@zhivex-ai/harness 1.0.0 → 1.1.0-rc.3
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.
- package/CHANGELOG.md +109 -1
- package/README.md +52 -306
- package/SECURITY.md +1 -1
- package/SUPPORT.md +7 -3
- package/benchmarks/README.md +3 -3
- package/contracts/client-protocol-v1.schema.json +341 -0
- package/contracts/public-api.json +100 -1
- package/contracts/stable-api-signatures.json +14 -14
- package/dist/approvals/approval-diff.d.ts +45 -0
- package/dist/approvals/approval-history.d.ts +38 -0
- package/dist/approvals/approval-preview.d.ts +26 -0
- package/dist/cli/arguments.d.ts +103 -0
- package/dist/cli/changes.d.ts +2 -0
- package/dist/cli/cli-credentials.d.ts +38 -0
- package/dist/cli/cli-help.d.ts +3 -0
- package/dist/{cli-options.d.ts → cli/cli-options.d.ts} +1 -1
- package/dist/{cli-profiles.d.ts → cli/cli-profiles.d.ts} +1 -1
- package/dist/cli/configured-harness.d.ts +12 -0
- package/dist/cli/console/console-commands.d.ts +6 -0
- package/dist/{console-context.d.ts → cli/console/console-context.d.ts} +7 -3
- package/dist/cli/console/console-input.d.ts +68 -0
- package/dist/cli/console/console-model-catalog.d.ts +25 -0
- package/dist/cli/console/console-navigation.d.ts +27 -0
- package/dist/cli/console/console-presentation.d.ts +18 -0
- package/dist/cli/console/console-welcome.d.ts +13 -0
- package/dist/cli/console.d.ts +2 -0
- package/dist/cli/doctor.d.ts +50 -0
- package/dist/cli/errors.d.ts +7 -0
- package/dist/cli/help-text.d.ts +2 -0
- package/dist/cli/presentation.d.ts +23 -0
- package/dist/cli/resume-metadata.d.ts +12 -0
- package/dist/cli/routing.d.ts +6 -0
- package/dist/cli/run-commands.d.ts +4 -0
- package/dist/cli/run-document.d.ts +91 -0
- package/dist/cli/run-management.d.ts +2 -0
- package/dist/cli/service-client-cli.d.ts +2 -0
- package/dist/cli/session-management.d.ts +8 -0
- package/dist/cli/setup.d.ts +16 -0
- package/dist/cli/state.d.ts +2 -0
- package/dist/cli/terminal/terminal-continuation.d.ts +3 -0
- package/dist/{terminal-markdown.d.ts → cli/terminal/terminal-markdown.d.ts} +1 -0
- package/dist/{terminal-ui.d.ts → cli/terminal/terminal-ui.d.ts} +7 -0
- package/dist/cli/terminal/tool-activity.d.ts +19 -0
- package/dist/cli.d.ts +33 -238
- package/dist/cli.js +16049 -12066
- package/dist/client/adapter.d.ts +4 -0
- package/dist/client/index.d.ts +15 -0
- package/dist/client/local-service.d.ts +34 -0
- package/dist/client/protocol.d.ts +205 -0
- package/dist/client/service-events.d.ts +50 -0
- package/dist/{context-engineering.d.ts → context/context-engineering.d.ts} +2 -2
- package/dist/{execution-environment.d.ts → execution/execution-environment.d.ts} +8 -4
- package/dist/hostile-repository-demo.js +2025 -897
- package/dist/index.d.ts +48 -41
- package/dist/index.js +7541 -5906
- package/dist/internal/desktop/persistence.d.ts +14 -0
- package/dist/internal/desktop/protocol.d.ts +8 -0
- package/dist/internal/desktop/providers.d.ts +6 -0
- package/dist/internal/desktop/runtime.d.ts +7 -0
- package/dist/{operations.d.ts → persistence/operations.d.ts} +24 -2
- package/dist/persistence/persistence-secrets.d.ts +2 -0
- package/dist/persistence/recorded-workspace.d.ts +4 -0
- package/dist/{sessions.d.ts → persistence/sessions.d.ts} +3 -0
- package/dist/persistence/sqlite-access.d.ts +26 -0
- package/dist/{sqlite-database.d.ts → persistence/sqlite-database.d.ts} +2 -0
- package/dist/{state-backup.d.ts → persistence/state-backup.d.ts} +9 -1
- package/dist/{capabilities.d.ts → providers/capabilities.d.ts} +1 -1
- package/dist/{routing.d.ts → providers/routing.d.ts} +1 -1
- package/dist/{config.d.ts → runtime/config.d.ts} +1 -1
- package/dist/{harness.d.ts → runtime/harness.d.ts} +12 -135
- package/dist/{orchestration.d.ts → runtime/orchestration.d.ts} +3 -0
- package/dist/{runtime-policy.d.ts → runtime/runtime-policy.d.ts} +1 -1
- package/dist/runtime/usage-ledger.d.ts +84 -0
- package/dist/service-cli.d.ts +2 -0
- package/dist/service-cli.js +10899 -0
- package/dist/tools/execution.d.ts +47 -0
- package/dist/tools/shared.d.ts +67 -0
- package/dist/tools/workspace.d.ts +85 -0
- package/dist/verify-historical-migrations.js +5786 -5265
- package/dist/{workspace.d.ts → workspace/workspace.d.ts} +22 -0
- package/dist/zhx.js +16049 -12066
- package/docs/CLI.md +151 -30
- package/docs/CLIENT_PROTOCOL.md +212 -0
- package/docs/CREDENTIALS.md +59 -0
- package/docs/DURABLE_OPERATIONS.md +2 -2
- package/docs/FIRST_USE.md +82 -0
- package/docs/FIRST_USE_EXAMPLE.md +40 -0
- package/docs/HOSTILE_REPOSITORY_DEMO.md +1 -1
- package/docs/LOCAL_SERVICE.md +142 -0
- package/docs/README.md +24 -23
- package/docs/THREAT_MODEL.md +1 -1
- package/docs/TIME_TO_SAFE_FIX.md +2 -2
- package/docs/USAGE.md +181 -0
- package/evaluations/daily-cli-cohort.json +33 -0
- package/evaluations/representative-assembly-matrix.json +28 -1
- package/examples/README.md +26 -0
- package/examples/first-use/greeting.test.ts +3 -0
- package/examples/first-use/greeting.ts +1 -0
- package/examples/first-use/package.json +7 -0
- package/package.json +39 -11
- package/ROADMAP.md +0 -414
- package/benchmarks/baselines/context-navigation-2026-09-09.json +0 -94
- package/benchmarks/baselines/efficiency-2026-09-08.json +0 -233
- package/benchmarks/baselines/external-final-2026-09-08.json +0 -588
- package/benchmarks/baselines/external-intermediate-2026-09-08.json +0 -116
- package/benchmarks/baselines/external-matched-openai-2026-09-08.json +0 -1922
- package/benchmarks/baselines/external-matched-qwen-2026-09-08.json +0 -1706
- package/benchmarks/baselines/external-pilot-2026-09-08.json +0 -439
- package/benchmarks/baselines/external-qwen-flash-2026-09-08.json +0 -1570
- package/benchmarks/baselines/external-qwen-remediation-2026-09-09.json +0 -1509
- package/benchmarks/baselines/external-qwen-sdk-114-progress-2026-09-09.json +0 -2509
- package/benchmarks/baselines/external-sdk-upgrade-2026-09-08.json +0 -564
- package/benchmarks/baselines/external-verifier-recovery-2026-09-09.json +0 -794
- package/benchmarks/baselines/full-harness-review-2026-09-10.json +0 -134
- package/benchmarks/baselines/harness-audit-2026-09-09.json +0 -136
- package/benchmarks/baselines/harness-remediation-2026-09-09.json +0 -160
- package/benchmarks/baselines/oci-remediation-2026-09-10.json +0 -535
- package/benchmarks/baselines/rc14-holdout-2026-09-10.json +0 -854
- package/benchmarks/baselines/repair-architecture-audit-2026-09-09.json +0 -75
- package/benchmarks/baselines/time-to-safe-fix-live-expanded-2026-08-21.json +0 -186
- package/benchmarks/baselines/time-to-safe-fix-live-smoke-2026-08-21.json +0 -186
- package/dist/console-input.d.ts +0 -28
- package/docs/GA_READINESS.md +0 -29
- package/docs/GEMINI_1_0_DECISION.md +0 -9
- package/docs/LIVE_CERTIFICATION.md +0 -107
- package/docs/RC13_SECURITY_REVIEW.md +0 -28
- package/docs/RC14_DELEGATED_SECURITY_DECISION.md +0 -80
- package/docs/RC14_SECURITY_REVIEW.md +0 -40
- package/docs/RELEASE.md +0 -96
- package/docs/ROLLBACK.md +0 -9
- package/docs/SECURITY_REVIEW_EVIDENCE.md +0 -17
- package/docs/ga-readiness.json +0 -366
- package/docs/reports/EFFICIENCY_REVIEW.md +0 -156
- package/docs/reports/EXTERNAL_COMPARISON.md +0 -434
- package/docs/reports/FULL_HARNESS_REVIEW_2026-09-10.md +0 -200
- package/docs/reports/GA_BUDGET_REMEDIATION_2026-09-19.md +0 -38
- package/docs/reports/GA_REPAIR_RECOVERY_2026-09-19.md +0 -700
- package/docs/reports/GA_VALIDATION_2026-09-19.md +0 -111
- package/docs/reports/HARNESS_ARCHITECTURE_REVIEW_2026-09-09.md +0 -202
- package/docs/reports/HARNESS_AUDIT_2026-09-09.md +0 -306
- package/docs/reports/HARNESS_REMEDIATION_2026-09-09.md +0 -74
- package/docs/reports/HARNESS_REMEDIATION_2026-09-10.md +0 -90
- package/docs/reports/MODEL_COMPARISON_2026-09-09.md +0 -184
- package/docs/reports/QWEN_REMEDIATION_COMPARISON_2026-09-09.md +0 -142
- package/docs/reports/RC14_VALIDATION_2026-09-10.md +0 -45
- package/docs/reports/README.md +0 -26
- package/docs/reports/evidence/ga-budget-remediation-2026-09-19.json +0 -133
- package/docs/reports/evidence/ga-catalog-ablation-2026-09-19.json +0 -699
- package/docs/reports/evidence/ga-catalog-ablation-plan-2026-09-19.json +0 -40
- package/docs/reports/evidence/ga-catalog-observed-2026-09-19.json +0 -985
- package/docs/reports/evidence/ga-catalog-observed-plan-2026-09-19.json +0 -28
- package/docs/reports/evidence/ga-catalog-observed-status-2026-09-19.json +0 -15
- package/docs/reports/evidence/ga-complete-local-check-2026-09-19.json +0 -47
- package/docs/reports/evidence/ga-completion-reminder-2026-09-19.json +0 -22
- package/docs/reports/evidence/ga-fresh-cohort-plan-2026-09-19.json +0 -54
- package/docs/reports/evidence/ga-fresh-cohort-preparation-2026-09-19.json +0 -242
- package/docs/reports/evidence/ga-fresh-cohort-result-2026-09-19.json +0 -4318
- package/docs/reports/evidence/ga-local-validation-followup-2026-09-19.json +0 -66
- package/docs/reports/evidence/ga-node-consumers-2026-09-19.json +0 -31
- package/docs/reports/evidence/ga-package-content-validation-2026-09-19.json +0 -22
- package/docs/reports/evidence/ga-plan-completion-2026-09-19.json +0 -39
- package/docs/reports/evidence/ga-plan-reserve-2026-09-19.json +0 -573
- package/docs/reports/evidence/ga-plan-reserve-plan-2026-09-19.json +0 -24
- package/docs/reports/evidence/ga-plan-transition-2026-09-19.json +0 -19
- package/docs/reports/evidence/ga-private-verifier-plan-2026-09-19.json +0 -24
- package/docs/reports/evidence/ga-private-verifier-root-cause-2026-09-19.json +0 -41
- package/docs/reports/evidence/ga-published-sdk-2026-09-19.json +0 -35
- package/docs/reports/evidence/ga-python-checkout-imports-2026-09-19.json +0 -95
- package/docs/reports/evidence/ga-python-origin-delivery-2026-09-19.json +0 -211
- package/docs/reports/evidence/ga-python-origin-delivery-plan-2026-09-19.json +0 -27
- package/docs/reports/evidence/ga-qwen-patched-package-2026-09-19.json +0 -1092
- package/docs/reports/evidence/ga-qwen-patched-package-plan-2026-09-19.json +0 -18
- package/docs/reports/evidence/ga-qwen-published-2026-09-19.json +0 -571
- package/docs/reports/evidence/ga-qwen-published-plan-2026-09-19.json +0 -18
- package/docs/reports/evidence/ga-qwen-tool-choice-2026-09-19.json +0 -85
- package/docs/reports/evidence/ga-qwen-tool-choice-patched-2026-09-19.json +0 -85
- package/docs/reports/evidence/ga-recovery-development-2026-09-19.json +0 -1066
- package/docs/reports/evidence/ga-recovery-graded-development-2026-09-19.json +0 -1313
- package/docs/reports/evidence/ga-recovery-graded-v2-2026-09-19.json +0 -1151
- package/docs/reports/evidence/ga-recovery-v3-2026-09-19.json +0 -3943
- package/docs/reports/evidence/ga-recovery-v4-2026-09-19.json +0 -1055
- package/docs/reports/evidence/ga-recovery-v5-2026-09-19.json +0 -1245
- package/docs/reports/evidence/ga-recovery-v6-2026-09-19.json +0 -699
- package/docs/reports/evidence/ga-recovery-v7-2026-09-19.json +0 -854
- package/docs/reports/evidence/ga-recovery-v8-2026-09-19.json +0 -596
- package/docs/reports/evidence/ga-required-delivery-2026-09-19.json +0 -25
- package/docs/reports/evidence/ga-required-delivery-live-2026-09-19.json +0 -1337
- package/docs/reports/evidence/ga-required-delivery-live-plan-2026-09-19.json +0 -25
- package/docs/reports/evidence/ga-retained-candidate-grading-2026-09-19.json +0 -30
- package/docs/reports/evidence/ga-tool-selection-live-2026-09-19.json +0 -559
- package/docs/reports/evidence/ga-tool-selection-live-plan-2026-09-19.json +0 -28
- package/docs/reports/evidence/ga-tool-selection-recovery-2026-09-19.json +0 -27
- package/docs/reports/evidence/ga-validation-2026-09-19.json +0 -236
- package/docs/reports/evidence/ga-verifier-discovery-live-2026-09-19.json +0 -1369
- package/docs/reports/evidence/ga-verifier-discovery-plan-2026-09-19.json +0 -29
- package/docs/reports/evidence/ga-verifier-failure-identification-2026-09-19.json +0 -30
- package/docs/reports/evidence/ga-verifier-feedback-2026-09-19.json +0 -65
- package/docs/reports/evidence/ga-verifier-hints-live-2026-09-19.json +0 -1297
- package/docs/reports/evidence/ga-verifier-hints-plan-2026-09-19.json +0 -27
- package/docs/reports/evidence/ga-work-boundary-context-analysis-2026-09-19.json +0 -113
- package/docs/reports/evidence/ga-work-boundary-live-2026-09-19.json +0 -301
- package/docs/reports/evidence/ga-work-boundary-live-plan-2026-09-19.json +0 -84
- package/docs/reports/evidence/ga-work-budget-planning-2026-09-19.json +0 -32
- package/docs/reports/evidence/ga-working-context-2026-09-19.json +0 -22
- package/docs/reports/evidence/rc14-delegated-security-review-2026-09-20.json +0 -81
- package/docs/reports/evidence/rc14-ga-gate-2026-09-20.json +0 -18
- package/docs/reports/evidence/rc14-representative-2026-09-20.json +0 -509
- package/docs/reports/evidence/rc14-security-regressions-2026-09-20.json +0 -64
- package/docs/reports/evidence/security-current-2026-09-19.json +0 -91
- package/docs/reports/evidence/security-remediation-2026-09-19.json +0 -97
- /package/dist/{cli-stream.d.ts → cli/cli-stream.d.ts} +0 -0
- /package/dist/{json-contracts.d.ts → client/json-contracts.d.ts} +0 -0
- /package/dist/{compaction.d.ts → context/compaction.d.ts} +0 -0
- /package/dist/{context-metrics.d.ts → context/context-metrics.d.ts} +0 -0
- /package/dist/{task-memory.d.ts → context/task-memory.d.ts} +0 -0
- /package/dist/{package-manager.d.ts → execution/package-manager.d.ts} +0 -0
- /package/dist/{process-runtime.d.ts → execution/process-runtime.d.ts} +0 -0
- /package/dist/{mcp.d.ts → integrations/mcp.d.ts} +0 -0
- /package/dist/{state-directory.d.ts → persistence/state-directory.d.ts} +0 -0
- /package/dist/{providers.d.ts → providers/providers.d.ts} +0 -0
- /package/dist/{errors.d.ts → runtime/errors.d.ts} +0 -0
- /package/dist/{model-budget.d.ts → runtime/model-budget.d.ts} +0 -0
- /package/dist/{repair-controller.d.ts → runtime/repair-controller.d.ts} +0 -0
- /package/dist/{repair-progress.d.ts → runtime/repair-progress.d.ts} +0 -0
- /package/dist/{repair-verifier.d.ts → runtime/repair-verifier.d.ts} +0 -0
- /package/dist/{run-interruption.d.ts → runtime/run-interruption.d.ts} +0 -0
- /package/dist/{runtime-checkpoints.d.ts → runtime/runtime-checkpoints.d.ts} +0 -0
- /package/dist/{runtime-diagnostics.d.ts → runtime/runtime-diagnostics.d.ts} +0 -0
- /package/dist/{time-to-safe-fix.d.ts → runtime/time-to-safe-fix.d.ts} +0 -0
- /package/dist/{tool-registry.d.ts → tools/tool-registry.d.ts} +0 -0
- /package/dist/{bounded-reads.d.ts → workspace/bounded-reads.d.ts} +0 -0
- /package/dist/{change-envelope.d.ts → workspace/change-envelope.d.ts} +0 -0
- /package/dist/{edit-contracts.d.ts → workspace/edit-contracts.d.ts} +0 -0
- /package/dist/{file-security.d.ts → workspace/file-security.d.ts} +0 -0
- /package/dist/{replacement-edits.d.ts → workspace/replacement-edits.d.ts} +0 -0
- /package/dist/{workspace-mutation-lock.d.ts → workspace/workspace-mutation-lock.d.ts} +0 -0
|
@@ -1,854 +0,0 @@
|
|
|
1
|
-
{
|
|
2
|
-
"schemaVersion": 1,
|
|
3
|
-
"kind": "rc14-holdout-validation",
|
|
4
|
-
"sourceCommit": "0609f11681cf44d0fc68b7671ed4e37ceaf48309",
|
|
5
|
-
"sourceHashes": {
|
|
6
|
-
"bun.lock": "b0440362d043879ece281bcee4c7adfa775a3724d2a9477079e187babd3b2d26",
|
|
7
|
-
"evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
|
|
8
|
-
"package.json": "0ed0eea4c944350f65a7e1e7304d8c0ec7b2ae3d07945490b51c95ae49b54777",
|
|
9
|
-
"scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
|
|
10
|
-
"scripts/swebench/candidate.py": "f5e3410b91b950acfbe50fc7511891816a8fb1c223082397e2c6300cf1e866cf",
|
|
11
|
-
"scripts/swebench/candidate.ts": "bf102becfcce0c9a3f0df0a8f1ce0e5aaea7db2d252f6d9454a8370c24e53982",
|
|
12
|
-
"scripts/swebench/common.py": "12f7851e8106b22f2124751ce56e3742348ebb3c40ef0bc89ee9d2fefa0aff77",
|
|
13
|
-
"scripts/swebench/context-metrics.ts": "7b84da6789413aa634d997c5963ee6796584e06c3eafe5f68f424fb7e27815e9",
|
|
14
|
-
"scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
|
|
15
|
-
"scripts/swebench/model-budget.ts": "e46dd406ee5d58a051a15c7d7133513666292303d8c6534f861a6db0e711f47c",
|
|
16
|
-
"scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
|
|
17
|
-
"scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
|
|
18
|
-
"scripts/swebench/report.py": "d0b8d45f32c48e0ba2cb977f50838065a497a0ecaf3b4243ed94139393ce4d6b",
|
|
19
|
-
"scripts/swebench/run.py": "24ffa302e673bbf240d6396474631bb4ce80447797584c3ed27f2d1e55e452cb",
|
|
20
|
-
"scripts/swebench/telemetry.ts": "743c9baf273e9240d704d098b6bd5eea9e76ef3e49ae636aaafa9bfd7a85b45b",
|
|
21
|
-
"scripts/swebench/test_candidate.py": "e77d16379b748baa2daf184a45ea62587d938b6a7d48dcf9f6061f8af21fafd9",
|
|
22
|
-
"scripts/swebench/test_contracts.py": "a8c1348f1dff2e563c139e6d3e83eb141b4f2a71f91a58e2c5c051ed2a431024",
|
|
23
|
-
"scripts/swebench/zhivex-driver.ts": "e33ed929256133fb29be67e3feb282f0d0fd52086576a668021852c151ff35d1",
|
|
24
|
-
"scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
|
|
25
|
-
"src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
|
|
26
|
-
"src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
|
|
27
|
-
"src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
|
|
28
|
-
"src/cli-options.ts": "f48d9be132fed38f2ddbcfecbd8222c664a0df2f20edc28ce8d1b9b1a1900863",
|
|
29
|
-
"src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
|
|
30
|
-
"src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
|
|
31
|
-
"src/cli.ts": "8f44daed380575481e3aa22ec559317f58d88aca30c7fc80fe860163ef5ea336",
|
|
32
|
-
"src/compaction.ts": "9e06a0abeeef155377f40c160b21c56256bca95ec18c5ee897681a5fc73db169",
|
|
33
|
-
"src/config.ts": "a31e9f91c0eb59b4d77496431639aa4b47b149e1968c6e0a002a508de4919c35",
|
|
34
|
-
"src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
|
|
35
|
-
"src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
|
|
36
|
-
"src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
|
|
37
|
-
"src/context-metrics.ts": "3eb30f9db808f918ed8d73d4f8ac19a13acb09ed5c1ee70fdae05201714b9198",
|
|
38
|
-
"src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
|
|
39
|
-
"src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
|
|
40
|
-
"src/execution-environment.ts": "980ec7cb0aa74f5dbfead7dedc4ca221a27b997de2bc2d463d7073813fe83bbc",
|
|
41
|
-
"src/file-security.ts": "b8c9f4988c4339c73a60ab4fa8d394b3eecc19117a0c61a8d19ab6483ca253df",
|
|
42
|
-
"src/harness.ts": "62ab0f6f2a428ce23f3f3f8847402c75ffe7180c939a0236af87612c04fae1ba",
|
|
43
|
-
"src/index.ts": "4bbfcd11376d48c4f1c1ec17d6b5c9f51baa9331fa999432be178bae94530a5e",
|
|
44
|
-
"src/json-contracts.ts": "0780145636b47829a1aa6087c574d5654eb31b7534ea7d9c38c188d648c90272",
|
|
45
|
-
"src/mcp.ts": "7722de031921c620ec03b9e42a6596ef8415aa8e4bb5f93ea84d2427c466e554",
|
|
46
|
-
"src/model-budget.ts": "e5a5d3dd8c97df7f18f36ce0a82ed250e4a0c7d49b12580a7393db6ff2183a55",
|
|
47
|
-
"src/operations.ts": "bcccaf4d1d0ec117c2ed35df92185564d33399e5fdc5c12225ac61f14fd0b4b1",
|
|
48
|
-
"src/orchestration.ts": "332264f91fc9630e8d5d32e4ca2aebdf063fdb87acc457c4854c711c725d6e26",
|
|
49
|
-
"src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
|
|
50
|
-
"src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
|
|
51
|
-
"src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
|
|
52
|
-
"src/repair-controller.ts": "ce12b1c9fb9a5fa5d370b2eda5c1233cc8c33706477ef66302b52bf7e8c40e09",
|
|
53
|
-
"src/repair-progress.ts": "38270f07c4e1d81d745aa3a01837cea4420b7caf310df59c7447e56b1a3e0440",
|
|
54
|
-
"src/repair-verifier.ts": "b70fb11feee8cdc7933947ccac0b78f8bb5c0eaccc3b29ad527197d77f40ffc9",
|
|
55
|
-
"src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
|
|
56
|
-
"src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
|
|
57
|
-
"src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
|
|
58
|
-
"src/runtime-checkpoints.ts": "a9b555d4c6b5f369f86a3d79ec1a3aab45886282cc7d0ef211aa084ee17e3897",
|
|
59
|
-
"src/runtime-diagnostics.ts": "e1e5a3a558fa2aaeb477335dc336551f022c1234fe5e742eefc6599c8803e364",
|
|
60
|
-
"src/runtime-policy.ts": "bed4efad0010aa63d7e083b0c8014cb36a915626ede00b8a40e1664310972f3c",
|
|
61
|
-
"src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
|
|
62
|
-
"src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
|
|
63
|
-
"src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
|
|
64
|
-
"src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
|
|
65
|
-
"src/task-memory.ts": "5b62c863be5a23c33608633cd7aac0464f5536d01cba930aca58b11870170731",
|
|
66
|
-
"src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
|
|
67
|
-
"src/terminal-ui.ts": "e50025a8649d1f60efdf3af895b8639730ec2e86553fb80a61a776b11076c705",
|
|
68
|
-
"src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
|
|
69
|
-
"src/tool-registry.ts": "7e90fff11940e525883c9ec0d2adbea2e6616f8bd9bc4ec77e1b7277e4217f2c",
|
|
70
|
-
"src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
|
|
71
|
-
"src/workspace.ts": "0f472998a55672a326bfb32404a10073a90c9a34c24758471fede0345f582961"
|
|
72
|
-
},
|
|
73
|
-
"sourceAndSnapshotsVerified": true,
|
|
74
|
-
"datasetRevision": "c104f840cc67f8b6eec6f759ebc8b2693d585d4a",
|
|
75
|
-
"taskIds": [
|
|
76
|
-
"django__django-15128",
|
|
77
|
-
"django__django-15851"
|
|
78
|
-
],
|
|
79
|
-
"excludedPriorTaskIds": [
|
|
80
|
-
"astropy__astropy-14995",
|
|
81
|
-
"django__django-11265",
|
|
82
|
-
"django__django-12262",
|
|
83
|
-
"django__django-13410",
|
|
84
|
-
"django__django-14915",
|
|
85
|
-
"django__django-15554",
|
|
86
|
-
"django__django-15569",
|
|
87
|
-
"pydata__xarray-4695",
|
|
88
|
-
"pydata__xarray-6721",
|
|
89
|
-
"pytest-dev__pytest-8399",
|
|
90
|
-
"scikit-learn__scikit-learn-10908",
|
|
91
|
-
"scikit-learn__scikit-learn-14141",
|
|
92
|
-
"sympy__sympy-15809",
|
|
93
|
-
"sympy__sympy-20438",
|
|
94
|
-
"sympy__sympy-21930"
|
|
95
|
-
],
|
|
96
|
-
"seed": 20260910,
|
|
97
|
-
"repetitions": 3,
|
|
98
|
-
"localArtifactSha512": "sha512-BIBg1hcrQLAvJJboiFxWv6lzWb3tamr5qO83fWEa9MyrhTkvTJ9FYy62iL+o2N+5Jx5Mj4GOICrX1NpGd827Wg==",
|
|
99
|
-
"artifactBoundary": "Local tarball inspected and installed separately. Holdout executes the hash-frozen source runtime, not the protected release artifact. No publication certification claimed.",
|
|
100
|
-
"runs": {
|
|
101
|
-
"openai": {
|
|
102
|
-
"model": "gpt-5.6-luna",
|
|
103
|
-
"reasoning": "low",
|
|
104
|
-
"transport": "openai-responses",
|
|
105
|
-
"limits": {
|
|
106
|
-
"steps": 24,
|
|
107
|
-
"outputPerTurn": 2048,
|
|
108
|
-
"inputTokens": 100000,
|
|
109
|
-
"outputTokens": 16000,
|
|
110
|
-
"timeoutSeconds": 300,
|
|
111
|
-
"memoryMb": 2048,
|
|
112
|
-
"cpus": 2
|
|
113
|
-
},
|
|
114
|
-
"report": {
|
|
115
|
-
"schemaVersion": 1,
|
|
116
|
-
"kind": "swebench-comparison",
|
|
117
|
-
"manifestSha256": "53d564ab2d4e608b9740c2292c9056f6d19ef789c9de7c1508c838b38ced8519",
|
|
118
|
-
"complete": true,
|
|
119
|
-
"candidates": {
|
|
120
|
-
"zhivex": {
|
|
121
|
-
"planned": 6,
|
|
122
|
-
"recorded": 6,
|
|
123
|
-
"resolved": 1,
|
|
124
|
-
"candidateCaptured": 6,
|
|
125
|
-
"candidateResolved": 3,
|
|
126
|
-
"candidateGraded": 6,
|
|
127
|
-
"resolutionRate": 0.16666666666666666,
|
|
128
|
-
"missing": 0,
|
|
129
|
-
"gradingFailures": 0,
|
|
130
|
-
"usageComplete": true,
|
|
131
|
-
"totalCostUsd": null,
|
|
132
|
-
"costPerResolvedUsd": null,
|
|
133
|
-
"resolvedP50Ms": 80594.56762496848
|
|
134
|
-
},
|
|
135
|
-
"mini-swe-agent": {
|
|
136
|
-
"planned": 6,
|
|
137
|
-
"recorded": 6,
|
|
138
|
-
"resolved": 4,
|
|
139
|
-
"candidateCaptured": 6,
|
|
140
|
-
"candidateResolved": 4,
|
|
141
|
-
"candidateGraded": 6,
|
|
142
|
-
"resolutionRate": 0.6666666666666666,
|
|
143
|
-
"missing": 0,
|
|
144
|
-
"gradingFailures": 0,
|
|
145
|
-
"usageComplete": true,
|
|
146
|
-
"totalCostUsd": null,
|
|
147
|
-
"costPerResolvedUsd": null,
|
|
148
|
-
"resolvedP50Ms": 50588.63154100254
|
|
149
|
-
}
|
|
150
|
-
},
|
|
151
|
-
"pairedTaskResolutionDifference": -0.5,
|
|
152
|
-
"pairedTaskBootstrap95": [
|
|
153
|
-
-0.6666666666666667,
|
|
154
|
-
-0.3333333333333333
|
|
155
|
-
],
|
|
156
|
-
"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."
|
|
157
|
-
},
|
|
158
|
-
"samples": [
|
|
159
|
-
{
|
|
160
|
-
"candidate": "zhivex",
|
|
161
|
-
"instanceId": "django__django-15128",
|
|
162
|
-
"repetition": 0,
|
|
163
|
-
"status": "failed",
|
|
164
|
-
"resolved": false,
|
|
165
|
-
"candidateResolved": false,
|
|
166
|
-
"candidateCaptured": true,
|
|
167
|
-
"gradingStatus": "completed",
|
|
168
|
-
"candidateGradingStatus": "completed",
|
|
169
|
-
"inputTokens": 57917,
|
|
170
|
-
"outputTokens": 931,
|
|
171
|
-
"cachedInputTokens": 19808,
|
|
172
|
-
"usageComplete": true,
|
|
173
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
174
|
-
"repairStages": {
|
|
175
|
-
"candidatePresent": false,
|
|
176
|
-
"importedPatchPresent": false,
|
|
177
|
-
"verificationAttempted": false
|
|
178
|
-
},
|
|
179
|
-
"repairProgress": {
|
|
180
|
-
"repeatedResults": 1,
|
|
181
|
-
"suppressedResults": 0,
|
|
182
|
-
"blockedBroadCalls": 0,
|
|
183
|
-
"enteredClosure": false,
|
|
184
|
-
"phase": "explore",
|
|
185
|
-
"blockedPhaseCalls": 0,
|
|
186
|
-
"planRecorded": false
|
|
187
|
-
}
|
|
188
|
-
},
|
|
189
|
-
{
|
|
190
|
-
"candidate": "mini-swe-agent",
|
|
191
|
-
"instanceId": "django__django-15128",
|
|
192
|
-
"repetition": 0,
|
|
193
|
-
"status": "Submitted",
|
|
194
|
-
"resolved": true,
|
|
195
|
-
"candidateResolved": true,
|
|
196
|
-
"candidateCaptured": true,
|
|
197
|
-
"gradingStatus": "completed",
|
|
198
|
-
"candidateGradingStatus": "completed",
|
|
199
|
-
"inputTokens": 76213,
|
|
200
|
-
"outputTokens": 1852,
|
|
201
|
-
"cachedInputTokens": 63396,
|
|
202
|
-
"usageComplete": true,
|
|
203
|
-
"failure": null,
|
|
204
|
-
"repairStages": {
|
|
205
|
-
"candidatePresent": true,
|
|
206
|
-
"importedPatchPresent": null,
|
|
207
|
-
"verificationAttempted": null
|
|
208
|
-
},
|
|
209
|
-
"repairProgress": null
|
|
210
|
-
},
|
|
211
|
-
{
|
|
212
|
-
"candidate": "mini-swe-agent",
|
|
213
|
-
"instanceId": "django__django-15128",
|
|
214
|
-
"repetition": 1,
|
|
215
|
-
"status": "Submitted",
|
|
216
|
-
"resolved": false,
|
|
217
|
-
"candidateResolved": false,
|
|
218
|
-
"candidateCaptured": true,
|
|
219
|
-
"gradingStatus": "completed",
|
|
220
|
-
"candidateGradingStatus": "completed",
|
|
221
|
-
"inputTokens": 84642,
|
|
222
|
-
"outputTokens": 2508,
|
|
223
|
-
"cachedInputTokens": 72996,
|
|
224
|
-
"usageComplete": true,
|
|
225
|
-
"failure": null,
|
|
226
|
-
"repairStages": {
|
|
227
|
-
"candidatePresent": true,
|
|
228
|
-
"importedPatchPresent": null,
|
|
229
|
-
"verificationAttempted": null
|
|
230
|
-
},
|
|
231
|
-
"repairProgress": null
|
|
232
|
-
},
|
|
233
|
-
{
|
|
234
|
-
"candidate": "zhivex",
|
|
235
|
-
"instanceId": "django__django-15128",
|
|
236
|
-
"repetition": 1,
|
|
237
|
-
"status": "failed",
|
|
238
|
-
"resolved": false,
|
|
239
|
-
"candidateResolved": false,
|
|
240
|
-
"candidateCaptured": true,
|
|
241
|
-
"gradingStatus": "completed",
|
|
242
|
-
"candidateGradingStatus": "completed",
|
|
243
|
-
"inputTokens": 60677,
|
|
244
|
-
"outputTokens": 776,
|
|
245
|
-
"cachedInputTokens": 23581,
|
|
246
|
-
"usageComplete": true,
|
|
247
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
248
|
-
"repairStages": {
|
|
249
|
-
"candidatePresent": false,
|
|
250
|
-
"importedPatchPresent": false,
|
|
251
|
-
"verificationAttempted": false
|
|
252
|
-
},
|
|
253
|
-
"repairProgress": {
|
|
254
|
-
"repeatedResults": 0,
|
|
255
|
-
"suppressedResults": 0,
|
|
256
|
-
"blockedBroadCalls": 0,
|
|
257
|
-
"enteredClosure": false,
|
|
258
|
-
"phase": "explore",
|
|
259
|
-
"blockedPhaseCalls": 0,
|
|
260
|
-
"planRecorded": false
|
|
261
|
-
}
|
|
262
|
-
},
|
|
263
|
-
{
|
|
264
|
-
"candidate": "zhivex",
|
|
265
|
-
"instanceId": "django__django-15128",
|
|
266
|
-
"repetition": 2,
|
|
267
|
-
"status": "failed",
|
|
268
|
-
"resolved": false,
|
|
269
|
-
"candidateResolved": false,
|
|
270
|
-
"candidateCaptured": true,
|
|
271
|
-
"gradingStatus": "completed",
|
|
272
|
-
"candidateGradingStatus": "completed",
|
|
273
|
-
"inputTokens": 65281,
|
|
274
|
-
"outputTokens": 766,
|
|
275
|
-
"cachedInputTokens": 28295,
|
|
276
|
-
"usageComplete": true,
|
|
277
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
278
|
-
"repairStages": {
|
|
279
|
-
"candidatePresent": false,
|
|
280
|
-
"importedPatchPresent": false,
|
|
281
|
-
"verificationAttempted": false
|
|
282
|
-
},
|
|
283
|
-
"repairProgress": {
|
|
284
|
-
"repeatedResults": 0,
|
|
285
|
-
"suppressedResults": 0,
|
|
286
|
-
"blockedBroadCalls": 0,
|
|
287
|
-
"enteredClosure": false,
|
|
288
|
-
"phase": "explore",
|
|
289
|
-
"blockedPhaseCalls": 0,
|
|
290
|
-
"planRecorded": false
|
|
291
|
-
}
|
|
292
|
-
},
|
|
293
|
-
{
|
|
294
|
-
"candidate": "mini-swe-agent",
|
|
295
|
-
"instanceId": "django__django-15128",
|
|
296
|
-
"repetition": 2,
|
|
297
|
-
"status": "Submitted",
|
|
298
|
-
"resolved": false,
|
|
299
|
-
"candidateResolved": false,
|
|
300
|
-
"candidateCaptured": true,
|
|
301
|
-
"gradingStatus": "completed",
|
|
302
|
-
"candidateGradingStatus": "completed",
|
|
303
|
-
"inputTokens": 79373,
|
|
304
|
-
"outputTokens": 1971,
|
|
305
|
-
"cachedInputTokens": 68544,
|
|
306
|
-
"usageComplete": true,
|
|
307
|
-
"failure": null,
|
|
308
|
-
"repairStages": {
|
|
309
|
-
"candidatePresent": true,
|
|
310
|
-
"importedPatchPresent": null,
|
|
311
|
-
"verificationAttempted": null
|
|
312
|
-
},
|
|
313
|
-
"repairProgress": null
|
|
314
|
-
},
|
|
315
|
-
{
|
|
316
|
-
"candidate": "mini-swe-agent",
|
|
317
|
-
"instanceId": "django__django-15851",
|
|
318
|
-
"repetition": 0,
|
|
319
|
-
"status": "Submitted",
|
|
320
|
-
"resolved": true,
|
|
321
|
-
"candidateResolved": true,
|
|
322
|
-
"candidateCaptured": true,
|
|
323
|
-
"gradingStatus": "completed",
|
|
324
|
-
"candidateGradingStatus": "completed",
|
|
325
|
-
"inputTokens": 26367,
|
|
326
|
-
"outputTokens": 1195,
|
|
327
|
-
"cachedInputTokens": 20848,
|
|
328
|
-
"usageComplete": true,
|
|
329
|
-
"failure": null,
|
|
330
|
-
"repairStages": {
|
|
331
|
-
"candidatePresent": true,
|
|
332
|
-
"importedPatchPresent": null,
|
|
333
|
-
"verificationAttempted": null
|
|
334
|
-
},
|
|
335
|
-
"repairProgress": null
|
|
336
|
-
},
|
|
337
|
-
{
|
|
338
|
-
"candidate": "zhivex",
|
|
339
|
-
"instanceId": "django__django-15851",
|
|
340
|
-
"repetition": 0,
|
|
341
|
-
"status": "failed",
|
|
342
|
-
"resolved": false,
|
|
343
|
-
"candidateResolved": true,
|
|
344
|
-
"candidateCaptured": true,
|
|
345
|
-
"gradingStatus": "completed",
|
|
346
|
-
"candidateGradingStatus": "completed",
|
|
347
|
-
"inputTokens": 72876,
|
|
348
|
-
"outputTokens": 1512,
|
|
349
|
-
"cachedInputTokens": 27236,
|
|
350
|
-
"usageComplete": true,
|
|
351
|
-
"failure": "EXECUTION_FAILED",
|
|
352
|
-
"repairStages": {
|
|
353
|
-
"candidatePresent": true,
|
|
354
|
-
"importedPatchPresent": false,
|
|
355
|
-
"verificationAttempted": true
|
|
356
|
-
},
|
|
357
|
-
"repairProgress": {
|
|
358
|
-
"repeatedResults": 0,
|
|
359
|
-
"suppressedResults": 0,
|
|
360
|
-
"blockedBroadCalls": 0,
|
|
361
|
-
"enteredClosure": true,
|
|
362
|
-
"phase": "verify",
|
|
363
|
-
"blockedPhaseCalls": 0,
|
|
364
|
-
"planRecorded": true
|
|
365
|
-
}
|
|
366
|
-
},
|
|
367
|
-
{
|
|
368
|
-
"candidate": "zhivex",
|
|
369
|
-
"instanceId": "django__django-15851",
|
|
370
|
-
"repetition": 1,
|
|
371
|
-
"status": "completed",
|
|
372
|
-
"resolved": true,
|
|
373
|
-
"candidateResolved": true,
|
|
374
|
-
"candidateCaptured": true,
|
|
375
|
-
"gradingStatus": "completed",
|
|
376
|
-
"candidateGradingStatus": "completed",
|
|
377
|
-
"inputTokens": 42840,
|
|
378
|
-
"outputTokens": 897,
|
|
379
|
-
"cachedInputTokens": 17868,
|
|
380
|
-
"usageComplete": true,
|
|
381
|
-
"failure": null,
|
|
382
|
-
"repairStages": {
|
|
383
|
-
"candidatePresent": true,
|
|
384
|
-
"importedPatchPresent": true,
|
|
385
|
-
"verificationAttempted": true
|
|
386
|
-
},
|
|
387
|
-
"repairProgress": {
|
|
388
|
-
"repeatedResults": 0,
|
|
389
|
-
"suppressedResults": 0,
|
|
390
|
-
"blockedBroadCalls": 0,
|
|
391
|
-
"enteredClosure": false,
|
|
392
|
-
"phase": "verify",
|
|
393
|
-
"blockedPhaseCalls": 0,
|
|
394
|
-
"planRecorded": true
|
|
395
|
-
}
|
|
396
|
-
},
|
|
397
|
-
{
|
|
398
|
-
"candidate": "mini-swe-agent",
|
|
399
|
-
"instanceId": "django__django-15851",
|
|
400
|
-
"repetition": 1,
|
|
401
|
-
"status": "Submitted",
|
|
402
|
-
"resolved": true,
|
|
403
|
-
"candidateResolved": true,
|
|
404
|
-
"candidateCaptured": true,
|
|
405
|
-
"gradingStatus": "completed",
|
|
406
|
-
"candidateGradingStatus": "completed",
|
|
407
|
-
"inputTokens": 14726,
|
|
408
|
-
"outputTokens": 989,
|
|
409
|
-
"cachedInputTokens": 12580,
|
|
410
|
-
"usageComplete": true,
|
|
411
|
-
"failure": null,
|
|
412
|
-
"repairStages": {
|
|
413
|
-
"candidatePresent": true,
|
|
414
|
-
"importedPatchPresent": null,
|
|
415
|
-
"verificationAttempted": null
|
|
416
|
-
},
|
|
417
|
-
"repairProgress": null
|
|
418
|
-
},
|
|
419
|
-
{
|
|
420
|
-
"candidate": "mini-swe-agent",
|
|
421
|
-
"instanceId": "django__django-15851",
|
|
422
|
-
"repetition": 2,
|
|
423
|
-
"status": "Submitted",
|
|
424
|
-
"resolved": true,
|
|
425
|
-
"candidateResolved": true,
|
|
426
|
-
"candidateCaptured": true,
|
|
427
|
-
"gradingStatus": "completed",
|
|
428
|
-
"candidateGradingStatus": "completed",
|
|
429
|
-
"inputTokens": 25208,
|
|
430
|
-
"outputTokens": 1233,
|
|
431
|
-
"cachedInputTokens": 21065,
|
|
432
|
-
"usageComplete": true,
|
|
433
|
-
"failure": null,
|
|
434
|
-
"repairStages": {
|
|
435
|
-
"candidatePresent": true,
|
|
436
|
-
"importedPatchPresent": null,
|
|
437
|
-
"verificationAttempted": null
|
|
438
|
-
},
|
|
439
|
-
"repairProgress": null
|
|
440
|
-
},
|
|
441
|
-
{
|
|
442
|
-
"candidate": "zhivex",
|
|
443
|
-
"instanceId": "django__django-15851",
|
|
444
|
-
"repetition": 2,
|
|
445
|
-
"status": "failed",
|
|
446
|
-
"resolved": false,
|
|
447
|
-
"candidateResolved": true,
|
|
448
|
-
"candidateCaptured": true,
|
|
449
|
-
"gradingStatus": "completed",
|
|
450
|
-
"candidateGradingStatus": "completed",
|
|
451
|
-
"inputTokens": 73076,
|
|
452
|
-
"outputTokens": 1801,
|
|
453
|
-
"cachedInputTokens": 28815,
|
|
454
|
-
"usageComplete": true,
|
|
455
|
-
"failure": "EXECUTION_FAILED",
|
|
456
|
-
"repairStages": {
|
|
457
|
-
"candidatePresent": true,
|
|
458
|
-
"importedPatchPresent": false,
|
|
459
|
-
"verificationAttempted": true
|
|
460
|
-
},
|
|
461
|
-
"repairProgress": {
|
|
462
|
-
"repeatedResults": 0,
|
|
463
|
-
"suppressedResults": 0,
|
|
464
|
-
"blockedBroadCalls": 0,
|
|
465
|
-
"enteredClosure": true,
|
|
466
|
-
"phase": "verify",
|
|
467
|
-
"blockedPhaseCalls": 0,
|
|
468
|
-
"planRecorded": true
|
|
469
|
-
}
|
|
470
|
-
}
|
|
471
|
-
]
|
|
472
|
-
},
|
|
473
|
-
"qwen": {
|
|
474
|
-
"model": "qwen3.8-flash",
|
|
475
|
-
"reasoning": "none",
|
|
476
|
-
"transport": "qwen-chat-completions",
|
|
477
|
-
"limits": {
|
|
478
|
-
"steps": 24,
|
|
479
|
-
"outputPerTurn": 2048,
|
|
480
|
-
"inputTokens": 100000,
|
|
481
|
-
"outputTokens": 16000,
|
|
482
|
-
"timeoutSeconds": 300,
|
|
483
|
-
"memoryMb": 2048,
|
|
484
|
-
"cpus": 2
|
|
485
|
-
},
|
|
486
|
-
"report": {
|
|
487
|
-
"schemaVersion": 1,
|
|
488
|
-
"kind": "swebench-comparison",
|
|
489
|
-
"manifestSha256": "70e9d46277c326551ad59827de191eebcc8cc203493e3b05ef8f188ea6523e5a",
|
|
490
|
-
"complete": true,
|
|
491
|
-
"candidates": {
|
|
492
|
-
"zhivex": {
|
|
493
|
-
"planned": 6,
|
|
494
|
-
"recorded": 6,
|
|
495
|
-
"resolved": 0,
|
|
496
|
-
"candidateCaptured": 6,
|
|
497
|
-
"candidateResolved": 2,
|
|
498
|
-
"candidateGraded": 6,
|
|
499
|
-
"resolutionRate": 0.0,
|
|
500
|
-
"missing": 0,
|
|
501
|
-
"gradingFailures": 0,
|
|
502
|
-
"usageComplete": true,
|
|
503
|
-
"totalCostUsd": null,
|
|
504
|
-
"costPerResolvedUsd": null,
|
|
505
|
-
"resolvedP50Ms": null
|
|
506
|
-
},
|
|
507
|
-
"mini-swe-agent": {
|
|
508
|
-
"planned": 6,
|
|
509
|
-
"recorded": 6,
|
|
510
|
-
"resolved": 3,
|
|
511
|
-
"candidateCaptured": 6,
|
|
512
|
-
"candidateResolved": 3,
|
|
513
|
-
"candidateGraded": 6,
|
|
514
|
-
"resolutionRate": 0.5,
|
|
515
|
-
"missing": 0,
|
|
516
|
-
"gradingFailures": 0,
|
|
517
|
-
"usageComplete": true,
|
|
518
|
-
"totalCostUsd": null,
|
|
519
|
-
"costPerResolvedUsd": null,
|
|
520
|
-
"resolvedP50Ms": 55649.54887499334
|
|
521
|
-
}
|
|
522
|
-
},
|
|
523
|
-
"pairedTaskResolutionDifference": -0.5,
|
|
524
|
-
"pairedTaskBootstrap95": [
|
|
525
|
-
-1.0,
|
|
526
|
-
0.0
|
|
527
|
-
],
|
|
528
|
-
"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."
|
|
529
|
-
},
|
|
530
|
-
"samples": [
|
|
531
|
-
{
|
|
532
|
-
"candidate": "zhivex",
|
|
533
|
-
"instanceId": "django__django-15128",
|
|
534
|
-
"repetition": 0,
|
|
535
|
-
"status": "completed",
|
|
536
|
-
"resolved": false,
|
|
537
|
-
"candidateResolved": false,
|
|
538
|
-
"candidateCaptured": true,
|
|
539
|
-
"gradingStatus": "completed",
|
|
540
|
-
"candidateGradingStatus": "completed",
|
|
541
|
-
"inputTokens": 40710,
|
|
542
|
-
"outputTokens": 455,
|
|
543
|
-
"cachedInputTokens": 24576,
|
|
544
|
-
"usageComplete": true,
|
|
545
|
-
"failure": null,
|
|
546
|
-
"repairStages": {
|
|
547
|
-
"candidatePresent": false,
|
|
548
|
-
"importedPatchPresent": false,
|
|
549
|
-
"verificationAttempted": false
|
|
550
|
-
},
|
|
551
|
-
"repairProgress": {
|
|
552
|
-
"repeatedResults": 0,
|
|
553
|
-
"suppressedResults": 0,
|
|
554
|
-
"blockedBroadCalls": 0,
|
|
555
|
-
"enteredClosure": false,
|
|
556
|
-
"phase": "explore",
|
|
557
|
-
"blockedPhaseCalls": 0,
|
|
558
|
-
"planRecorded": false
|
|
559
|
-
}
|
|
560
|
-
},
|
|
561
|
-
{
|
|
562
|
-
"candidate": "mini-swe-agent",
|
|
563
|
-
"instanceId": "django__django-15128",
|
|
564
|
-
"repetition": 0,
|
|
565
|
-
"status": "LimitsExceeded",
|
|
566
|
-
"resolved": false,
|
|
567
|
-
"candidateResolved": false,
|
|
568
|
-
"candidateCaptured": true,
|
|
569
|
-
"gradingStatus": "completed",
|
|
570
|
-
"candidateGradingStatus": "completed",
|
|
571
|
-
"inputTokens": 105777,
|
|
572
|
-
"outputTokens": 1616,
|
|
573
|
-
"cachedInputTokens": 89344,
|
|
574
|
-
"usageComplete": true,
|
|
575
|
-
"failure": null,
|
|
576
|
-
"repairStages": {
|
|
577
|
-
"candidatePresent": false,
|
|
578
|
-
"importedPatchPresent": null,
|
|
579
|
-
"verificationAttempted": null
|
|
580
|
-
},
|
|
581
|
-
"repairProgress": null
|
|
582
|
-
},
|
|
583
|
-
{
|
|
584
|
-
"candidate": "mini-swe-agent",
|
|
585
|
-
"instanceId": "django__django-15128",
|
|
586
|
-
"repetition": 1,
|
|
587
|
-
"status": "LimitsExceeded",
|
|
588
|
-
"resolved": false,
|
|
589
|
-
"candidateResolved": false,
|
|
590
|
-
"candidateCaptured": true,
|
|
591
|
-
"gradingStatus": "completed",
|
|
592
|
-
"candidateGradingStatus": "completed",
|
|
593
|
-
"inputTokens": 110108,
|
|
594
|
-
"outputTokens": 2110,
|
|
595
|
-
"cachedInputTokens": 94208,
|
|
596
|
-
"usageComplete": true,
|
|
597
|
-
"failure": null,
|
|
598
|
-
"repairStages": {
|
|
599
|
-
"candidatePresent": false,
|
|
600
|
-
"importedPatchPresent": null,
|
|
601
|
-
"verificationAttempted": null
|
|
602
|
-
},
|
|
603
|
-
"repairProgress": null
|
|
604
|
-
},
|
|
605
|
-
{
|
|
606
|
-
"candidate": "zhivex",
|
|
607
|
-
"instanceId": "django__django-15128",
|
|
608
|
-
"repetition": 1,
|
|
609
|
-
"status": "failed",
|
|
610
|
-
"resolved": false,
|
|
611
|
-
"candidateResolved": false,
|
|
612
|
-
"candidateCaptured": true,
|
|
613
|
-
"gradingStatus": "completed",
|
|
614
|
-
"candidateGradingStatus": "completed",
|
|
615
|
-
"inputTokens": 63362,
|
|
616
|
-
"outputTokens": 647,
|
|
617
|
-
"cachedInputTokens": 47232,
|
|
618
|
-
"usageComplete": true,
|
|
619
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
620
|
-
"repairStages": {
|
|
621
|
-
"candidatePresent": false,
|
|
622
|
-
"importedPatchPresent": false,
|
|
623
|
-
"verificationAttempted": false
|
|
624
|
-
},
|
|
625
|
-
"repairProgress": {
|
|
626
|
-
"repeatedResults": 0,
|
|
627
|
-
"suppressedResults": 0,
|
|
628
|
-
"blockedBroadCalls": 0,
|
|
629
|
-
"enteredClosure": false,
|
|
630
|
-
"phase": "explore",
|
|
631
|
-
"blockedPhaseCalls": 0,
|
|
632
|
-
"planRecorded": false
|
|
633
|
-
}
|
|
634
|
-
},
|
|
635
|
-
{
|
|
636
|
-
"candidate": "zhivex",
|
|
637
|
-
"instanceId": "django__django-15128",
|
|
638
|
-
"repetition": 2,
|
|
639
|
-
"status": "failed",
|
|
640
|
-
"resolved": false,
|
|
641
|
-
"candidateResolved": false,
|
|
642
|
-
"candidateCaptured": true,
|
|
643
|
-
"gradingStatus": "completed",
|
|
644
|
-
"candidateGradingStatus": "completed",
|
|
645
|
-
"inputTokens": 62385,
|
|
646
|
-
"outputTokens": 635,
|
|
647
|
-
"cachedInputTokens": 43520,
|
|
648
|
-
"usageComplete": true,
|
|
649
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
650
|
-
"repairStages": {
|
|
651
|
-
"candidatePresent": false,
|
|
652
|
-
"importedPatchPresent": false,
|
|
653
|
-
"verificationAttempted": false
|
|
654
|
-
},
|
|
655
|
-
"repairProgress": {
|
|
656
|
-
"repeatedResults": 0,
|
|
657
|
-
"suppressedResults": 0,
|
|
658
|
-
"blockedBroadCalls": 0,
|
|
659
|
-
"enteredClosure": false,
|
|
660
|
-
"phase": "explore",
|
|
661
|
-
"blockedPhaseCalls": 0,
|
|
662
|
-
"planRecorded": false
|
|
663
|
-
}
|
|
664
|
-
},
|
|
665
|
-
{
|
|
666
|
-
"candidate": "mini-swe-agent",
|
|
667
|
-
"instanceId": "django__django-15128",
|
|
668
|
-
"repetition": 2,
|
|
669
|
-
"status": "LimitsExceeded",
|
|
670
|
-
"resolved": false,
|
|
671
|
-
"candidateResolved": false,
|
|
672
|
-
"candidateCaptured": true,
|
|
673
|
-
"gradingStatus": "completed",
|
|
674
|
-
"candidateGradingStatus": "completed",
|
|
675
|
-
"inputTokens": 102448,
|
|
676
|
-
"outputTokens": 1358,
|
|
677
|
-
"cachedInputTokens": 82944,
|
|
678
|
-
"usageComplete": true,
|
|
679
|
-
"failure": null,
|
|
680
|
-
"repairStages": {
|
|
681
|
-
"candidatePresent": false,
|
|
682
|
-
"importedPatchPresent": null,
|
|
683
|
-
"verificationAttempted": null
|
|
684
|
-
},
|
|
685
|
-
"repairProgress": null
|
|
686
|
-
},
|
|
687
|
-
{
|
|
688
|
-
"candidate": "mini-swe-agent",
|
|
689
|
-
"instanceId": "django__django-15851",
|
|
690
|
-
"repetition": 0,
|
|
691
|
-
"status": "Submitted",
|
|
692
|
-
"resolved": true,
|
|
693
|
-
"candidateResolved": true,
|
|
694
|
-
"candidateCaptured": true,
|
|
695
|
-
"gradingStatus": "completed",
|
|
696
|
-
"candidateGradingStatus": "completed",
|
|
697
|
-
"inputTokens": 23132,
|
|
698
|
-
"outputTokens": 686,
|
|
699
|
-
"cachedInputTokens": 16384,
|
|
700
|
-
"usageComplete": true,
|
|
701
|
-
"failure": null,
|
|
702
|
-
"repairStages": {
|
|
703
|
-
"candidatePresent": true,
|
|
704
|
-
"importedPatchPresent": null,
|
|
705
|
-
"verificationAttempted": null
|
|
706
|
-
},
|
|
707
|
-
"repairProgress": null
|
|
708
|
-
},
|
|
709
|
-
{
|
|
710
|
-
"candidate": "zhivex",
|
|
711
|
-
"instanceId": "django__django-15851",
|
|
712
|
-
"repetition": 0,
|
|
713
|
-
"status": "failed",
|
|
714
|
-
"resolved": false,
|
|
715
|
-
"candidateResolved": true,
|
|
716
|
-
"candidateCaptured": true,
|
|
717
|
-
"gradingStatus": "completed",
|
|
718
|
-
"candidateGradingStatus": "completed",
|
|
719
|
-
"inputTokens": 65852,
|
|
720
|
-
"outputTokens": 1137,
|
|
721
|
-
"cachedInputTokens": 44032,
|
|
722
|
-
"usageComplete": true,
|
|
723
|
-
"failure": "AGENT_FAILED",
|
|
724
|
-
"repairStages": {
|
|
725
|
-
"candidatePresent": true,
|
|
726
|
-
"importedPatchPresent": false,
|
|
727
|
-
"verificationAttempted": true
|
|
728
|
-
},
|
|
729
|
-
"repairProgress": {
|
|
730
|
-
"repeatedResults": 0,
|
|
731
|
-
"suppressedResults": 0,
|
|
732
|
-
"blockedBroadCalls": 0,
|
|
733
|
-
"enteredClosure": false,
|
|
734
|
-
"phase": "verify",
|
|
735
|
-
"blockedPhaseCalls": 0,
|
|
736
|
-
"planRecorded": true
|
|
737
|
-
}
|
|
738
|
-
},
|
|
739
|
-
{
|
|
740
|
-
"candidate": "zhivex",
|
|
741
|
-
"instanceId": "django__django-15851",
|
|
742
|
-
"repetition": 1,
|
|
743
|
-
"status": "failed",
|
|
744
|
-
"resolved": false,
|
|
745
|
-
"candidateResolved": true,
|
|
746
|
-
"candidateCaptured": true,
|
|
747
|
-
"gradingStatus": "completed",
|
|
748
|
-
"candidateGradingStatus": "completed",
|
|
749
|
-
"inputTokens": 98158,
|
|
750
|
-
"outputTokens": 1084,
|
|
751
|
-
"cachedInputTokens": 75776,
|
|
752
|
-
"usageComplete": true,
|
|
753
|
-
"failure": "INPUT_TOKEN_BUDGET",
|
|
754
|
-
"repairStages": {
|
|
755
|
-
"candidatePresent": true,
|
|
756
|
-
"importedPatchPresent": false,
|
|
757
|
-
"verificationAttempted": true
|
|
758
|
-
},
|
|
759
|
-
"repairProgress": {
|
|
760
|
-
"repeatedResults": 0,
|
|
761
|
-
"suppressedResults": 0,
|
|
762
|
-
"blockedBroadCalls": 0,
|
|
763
|
-
"enteredClosure": true,
|
|
764
|
-
"phase": "verify",
|
|
765
|
-
"blockedPhaseCalls": 1,
|
|
766
|
-
"planRecorded": true
|
|
767
|
-
}
|
|
768
|
-
},
|
|
769
|
-
{
|
|
770
|
-
"candidate": "mini-swe-agent",
|
|
771
|
-
"instanceId": "django__django-15851",
|
|
772
|
-
"repetition": 1,
|
|
773
|
-
"status": "Submitted",
|
|
774
|
-
"resolved": true,
|
|
775
|
-
"candidateResolved": true,
|
|
776
|
-
"candidateCaptured": true,
|
|
777
|
-
"gradingStatus": "completed",
|
|
778
|
-
"candidateGradingStatus": "completed",
|
|
779
|
-
"inputTokens": 23878,
|
|
780
|
-
"outputTokens": 745,
|
|
781
|
-
"cachedInputTokens": 17408,
|
|
782
|
-
"usageComplete": true,
|
|
783
|
-
"failure": null,
|
|
784
|
-
"repairStages": {
|
|
785
|
-
"candidatePresent": true,
|
|
786
|
-
"importedPatchPresent": null,
|
|
787
|
-
"verificationAttempted": null
|
|
788
|
-
},
|
|
789
|
-
"repairProgress": null
|
|
790
|
-
},
|
|
791
|
-
{
|
|
792
|
-
"candidate": "mini-swe-agent",
|
|
793
|
-
"instanceId": "django__django-15851",
|
|
794
|
-
"repetition": 2,
|
|
795
|
-
"status": "Submitted",
|
|
796
|
-
"resolved": true,
|
|
797
|
-
"candidateResolved": true,
|
|
798
|
-
"candidateCaptured": true,
|
|
799
|
-
"gradingStatus": "completed",
|
|
800
|
-
"candidateGradingStatus": "completed",
|
|
801
|
-
"inputTokens": 27959,
|
|
802
|
-
"outputTokens": 738,
|
|
803
|
-
"cachedInputTokens": 20736,
|
|
804
|
-
"usageComplete": true,
|
|
805
|
-
"failure": null,
|
|
806
|
-
"repairStages": {
|
|
807
|
-
"candidatePresent": true,
|
|
808
|
-
"importedPatchPresent": null,
|
|
809
|
-
"verificationAttempted": null
|
|
810
|
-
},
|
|
811
|
-
"repairProgress": null
|
|
812
|
-
},
|
|
813
|
-
{
|
|
814
|
-
"candidate": "zhivex",
|
|
815
|
-
"instanceId": "django__django-15851",
|
|
816
|
-
"repetition": 2,
|
|
817
|
-
"status": "failed",
|
|
818
|
-
"resolved": false,
|
|
819
|
-
"candidateResolved": false,
|
|
820
|
-
"candidateCaptured": true,
|
|
821
|
-
"gradingStatus": "completed",
|
|
822
|
-
"candidateGradingStatus": "completed",
|
|
823
|
-
"inputTokens": 62918,
|
|
824
|
-
"outputTokens": 759,
|
|
825
|
-
"cachedInputTokens": 48128,
|
|
826
|
-
"usageComplete": true,
|
|
827
|
-
"failure": "WORK_TOKEN_BUDGET",
|
|
828
|
-
"repairStages": {
|
|
829
|
-
"candidatePresent": false,
|
|
830
|
-
"importedPatchPresent": false,
|
|
831
|
-
"verificationAttempted": false
|
|
832
|
-
},
|
|
833
|
-
"repairProgress": {
|
|
834
|
-
"repeatedResults": 0,
|
|
835
|
-
"suppressedResults": 0,
|
|
836
|
-
"blockedBroadCalls": 0,
|
|
837
|
-
"enteredClosure": false,
|
|
838
|
-
"phase": "explore",
|
|
839
|
-
"blockedPhaseCalls": 0,
|
|
840
|
-
"planRecorded": true
|
|
841
|
-
}
|
|
842
|
-
}
|
|
843
|
-
]
|
|
844
|
-
}
|
|
845
|
-
},
|
|
846
|
-
"limitations": [
|
|
847
|
-
"Two new tasks, both Django; not a general performance estimate.",
|
|
848
|
-
"Providers ran concurrently on one host; timing is not a controlled comparison.",
|
|
849
|
-
"Model, reasoning and transport vary across providers.",
|
|
850
|
-
"No before/after controller ablation was run.",
|
|
851
|
-
"No supplied prices; no monetary cost claim.",
|
|
852
|
-
"Correct private candidates are diagnostic only; only independently accepted delivered patches count."
|
|
853
|
-
]
|
|
854
|
-
}
|