- "apply_reviewed_replacement"
1023
- ],
1024
- "inputTokens": 12251,
1025
- "outputTokens": 190,
1026
- "cachedInputTokens": 10240,
1027
- "requestCharacters": 30183,
1028
- "tools": []
1029
- },
1030
- {
1031
- "index": 8,
1032
- "status": "completed",
1033
- "finishReason": "tool-calls",
1034
- "requestedTools": [
1035
- "verify_and_apply_environment_patch"
1036
- ],
1037
- "inputTokens": 0,
1038
- "outputTokens": 0,
1039
- "cachedInputTokens": 0,
1040
- "requestCharacters": 11836,
1041
- "tools": [
1042
- {
1043
- "name": "apply_reviewed_replacement",
1044
- "isError": false,
1045
- "characters": 483
1046
- },
1047
- {
1048
- "name": "verify_and_apply_environment_patch",
1049
- "isError": true,
1050
- "characters": 276,
1051
- "exitCode": 1,
1052
- "timedOut": false,
1053
- "verificationOutcome": "failed",
1054
- "diagnostic": {
1055
- "code": "EXECUTION_FAILED",
1056
- "category": "execution",
1057
- "failureType": "execution",
1058
- "retryable": false,
1059
- "chain": [
1060
- {
1061
- "validationCode": null,
1062
- "providerReason": null,
1063
- "validationIssues": [],
1064
- "status": null,
1065
- "errorClass": "Object",
1066
- "providerHints": [],
1067
- "fingerprint": "d22955ecb1810cd7e960e732636eec570e04ab3b19eab276ebf23e2096a3c941",
1068
- "reason": null
1069
- }
1070
- ]
1071
- }
1072
- }
1073
- ]
1074
- },
1075
- {
1076
- "index": 9,
1077
- "status": "waiting_approval",
1078
- "finishReason": "tool-calls",
1079
- "requestedTools": [
1080
- "run_environment_command"
1081
- ],
1082
- "inputTokens": 6415,
1083
- "outputTokens": 67,
1084
- "cachedInputTokens": 3328,
1085
- "requestCharacters": 13127,
1086
- "tools": []
1087
- },
1088
- {
1089
- "index": 10,
1090
- "status": "waiting_approval",
1091
- "finishReason": "tool-calls",
1092
- "requestedTools": [
1093
- "run_environment_shell"
1094
- ],
1095
- "inputTokens": 6815,
1096
- "outputTokens": 97,
1097
- "cachedInputTokens": 5120,
1098
- "requestCharacters": 14347,
1099
- "tools": [
1100
- {
1101
- "name": "run_environment_command",
1102
- "isError": false,
1103
- "characters": 415,
1104
- "phaseLatencies": {
1105
- "commandAndAttestationMs": 451.550208,
1106
- "totalMs": 2762.550416,
1107
- "sessionCreationMs": 2175.600875
1108
- },
1109
- "exitCode": 1,
1110
- "timedOut": false
1111
- }
1112
- ]
1113
- },
1114
- {
1115
- "index": 11,
1116
- "status": "completed",
1117
- "finishReason": "stop",
1118
- "requestedTools": [],
1119
- "inputTokens": 5187,
1120
- "outputTokens": 533,
1121
- "cachedInputTokens": 0,
1122
- "requestCharacters": 16684,
1123
- "tools": []
1124
- }
1125
- ],
1126
- "inputTokens": 77316,
1127
- "outputTokens": 1484,
1128
- "cachedInputTokens": 50560,
1129
- "modelCalls": 10,
1130
- "compactions": [
1131
- {
1132
- "beforeStep": 8,
1133
- "estimatedTokensBefore": 7414,
1134
- "estimatedTokensAfter": 2491
1135
- }
1136
- ],
1137
- "terminalDiagnostic": {
1138
- "code": "EXECUTION_FAILED",
1139
- "category": "execution",
1140
- "failureType": "execution",
1141
- "retryable": false,
1142
- "chain": [
1143
- {
1144
- "validationCode": null,
1145
- "providerReason": null,
1146
- "validationIssues": [],
1147
- "status": null,
1148
- "errorClass": "Object",
1149
- "providerHints": [],
1150
- "fingerprint": "df4ad18f75f381d4abf1c3a9a9cb1ce99246204330b132a8d70938de039a0be4",
1151
- "reason": null
1152
- }
1153
- ]
1154
- },
1155
- "thrownDiagnostic": null,
1156
- "guardrails": [],
1157
- "repairProgress": {
1158
- "repeatedResults": 0,
1159
- "suppressedResults": 0,
1160
- "blockedBroadCalls": 0,
1161
- "enteredClosure": true,
1162
- "phase": "verify",
1163
- "blockedPhaseCalls": 0,
1164
- "planRecorded": true
1165
- },
1166
- "contextMetrics": [
1167
- {
1168
- "systemCharacters": 4724,
1169
- "userCharacters": 2403,
1170
- "assistantCharacters": 0,
1171
- "toolResultCharacters": 0,
1172
- "otherMessageCharacters": 0,
1173
- "toolDefinitionCharacters": 9375,
1174
- "unmeasuredToolDefinitions": 0
1175
- },
1176
- {
1177
- "systemCharacters": 4724,
1178
- "userCharacters": 2403,
1179
- "assistantCharacters": 479,
1180
- "toolResultCharacters": 928,
1181
- "otherMessageCharacters": 0,
1182
- "toolDefinitionCharacters": 9375,
1183
- "unmeasuredToolDefinitions": 0
1184
- },
1185
- {
1186
- "systemCharacters": 4724,
1187
- "userCharacters": 2403,
1188
- "assistantCharacters": 671,
1189
- "toolResultCharacters": 3759,
1190
- "otherMessageCharacters": 0,
1191
- "toolDefinitionCharacters": 9375,
1192
- "unmeasuredToolDefinitions": 0
1193
- },
1194
- {
1195
- "systemCharacters": 4724,
1196
- "userCharacters": 2403,
1197
- "assistantCharacters": 886,
1198
- "toolResultCharacters": 10443,
1199
- "otherMessageCharacters": 0,
1200
- "toolDefinitionCharacters": 9375,
1201
- "unmeasuredToolDefinitions": 0
1202
- },
1203
- {
1204
- "systemCharacters": 4724,
1205
- "userCharacters": 2403,
1206
- "assistantCharacters": 1099,
1207
- "toolResultCharacters": 16981,
1208
- "otherMessageCharacters": 0,
1209
- "toolDefinitionCharacters": 9375,
1210
- "unmeasuredToolDefinitions": 0
1211
- },
1212
- {
1213
- "systemCharacters": 4724,
1214
- "userCharacters": 2766,
1215
- "assistantCharacters": 2161,
1216
- "toolResultCharacters": 18109,
1217
- "otherMessageCharacters": 0,
1218
- "toolDefinitionCharacters": 9375,
1219
- "unmeasuredToolDefinitions": 0
1220
- },
1221
- {
1222
- "systemCharacters": 4724,
1223
- "userCharacters": 2766,
1224
- "assistantCharacters": 2376,
1225
- "toolResultCharacters": 20300,
1226
- "otherMessageCharacters": 0,
1227
- "toolDefinitionCharacters": 9375,
1228
- "unmeasuredToolDefinitions": 0
1229
- },
1230
- {
1231
- "systemCharacters": 4724,
1232
- "userCharacters": 2232,
1233
- "assistantCharacters": 4926,
1234
- "toolResultCharacters": 1237,
1235
- "otherMessageCharacters": 0,
1236
- "toolDefinitionCharacters": 9375,
1237
- "unmeasuredToolDefinitions": 0
1238
- },
1239
- {
1240
- "systemCharacters": 4724,
1241
- "userCharacters": 2603,
1242
- "assistantCharacters": 5187,
1243
- "toolResultCharacters": 1823,
1244
- "otherMessageCharacters": 0,
1245
- "toolDefinitionCharacters": 9375,
1246
- "unmeasuredToolDefinitions": 0
1247
- },
1248
- {
1249
- "systemCharacters": 4724,
1250
- "userCharacters": 2614,
1251
- "assistantCharacters": 5636,
1252
- "toolResultCharacters": 3698,
1253
- "otherMessageCharacters": 0,
1254
- "toolDefinitionCharacters": 1635,
1255
- "unmeasuredToolDefinitions": 0
1256
- }
1257
- ],
1258
- "omittedContextMeasurements": 0,
1259
- "toolCalls": 11,
1260
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1261
- "protectedFilesChanged": [],
1262
- "gradingMethod": "empty-patch-no-solution",
1263
- "officialResolved": false,
1264
- "candidateCaptured": true,
1265
- "candidatePatchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1266
- "candidateGradingStatus": "completed",
1267
- "candidateResolved": false,
1268
- "repairStages": {
1269
- "candidatePresent": false,
1270
- "importedPatchPresent": false,
1271
- "verificationAttempted": true
1272
- },
1273
- "gradingDurationMs": 0.16370904631912708,
1274
- "totalDurationMs": 84125.71037502494
1275
- },
1276
- {
1277
- "candidate": "mini-swe-agent",
1278
- "instanceId": "django__django-15851",
1279
- "repetition": 0,
1280
- "resolved": true,
1281
- "gradingStatus": "completed",
1282
- "usageComplete": true,
1283
- "costUsd": null,
1284
- "schemaVersion": 1,
1285
- "status": "Submitted",
1286
- "failure": null,
1287
- "phase": "agent",
1288
- "durationMs": 22791.735082981177,
1289
- "approvals": 0,
1290
- "toolCalls": 9,
1291
- "inputTokens": 21728,
1292
- "outputTokens": 654,
1293
- "cachedInputTokens": 14336,
1294
- "modelCalls": 9,
1295
- "patchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
1296
- "protectedFilesChanged": [],
1297
- "officialResolved": true,
1298
- "candidateCaptured": true,
1299
- "candidatePatchSha256": "873ee0e2d09108303f5ebe1fa4d0aec08ee9e4ce52b93941c70b4c8c91c21db4",
1300
- "candidateGradingStatus": "completed",
1301
- "candidateResolved": true,
1302
- "repairStages": {
1303
- "candidatePresent": true,
1304
- "importedPatchPresent": null,
1305
- "verificationAttempted": null
1306
- },
1307
- "gradingDurationMs": 24652.685625012964,
1308
- "totalDurationMs": 49566.27295794897
1309
- }
1310
- ]
1311
- }
1312
- }
1313
- }