instar 1.3.982 → 1.3.984
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/dist/core/CredentialManualLevers.js +0 -0
- package/dist/core/CredentialManualLevers.js.map +1 -1
- package/dist/core/DeliverMessageHandler.js +0 -0
- package/dist/core/DeliverMessageHandler.js.map +1 -1
- package/dist/core/PeerEndpointResolver.d.ts +0 -0
- package/dist/core/PeerEndpointResolver.d.ts.map +1 -1
- package/dist/core/PeerEndpointResolver.js +0 -0
- package/dist/core/PeerEndpointResolver.js.map +1 -1
- package/dist/core/ProjectRoundExecution.d.ts +45 -2
- package/dist/core/ProjectRoundExecution.d.ts.map +1 -1
- package/dist/core/ProjectRoundExecution.js +56 -7
- package/dist/core/ProjectRoundExecution.js.map +1 -1
- package/dist/core/QueueDrainLoop.js +0 -0
- package/dist/core/QueueDrainLoop.js.map +1 -1
- package/dist/core/ScopeAccretionSweep.js +1 -1
- package/dist/core/ScopeAccretionSweep.js.map +1 -1
- package/dist/core/SessionOwnership.js +1 -1
- package/dist/core/SessionOwnership.js.map +1 -1
- package/dist/core/StageTransitionValidator.d.ts +26 -0
- package/dist/core/StageTransitionValidator.d.ts.map +1 -1
- package/dist/core/StageTransitionValidator.js +11 -1
- package/dist/core/StageTransitionValidator.js.map +1 -1
- package/dist/core/StandardsEnforcementAuditor.js +1 -1
- package/dist/core/StandardsEnforcementAuditor.js.map +1 -1
- package/dist/core/StoreSnapshot.js +2 -2
- package/dist/core/StoreSnapshot.js.map +1 -1
- package/dist/core/UnionReader.js +0 -0
- package/dist/core/UnionReader.js.map +1 -1
- package/dist/core/benchmarkDivergenceCore.js +1 -1
- package/dist/core/benchmarkDivergenceCore.js.map +1 -1
- package/dist/core/cartographerSummary.js +0 -0
- package/dist/core/cartographerSummary.js.map +1 -1
- package/dist/core/routingPriceAuthority.d.ts +0 -0
- package/dist/core/routingPriceAuthority.d.ts.map +1 -1
- package/dist/core/routingPriceAuthority.js +0 -0
- package/dist/core/routingPriceAuthority.js.map +1 -1
- package/dist/messaging/relayContentDedup.js +0 -0
- package/dist/messaging/relayContentDedup.js.map +1 -1
- package/dist/monitoring/BenchmarkDivergenceAnalyzer.js +10 -10
- package/dist/monitoring/BenchmarkDivergenceAnalyzer.js.map +1 -1
- package/dist/monitoring/ExternalHogArmMarker.js +0 -0
- package/dist/monitoring/ExternalHogArmMarker.js.map +1 -1
- package/dist/monitoring/ExternalHogClassifier.js +0 -0
- package/dist/monitoring/ExternalHogClassifier.js.map +1 -1
- package/dist/monitoring/ExternalHogSampler.js +0 -0
- package/dist/monitoring/ExternalHogSampler.js.map +1 -1
- package/dist/monitoring/FeatureMetricsLedger.js +1 -1
- package/dist/monitoring/FeatureMetricsLedger.js.map +1 -1
- package/dist/monitoring/GreenPrAutoMerger.js +1 -1
- package/dist/monitoring/GreenPrAutoMerger.js.map +1 -1
- package/dist/monitoring/PermissionPromptAutoResolver.js +1 -1
- package/dist/monitoring/PermissionPromptAutoResolver.js.map +1 -1
- package/dist/monitoring/ProviderCostReportStore.js +0 -0
- package/dist/monitoring/ProviderCostReportStore.js.map +1 -1
- package/dist/monitoring/blockerSettleAuthority.js +0 -0
- package/dist/monitoring/blockerSettleAuthority.js.map +1 -1
- package/dist/server/routes.d.ts.map +1 -1
- package/dist/server/routes.js +55 -4
- package/dist/server/routes.js.map +1 -1
- package/package.json +1 -1
- package/src/data/builtin-manifest.json +46 -46
- package/upgrades/1.3.982.md +17 -2
- package/upgrades/1.3.983.md +108 -0
- package/upgrades/1.3.984.md +116 -0
- package/upgrades/side-effects/behavioural-promise-unverifiable.md +17 -2
- package/upgrades/side-effects/merged-record-carries-its-evidence.md +230 -0
- package/upgrades/side-effects/raw-nul-bytes-hide-source-from-grep.md +199 -0
package/package.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
{
|
|
2
2
|
"$schema": "./builtin-manifest.schema.json",
|
|
3
3
|
"schemaVersion": 1,
|
|
4
|
-
"generatedAt": "2026-07-
|
|
5
|
-
"instarVersion": "1.3.
|
|
4
|
+
"generatedAt": "2026-07-26T08:47:20.129Z",
|
|
5
|
+
"instarVersion": "1.3.984",
|
|
6
6
|
"entryCount": 202,
|
|
7
7
|
"entries": {
|
|
8
8
|
"hook:session-start": {
|
|
@@ -418,7 +418,7 @@
|
|
|
418
418
|
"type": "route-group",
|
|
419
419
|
"domain": "monitoring",
|
|
420
420
|
"sourcePath": "src/server/routes.ts",
|
|
421
|
-
"contentHash": "
|
|
421
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
422
422
|
"since": "2025-01-01"
|
|
423
423
|
},
|
|
424
424
|
"route-group:agents": {
|
|
@@ -426,7 +426,7 @@
|
|
|
426
426
|
"type": "route-group",
|
|
427
427
|
"domain": "sessions",
|
|
428
428
|
"sourcePath": "src/server/routes.ts",
|
|
429
|
-
"contentHash": "
|
|
429
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
430
430
|
"since": "2025-01-01"
|
|
431
431
|
},
|
|
432
432
|
"route-group:backups": {
|
|
@@ -434,7 +434,7 @@
|
|
|
434
434
|
"type": "route-group",
|
|
435
435
|
"domain": "operations",
|
|
436
436
|
"sourcePath": "src/server/routes.ts",
|
|
437
|
-
"contentHash": "
|
|
437
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
438
438
|
"since": "2025-01-01"
|
|
439
439
|
},
|
|
440
440
|
"route-group:git": {
|
|
@@ -442,7 +442,7 @@
|
|
|
442
442
|
"type": "route-group",
|
|
443
443
|
"domain": "coordination",
|
|
444
444
|
"sourcePath": "src/server/routes.ts",
|
|
445
|
-
"contentHash": "
|
|
445
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
446
446
|
"since": "2025-01-01"
|
|
447
447
|
},
|
|
448
448
|
"route-group:memory": {
|
|
@@ -450,7 +450,7 @@
|
|
|
450
450
|
"type": "route-group",
|
|
451
451
|
"domain": "memory",
|
|
452
452
|
"sourcePath": "src/server/routes.ts",
|
|
453
|
-
"contentHash": "
|
|
453
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
454
454
|
"since": "2025-01-01"
|
|
455
455
|
},
|
|
456
456
|
"route-group:semantic": {
|
|
@@ -458,7 +458,7 @@
|
|
|
458
458
|
"type": "route-group",
|
|
459
459
|
"domain": "memory",
|
|
460
460
|
"sourcePath": "src/server/routes.ts",
|
|
461
|
-
"contentHash": "
|
|
461
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
462
462
|
"since": "2025-01-01"
|
|
463
463
|
},
|
|
464
464
|
"route-group:status": {
|
|
@@ -466,7 +466,7 @@
|
|
|
466
466
|
"type": "route-group",
|
|
467
467
|
"domain": "monitoring",
|
|
468
468
|
"sourcePath": "src/server/routes.ts",
|
|
469
|
-
"contentHash": "
|
|
469
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
470
470
|
"since": "2025-01-01"
|
|
471
471
|
},
|
|
472
472
|
"route-group:capabilities": {
|
|
@@ -474,7 +474,7 @@
|
|
|
474
474
|
"type": "route-group",
|
|
475
475
|
"domain": "mapping",
|
|
476
476
|
"sourcePath": "src/server/routes.ts",
|
|
477
|
-
"contentHash": "
|
|
477
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
478
478
|
"since": "2025-01-01"
|
|
479
479
|
},
|
|
480
480
|
"route-group:project-map": {
|
|
@@ -482,7 +482,7 @@
|
|
|
482
482
|
"type": "route-group",
|
|
483
483
|
"domain": "mapping",
|
|
484
484
|
"sourcePath": "src/server/routes.ts",
|
|
485
|
-
"contentHash": "
|
|
485
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
486
486
|
"since": "2025-01-01"
|
|
487
487
|
},
|
|
488
488
|
"route-group:coherence": {
|
|
@@ -490,7 +490,7 @@
|
|
|
490
490
|
"type": "route-group",
|
|
491
491
|
"domain": "coherence",
|
|
492
492
|
"sourcePath": "src/server/routes.ts",
|
|
493
|
-
"contentHash": "
|
|
493
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
494
494
|
"since": "2025-01-01"
|
|
495
495
|
},
|
|
496
496
|
"route-group:topic-bindings": {
|
|
@@ -498,7 +498,7 @@
|
|
|
498
498
|
"type": "route-group",
|
|
499
499
|
"domain": "sessions",
|
|
500
500
|
"sourcePath": "src/server/routes.ts",
|
|
501
|
-
"contentHash": "
|
|
501
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
502
502
|
"since": "2025-01-01"
|
|
503
503
|
},
|
|
504
504
|
"route-group:context": {
|
|
@@ -506,7 +506,7 @@
|
|
|
506
506
|
"type": "route-group",
|
|
507
507
|
"domain": "context",
|
|
508
508
|
"sourcePath": "src/server/routes.ts",
|
|
509
|
-
"contentHash": "
|
|
509
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
510
510
|
"since": "2025-01-01"
|
|
511
511
|
},
|
|
512
512
|
"route-group:scope-coherence": {
|
|
@@ -514,7 +514,7 @@
|
|
|
514
514
|
"type": "route-group",
|
|
515
515
|
"domain": "coherence",
|
|
516
516
|
"sourcePath": "src/server/routes.ts",
|
|
517
|
-
"contentHash": "
|
|
517
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
518
518
|
"since": "2025-01-01"
|
|
519
519
|
},
|
|
520
520
|
"route-group:canonical-state": {
|
|
@@ -522,7 +522,7 @@
|
|
|
522
522
|
"type": "route-group",
|
|
523
523
|
"domain": "state",
|
|
524
524
|
"sourcePath": "src/server/routes.ts",
|
|
525
|
-
"contentHash": "
|
|
525
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
526
526
|
"since": "2025-01-01"
|
|
527
527
|
},
|
|
528
528
|
"route-group:ci": {
|
|
@@ -530,7 +530,7 @@
|
|
|
530
530
|
"type": "route-group",
|
|
531
531
|
"domain": "monitoring",
|
|
532
532
|
"sourcePath": "src/server/routes.ts",
|
|
533
|
-
"contentHash": "
|
|
533
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
534
534
|
"since": "2025-01-01"
|
|
535
535
|
},
|
|
536
536
|
"route-group:sessions": {
|
|
@@ -538,7 +538,7 @@
|
|
|
538
538
|
"type": "route-group",
|
|
539
539
|
"domain": "sessions",
|
|
540
540
|
"sourcePath": "src/server/routes.ts",
|
|
541
|
-
"contentHash": "
|
|
541
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
542
542
|
"since": "2025-01-01"
|
|
543
543
|
},
|
|
544
544
|
"route-group:jobs": {
|
|
@@ -546,7 +546,7 @@
|
|
|
546
546
|
"type": "route-group",
|
|
547
547
|
"domain": "scheduling",
|
|
548
548
|
"sourcePath": "src/server/routes.ts",
|
|
549
|
-
"contentHash": "
|
|
549
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
550
550
|
"since": "2025-01-01"
|
|
551
551
|
},
|
|
552
552
|
"route-group:skip-ledger": {
|
|
@@ -554,7 +554,7 @@
|
|
|
554
554
|
"type": "route-group",
|
|
555
555
|
"domain": "scheduling",
|
|
556
556
|
"sourcePath": "src/server/routes.ts",
|
|
557
|
-
"contentHash": "
|
|
557
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
558
558
|
"since": "2025-01-01"
|
|
559
559
|
},
|
|
560
560
|
"route-group:telegram": {
|
|
@@ -562,7 +562,7 @@
|
|
|
562
562
|
"type": "route-group",
|
|
563
563
|
"domain": "communication",
|
|
564
564
|
"sourcePath": "src/server/routes.ts",
|
|
565
|
-
"contentHash": "
|
|
565
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
566
566
|
"since": "2025-01-01"
|
|
567
567
|
},
|
|
568
568
|
"route-group:attention": {
|
|
@@ -570,7 +570,7 @@
|
|
|
570
570
|
"type": "route-group",
|
|
571
571
|
"domain": "communication",
|
|
572
572
|
"sourcePath": "src/server/routes.ts",
|
|
573
|
-
"contentHash": "
|
|
573
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
574
574
|
"since": "2025-01-01"
|
|
575
575
|
},
|
|
576
576
|
"route-group:relationships": {
|
|
@@ -578,7 +578,7 @@
|
|
|
578
578
|
"type": "route-group",
|
|
579
579
|
"domain": "relationships",
|
|
580
580
|
"sourcePath": "src/server/routes.ts",
|
|
581
|
-
"contentHash": "
|
|
581
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
582
582
|
"since": "2025-01-01"
|
|
583
583
|
},
|
|
584
584
|
"route-group:feedback": {
|
|
@@ -586,7 +586,7 @@
|
|
|
586
586
|
"type": "route-group",
|
|
587
587
|
"domain": "feedback",
|
|
588
588
|
"sourcePath": "src/server/routes.ts",
|
|
589
|
-
"contentHash": "
|
|
589
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
590
590
|
"since": "2025-01-01"
|
|
591
591
|
},
|
|
592
592
|
"route-group:updates": {
|
|
@@ -594,7 +594,7 @@
|
|
|
594
594
|
"type": "route-group",
|
|
595
595
|
"domain": "updates",
|
|
596
596
|
"sourcePath": "src/server/routes.ts",
|
|
597
|
-
"contentHash": "
|
|
597
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
598
598
|
"since": "2025-01-01"
|
|
599
599
|
},
|
|
600
600
|
"route-group:dispatches": {
|
|
@@ -602,7 +602,7 @@
|
|
|
602
602
|
"type": "route-group",
|
|
603
603
|
"domain": "dispatches",
|
|
604
604
|
"sourcePath": "src/server/routes.ts",
|
|
605
|
-
"contentHash": "
|
|
605
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
606
606
|
"since": "2025-01-01"
|
|
607
607
|
},
|
|
608
608
|
"route-group:quota": {
|
|
@@ -610,7 +610,7 @@
|
|
|
610
610
|
"type": "route-group",
|
|
611
611
|
"domain": "monitoring",
|
|
612
612
|
"sourcePath": "src/server/routes.ts",
|
|
613
|
-
"contentHash": "
|
|
613
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
614
614
|
"since": "2025-01-01"
|
|
615
615
|
},
|
|
616
616
|
"route-group:publishing": {
|
|
@@ -618,7 +618,7 @@
|
|
|
618
618
|
"type": "route-group",
|
|
619
619
|
"domain": "publishing",
|
|
620
620
|
"sourcePath": "src/server/routes.ts",
|
|
621
|
-
"contentHash": "
|
|
621
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
622
622
|
"since": "2025-01-01"
|
|
623
623
|
},
|
|
624
624
|
"route-group:private-views": {
|
|
@@ -626,7 +626,7 @@
|
|
|
626
626
|
"type": "route-group",
|
|
627
627
|
"domain": "publishing",
|
|
628
628
|
"sourcePath": "src/server/routes.ts",
|
|
629
|
-
"contentHash": "
|
|
629
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
630
630
|
"since": "2025-01-01"
|
|
631
631
|
},
|
|
632
632
|
"route-group:tunnel": {
|
|
@@ -634,7 +634,7 @@
|
|
|
634
634
|
"type": "route-group",
|
|
635
635
|
"domain": "networking",
|
|
636
636
|
"sourcePath": "src/server/routes.ts",
|
|
637
|
-
"contentHash": "
|
|
637
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
638
638
|
"since": "2025-01-01"
|
|
639
639
|
},
|
|
640
640
|
"route-group:events": {
|
|
@@ -642,7 +642,7 @@
|
|
|
642
642
|
"type": "route-group",
|
|
643
643
|
"domain": "networking",
|
|
644
644
|
"sourcePath": "src/server/routes.ts",
|
|
645
|
-
"contentHash": "
|
|
645
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
646
646
|
"since": "2025-01-01"
|
|
647
647
|
},
|
|
648
648
|
"route-group:evolution": {
|
|
@@ -650,7 +650,7 @@
|
|
|
650
650
|
"type": "route-group",
|
|
651
651
|
"domain": "evolution",
|
|
652
652
|
"sourcePath": "src/server/routes.ts",
|
|
653
|
-
"contentHash": "
|
|
653
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
654
654
|
"since": "2025-01-01"
|
|
655
655
|
},
|
|
656
656
|
"route-group:watchdog": {
|
|
@@ -658,7 +658,7 @@
|
|
|
658
658
|
"type": "route-group",
|
|
659
659
|
"domain": "monitoring",
|
|
660
660
|
"sourcePath": "src/server/routes.ts",
|
|
661
|
-
"contentHash": "
|
|
661
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
662
662
|
"since": "2025-01-01"
|
|
663
663
|
},
|
|
664
664
|
"route-group:topic-memory": {
|
|
@@ -666,7 +666,7 @@
|
|
|
666
666
|
"type": "route-group",
|
|
667
667
|
"domain": "memory",
|
|
668
668
|
"sourcePath": "src/server/routes.ts",
|
|
669
|
-
"contentHash": "
|
|
669
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
670
670
|
"since": "2025-01-01"
|
|
671
671
|
},
|
|
672
672
|
"route-group:state-sync": {
|
|
@@ -674,7 +674,7 @@
|
|
|
674
674
|
"type": "route-group",
|
|
675
675
|
"domain": "coordination",
|
|
676
676
|
"sourcePath": "src/server/routes.ts",
|
|
677
|
-
"contentHash": "
|
|
677
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
678
678
|
"since": "2025-01-01"
|
|
679
679
|
},
|
|
680
680
|
"route-group:intent": {
|
|
@@ -682,7 +682,7 @@
|
|
|
682
682
|
"type": "route-group",
|
|
683
683
|
"domain": "intent",
|
|
684
684
|
"sourcePath": "src/server/routes.ts",
|
|
685
|
-
"contentHash": "
|
|
685
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
686
686
|
"since": "2025-01-01"
|
|
687
687
|
},
|
|
688
688
|
"route-group:triage": {
|
|
@@ -690,7 +690,7 @@
|
|
|
690
690
|
"type": "route-group",
|
|
691
691
|
"domain": "safety",
|
|
692
692
|
"sourcePath": "src/server/routes.ts",
|
|
693
|
-
"contentHash": "
|
|
693
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
694
694
|
"since": "2025-01-01"
|
|
695
695
|
},
|
|
696
696
|
"route-group:operations": {
|
|
@@ -698,7 +698,7 @@
|
|
|
698
698
|
"type": "route-group",
|
|
699
699
|
"domain": "safety",
|
|
700
700
|
"sourcePath": "src/server/routes.ts",
|
|
701
|
-
"contentHash": "
|
|
701
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
702
702
|
"since": "2025-01-01"
|
|
703
703
|
},
|
|
704
704
|
"route-group:sentinel": {
|
|
@@ -706,7 +706,7 @@
|
|
|
706
706
|
"type": "route-group",
|
|
707
707
|
"domain": "safety",
|
|
708
708
|
"sourcePath": "src/server/routes.ts",
|
|
709
|
-
"contentHash": "
|
|
709
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
710
710
|
"since": "2025-01-01"
|
|
711
711
|
},
|
|
712
712
|
"route-group:trust": {
|
|
@@ -714,7 +714,7 @@
|
|
|
714
714
|
"type": "route-group",
|
|
715
715
|
"domain": "safety",
|
|
716
716
|
"sourcePath": "src/server/routes.ts",
|
|
717
|
-
"contentHash": "
|
|
717
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
718
718
|
"since": "2025-01-01"
|
|
719
719
|
},
|
|
720
720
|
"route-group:monitoring": {
|
|
@@ -722,7 +722,7 @@
|
|
|
722
722
|
"type": "route-group",
|
|
723
723
|
"domain": "monitoring",
|
|
724
724
|
"sourcePath": "src/server/routes.ts",
|
|
725
|
-
"contentHash": "
|
|
725
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
726
726
|
"since": "2025-01-01"
|
|
727
727
|
},
|
|
728
728
|
"route-group:commitments": {
|
|
@@ -730,7 +730,7 @@
|
|
|
730
730
|
"type": "route-group",
|
|
731
731
|
"domain": "commitments",
|
|
732
732
|
"sourcePath": "src/server/routes.ts",
|
|
733
|
-
"contentHash": "
|
|
733
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
734
734
|
"since": "2025-01-01"
|
|
735
735
|
},
|
|
736
736
|
"route-group:episodes": {
|
|
@@ -738,7 +738,7 @@
|
|
|
738
738
|
"type": "route-group",
|
|
739
739
|
"domain": "memory",
|
|
740
740
|
"sourcePath": "src/server/routes.ts",
|
|
741
|
-
"contentHash": "
|
|
741
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
742
742
|
"since": "2025-01-01"
|
|
743
743
|
},
|
|
744
744
|
"route-group:messages": {
|
|
@@ -746,7 +746,7 @@
|
|
|
746
746
|
"type": "route-group",
|
|
747
747
|
"domain": "coordination",
|
|
748
748
|
"sourcePath": "src/server/routes.ts",
|
|
749
|
-
"contentHash": "
|
|
749
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
750
750
|
"since": "2025-01-01"
|
|
751
751
|
},
|
|
752
752
|
"route-group:system-reviews": {
|
|
@@ -754,7 +754,7 @@
|
|
|
754
754
|
"type": "route-group",
|
|
755
755
|
"domain": "monitoring",
|
|
756
756
|
"sourcePath": "src/server/routes.ts",
|
|
757
|
-
"contentHash": "
|
|
757
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
758
758
|
"since": "2025-01-01"
|
|
759
759
|
},
|
|
760
760
|
"route-group:machine-mesh": {
|
|
@@ -770,7 +770,7 @@
|
|
|
770
770
|
"type": "route-group",
|
|
771
771
|
"domain": "security",
|
|
772
772
|
"sourcePath": "src/server/routes.ts",
|
|
773
|
-
"contentHash": "
|
|
773
|
+
"contentHash": "3d7b20ccbe93ba5e4d23dbacbedfef0e66d8e97c44e0169a90cc21da36f2a5c3",
|
|
774
774
|
"since": "2025-01-01"
|
|
775
775
|
},
|
|
776
776
|
"cli:init": {
|
package/upgrades/1.3.982.md
CHANGED
|
@@ -42,6 +42,17 @@ genuinely reads live state.
|
|
|
42
42
|
|
|
43
43
|
## What to Tell Your User
|
|
44
44
|
|
|
45
|
+
> **CORRECTION (2026-07-26, after release).** The paragraph below over-claimed and is kept
|
|
46
|
+
> visible rather than quietly rewritten. **The change is forward-only.** Promises about
|
|
47
|
+
> behaviour recorded from now on stay open, and the runaway tally stopped for all of them —
|
|
48
|
+
> but promises that already existed **keep the verdict they were last stamped with**, because
|
|
49
|
+
> the fix works by declining to reach a verdict, which means it never revisits one. Measured
|
|
50
|
+
> on a live install after deploy: 98 behavioural records still showed 74 "kept" and 24
|
|
51
|
+
> "broken", unchanged. So if you had behavioural commitments before upgrading, their old and
|
|
52
|
+
> untrustworthy verdicts are still on screen, and the health line will still count those 74 as
|
|
53
|
+
> verified. Clearing them needs a deliberate decision (rewriting stored records, or having the
|
|
54
|
+
> display refuse a verdict for anything unverifiable) and has not been made.
|
|
55
|
+
|
|
45
56
|
If you track commitments, promises about behaviour now show as **pending** rather than
|
|
46
57
|
verified or violated, with both counters at zero. That is more honest, not a regression:
|
|
47
58
|
nothing was ever watching whether those promises were kept, so "verified" was reassurance
|
|
@@ -69,10 +80,14 @@ commitment"`. Live store corroboration: 74/74 with the field `verified`, 24/24 w
|
|
|
69
80
|
|
|
70
81
|
**Observed after,** same scenarios against the built code:
|
|
71
82
|
|
|
83
|
+
**Scope note added 2026-07-26:** every row below was measured on a **newly created** commitment.
|
|
84
|
+
A row that already existed before the upgrade keeps its stored verdict — see the correction above.
|
|
85
|
+
|
|
72
86
|
| scenario | before | after (25 sweeps) |
|
|
73
87
|
|---|---|---|
|
|
74
|
-
| behavioural, `behavioralRule` set | `verified`, +1 tick per sweep | `pending`, verifications 0, violations 0 |
|
|
75
|
-
| behavioural, field absent | `violated`, +1 tick per sweep | `pending`, verifications 0, violations 0 |
|
|
88
|
+
| behavioural, `behavioralRule` set (NEW row) | `verified`, +1 tick per sweep | `pending`, verifications 0, violations 0 |
|
|
89
|
+
| behavioural, field absent (NEW row) | `violated`, +1 tick per sweep | `pending`, verifications 0, violations 0 |
|
|
90
|
+
| behavioural row that PREDATES the change | `verified` or `violated`, ticking | ticking stops; **stored verdict unchanged** |
|
|
76
91
|
| rules file deleted | produced a verdict | no verdict; file self-heals; status unchanged |
|
|
77
92
|
| health, 1 behavioural row | `"1 commitment(s) tracked, all verified"` | `"1 commitment(s) tracked — 0 verified, 1 not automatically verifiable (no violations)"`, status `healthy` |
|
|
78
93
|
|
|
@@ -0,0 +1,108 @@
|
|
|
1
|
+
# Upgrade Guide — vNEXT
|
|
2
|
+
|
|
3
|
+
<!-- assembled-by: assemble-next-md -->
|
|
4
|
+
<!-- bump: patch -->
|
|
5
|
+
|
|
6
|
+
## What Changed
|
|
7
|
+
|
|
8
|
+
**A project item could read `merged` while carrying no record of what merged it — and
|
|
9
|
+
the two regression detectors that read that record were therefore selecting nothing and
|
|
10
|
+
reporting all clear.**
|
|
11
|
+
|
|
12
|
+
`/projects/:id/advance` proves four things before it will move an item to `merged`: the
|
|
13
|
+
PR state is `MERGED`, it has a format-valid merge commit sha, that sha is reachable from
|
|
14
|
+
canonical main, and the CI rollup is green. It then persisted **only** `pipelineStage`
|
|
15
|
+
and discarded the evidence. Measured on the live store: both Tier-1 children of the
|
|
16
|
+
`convergence-towards-coherence` project read `pipelineStage: "merged"` with no
|
|
17
|
+
`prNumber`, no `mergeCommitOid` and no `ciCheckedAt`.
|
|
18
|
+
|
|
19
|
+
The cause was a type, not carelessness. `StageTransitionResult`'s success case was
|
|
20
|
+
`{ ok: true }` — one bit — while every refusal carried `reason` + `code`. The validator
|
|
21
|
+
explained itself when it said no and said nothing when it said yes, so the caller had
|
|
22
|
+
nothing to write down. A passing verdict now carries a `MergedEvidence` record (PR
|
|
23
|
+
number, merge sha, the canonical-main ref it was actually proven against, and an ISO
|
|
24
|
+
timestamp), persisted in the **same** `update()` as the stage — so no window exists in
|
|
25
|
+
which an item reads `merged` with nothing behind it.
|
|
26
|
+
|
|
27
|
+
**The half that mattered more.** Two merged-state reconcilers find their candidates by
|
|
28
|
+
looking for `mergeCommitOid`: the lazy reconciler on `GET /projects/:id` (documented
|
|
29
|
+
"may mutate", running on every read) and `verifyMergedItemsViaGit`. Nothing ever wrote
|
|
30
|
+
that field, so the candidate set was always empty and the git verification never ran.
|
|
31
|
+
A regression detector that scans nothing and reports nothing is indistinguishable from
|
|
32
|
+
one that scans everything and finds nothing wrong — so merged work that was later
|
|
33
|
+
reverted or force-pushed off main would never have been noticed.
|
|
34
|
+
|
|
35
|
+
Because that path had never executed, three defects sat in it unexercised, **all three
|
|
36
|
+
already fixed in the advance path a few hundred lines away**: it called
|
|
37
|
+
`SafeGitExecutor.run` without `sourceTreeReadOk`, so SourceTreeGuard refuses the read
|
|
38
|
+
against an instar source tree; it hardcoded `origin/main`, which on a dev-agent home is
|
|
39
|
+
the agent's fork rather than where merges land; and its `catch {}` meant the caller
|
|
40
|
+
marked the item `regressed`, turning "I could not check" into "it was reverted". Any one
|
|
41
|
+
of those would have produced false demotions the moment evidence started being written,
|
|
42
|
+
so both halves ship together. `verifyMergedItemsViaGit` now returns three states —
|
|
43
|
+
`verified`, `regressed` (git's documented exit 1, the only genuine negative) and
|
|
44
|
+
`unverifiable` (everything else, with its reason) — and the reconciler demotes only on
|
|
45
|
+
`regressed`; an unverifiable item keeps its stage and takes only the `ciCheckedAt`
|
|
46
|
+
backoff, so the question is asked again later instead of answered wrongly now.
|
|
47
|
+
|
|
48
|
+
**What this does not do:** it adds no new regression checking. It makes the existing
|
|
49
|
+
reconcilers *possible*, by giving them the evidence they were written to read, and
|
|
50
|
+
*safe*, by stopping them guessing. Items merged before this change carry no evidence and
|
|
51
|
+
report honestly as `unverifiable` with the reason "no mergeCommitOid recorded on the
|
|
52
|
+
item" — no backfill is attempted, because reconstructing which PR merged a historical
|
|
53
|
+
item would mean fabricating evidence.
|
|
54
|
+
|
|
55
|
+
## What to Tell Your User
|
|
56
|
+
|
|
57
|
+
If you use project tracking, a finished item now records the pull request, the commit
|
|
58
|
+
and the time it was checked — so "merged" can name what merged it. Previously it could
|
|
59
|
+
not.
|
|
60
|
+
|
|
61
|
+
More usefully: the checks that watch for finished work being undone were, in practice,
|
|
62
|
+
examining an empty list and reporting no problems. They now have something to examine.
|
|
63
|
+
And when one of them cannot answer — no permission, wrong branch, missing commit — it
|
|
64
|
+
says so instead of declaring the work reverted, which is what it used to do.
|
|
65
|
+
|
|
66
|
+
## Summary of New Capabilities
|
|
67
|
+
|
|
68
|
+
- A passing stage-transition verdict carries the evidence it established, instead of a
|
|
69
|
+
bare success flag.
|
|
70
|
+
- A merged item records its PR number, merge commit and check timestamp in the same
|
|
71
|
+
write as its stage.
|
|
72
|
+
- The merged-state verifier distinguishes "reverted" from "could not check", and the
|
|
73
|
+
reconciler demotes only on the former.
|
|
74
|
+
- The reconciler checks the canonical-main ref resolved for the install rather than
|
|
75
|
+
assuming `origin/main`, so a fork-origin agent home no longer reads every healthy item
|
|
76
|
+
as regressed.
|
|
77
|
+
- `POST /projects/:id/advance` echoes the recorded evidence back, so a caller can confirm
|
|
78
|
+
it landed rather than trusting that it did.
|
|
79
|
+
|
|
80
|
+
## Evidence
|
|
81
|
+
|
|
82
|
+
**Reproduction (before):** read any merged child of a project —
|
|
83
|
+
`GET /projects/convergence-towards-coherence` returned both Tier-1 children with
|
|
84
|
+
`pipelineStage: "merged"` and an object whose keys contained no `prNumber`,
|
|
85
|
+
`mergeCommitOid` or `ciCheckedAt`. `grep` for writers of `mergeCommitOid` found none
|
|
86
|
+
outside `InitiativeTracker`'s own accessors, while two consumers select on it.
|
|
87
|
+
|
|
88
|
+
**Observed after,** same scenarios against the built code:
|
|
89
|
+
|
|
90
|
+
| scenario | before | after |
|
|
91
|
+
|---|---|---|
|
|
92
|
+
| passing `building → merged` verdict | `{ ok: true }` | carries `prNumber`, `mergeCommitOid`, `mergeBaseBranch`, `verifiedAt` |
|
|
93
|
+
| the item record after a merged advance | stage only | stage + PR number + merge sha + check timestamp, one write |
|
|
94
|
+
| verifier, git exit 0 | verified | verified |
|
|
95
|
+
| verifier, git exit 1 | not verified → caller demotes | `regressed` → caller demotes |
|
|
96
|
+
| verifier, git exit 128 / guard refusal | not verified → **caller demotes** | `unverifiable` with reason → **stage unchanged** |
|
|
97
|
+
| verifier, child with no recorded evidence | silently skipped | `unverifiable`, "no mergeCommitOid recorded" |
|
|
98
|
+
| verifier's branch | hardcoded `origin/main` | the caller-resolved canonical-main ref |
|
|
99
|
+
|
|
100
|
+
**Tests:** `tests/unit/merged-record-carries-its-evidence.test.ts` (9 new) plus 2 added
|
|
101
|
+
to `tests/unit/StageTransitionValidator.test.ts`. All 9 fail against pre-fix source with
|
|
102
|
+
failures that name the defect — "expected undefined to be defined" for the evidence-free
|
|
103
|
+
approval, "prNumber must be persisted alongside the stage, not discarded after
|
|
104
|
+
validation: expected false to be true", "the verifier readSync must exist: expected +0
|
|
105
|
+
to be 1" (it used `.run`, proving the missing source-tree declaration), and "an item is
|
|
106
|
+
demoted only when git said exit 1, never merely because it was absent from `verified`".
|
|
107
|
+
The 48 pre-existing validator tests and the 8 `ProjectRoundExecution` tests pass
|
|
108
|
+
unchanged. `npx tsc --noEmit` clean.
|
|
@@ -0,0 +1,116 @@
|
|
|
1
|
+
# Upgrade Guide — vNEXT
|
|
2
|
+
|
|
3
|
+
<!-- assembled-by: assemble-next-md -->
|
|
4
|
+
<!-- bump: patch -->
|
|
5
|
+
|
|
6
|
+
## What Changed
|
|
7
|
+
|
|
8
|
+
**Thirty source files were invisible to `grep`. A search across them returned nothing,
|
|
9
|
+
and nothing read as "there is nothing there."**
|
|
10
|
+
|
|
11
|
+
`grep` classifies a file as binary if it contains a NUL (0x00) byte, and on a binary file
|
|
12
|
+
it emits **nothing at all** — not a match, not a `Binary file X matches` line, not even a
|
|
13
|
+
`0` under `-c`. Thirty tracked text files each carried one, so every grep-based audit over
|
|
14
|
+
`src/` silently skipped them. A search that examined two-thirds of the tree was
|
|
15
|
+
byte-for-byte indistinguishable from one that examined all of it and found nothing wrong.
|
|
16
|
+
|
|
17
|
+
Twenty-two were live source. Among them: `blockerSettleAuthority.ts` (the gate that
|
|
18
|
+
decides whether a blocker is genuinely unresolvable), `SessionOwnership.ts`,
|
|
19
|
+
`GreenPrAutoMerger.ts`, `PermissionPromptAutoResolver.ts` (an always-on safety floor), all
|
|
20
|
+
three `ExternalHog*` modules — and `StandardsEnforcementAuditor.ts`, the module that
|
|
21
|
+
audits whether our standards carry structural guards. The auditor of guarantees was itself
|
|
22
|
+
invisible to the standard search instrument.
|
|
23
|
+
|
|
24
|
+
**A second consequence, worse in kind.** Git applies the same rule but only sniffs the
|
|
25
|
+
first 8000 bytes. For the **11 files** whose NUL fell inside that window, `git diff`
|
|
26
|
+
rendered `Bin 5407 -> 5412 bytes` instead of a line diff — so pull requests touching
|
|
27
|
+
safety-critical authority code were reviewed **without the reviewer being shown the
|
|
28
|
+
changed lines**. For the other 19, git saw text while `grep` did not. The two instruments
|
|
29
|
+
disagreed on the same tree, which is exactly why this survived so long: whichever one you
|
|
30
|
+
happened to reach for decided what you believed.
|
|
31
|
+
|
|
32
|
+
**None of it was corruption.** Every byte was a deliberate composite-key or hash
|
|
33
|
+
separator — the classic collision-proof choice, since a NUL cannot occur inside a model
|
|
34
|
+
name or a framework name:
|
|
35
|
+
|
|
36
|
+
```ts
|
|
37
|
+
const key = `${row.model}<a literal 0x00 byte>${row.framework}`;
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
The delimiter is correct. Writing it as the byte rather than its escape is the entire
|
|
41
|
+
defect. All thirty now use the six-character escape, which denotes the identical
|
|
42
|
+
character — so runtime behaviour is unchanged, and every hash derived from these
|
|
43
|
+
separators is byte-identical. Four sites feed them into digests
|
|
44
|
+
(`relayContentDedup`, `blockerSettleAuthority`, `UnionReader`, `ExternalHogArmMarker`);
|
|
45
|
+
this was the primary risk and it was verified rather than assumed. No migration, no cache
|
|
46
|
+
invalidation, no re-keying.
|
|
47
|
+
|
|
48
|
+
A new ratchet, `tests/unit/no-raw-nul-bytes-in-source.test.ts`, fails the build if one
|
|
49
|
+
returns. It reads bytes via `readFileSync` and never shells out to a search tool, so it
|
|
50
|
+
cannot be blinded by the defect it detects. It carries no exemption list, because no case
|
|
51
|
+
needs one: any runtime NUL a program wants is expressible as an escape.
|
|
52
|
+
|
|
53
|
+
**What this does not do:** it fixes one silent instrument, not the class. Other raw
|
|
54
|
+
control bytes (ESC, BEL) remain in a few hostile-input test fixtures — deliberately, after
|
|
55
|
+
verifying empirically that they do **not** cause the grep skip. Only 0x00 does. A lint
|
|
56
|
+
should enforce exactly the failure it is named for, and widening this one for tidiness
|
|
57
|
+
would have added churn while claiming safety it does not provide.
|
|
58
|
+
|
|
59
|
+
## What to Tell Your User
|
|
60
|
+
|
|
61
|
+
Nothing changes in how the agent behaves — this is a source-text and build-time fix with
|
|
62
|
+
no route, setting, or visible surface.
|
|
63
|
+
|
|
64
|
+
What it does change is trustworthiness of process: for months, searching the codebase for
|
|
65
|
+
a given pattern could quietly skip thirty files and report a clean result, and for eleven
|
|
66
|
+
of those files a code review showed "binary file changed" instead of the actual lines. Any
|
|
67
|
+
past conclusion of the form "we don't do X anywhere" may have been drawn from an
|
|
68
|
+
instrument that wasn't looking. Those files are readable again, and a check now fails the
|
|
69
|
+
build if it recurs.
|
|
70
|
+
|
|
71
|
+
## Summary of New Capabilities
|
|
72
|
+
|
|
73
|
+
- Thirty text files that were invisible to `grep` — twenty-two of them live source,
|
|
74
|
+
including several safety-critical authority modules — are searchable again.
|
|
75
|
+
- Eleven files that rendered as binary diffs in code review now show reviewable lines.
|
|
76
|
+
- A build-time ratchet refuses any text file containing a raw NUL byte, naming the file
|
|
77
|
+
and the consequence rather than emitting a bare assertion failure.
|
|
78
|
+
- The ratchet proves its own detector on every run, so a check that has gone dead is
|
|
79
|
+
distinguishable from one that has nothing to report.
|
|
80
|
+
|
|
81
|
+
## Evidence
|
|
82
|
+
|
|
83
|
+
**Reproduction (before):** `grep -c export src/monitoring/blockerSettleAuthority.ts`
|
|
84
|
+
printed nothing — no count, no error, exit status indistinguishable from a clean miss.
|
|
85
|
+
`file` reported the same TypeScript source as `data`, not text. Three one-line fixtures
|
|
86
|
+
isolate the cause: a clean file and a file containing ESC + BEL both return 1 match; a
|
|
87
|
+
file differing only by one NUL byte returns nothing at all.
|
|
88
|
+
|
|
89
|
+
**Observed after,** same scenarios against the fixed tree:
|
|
90
|
+
|
|
91
|
+
| scenario | before | after |
|
|
92
|
+
|---|---|---|
|
|
93
|
+
| `grep -c export` on `blockerSettleAuthority.ts` | *(silent)* | `1` |
|
|
94
|
+
| `grep -c export` on `StandardsEnforcementAuditor.ts` | *(silent)* | `13` |
|
|
95
|
+
| `grep -c export` on `FeatureMetricsLedger.ts` | *(silent)* | `20` |
|
|
96
|
+
| `file` on those sources | `data` | `ASCII text` |
|
|
97
|
+
| `git diff` on the 11 early-NUL files | `Bin NNNN -> NNNN bytes` | reviewable line diff |
|
|
98
|
+
| runtime value of the separator | U+0000 | U+0000 (unchanged) |
|
|
99
|
+
| digests derived from the separator | — | byte-identical, verified |
|
|
100
|
+
| ratchet against a reintroduced NUL | *(no check existed)* | fails, names the file |
|
|
101
|
+
|
|
102
|
+
**Tests:** `tests/unit/no-raw-nul-bytes-in-source.test.ts` (4 new). One asserts the
|
|
103
|
+
detector flags a raw-byte file and clears an escaped one — a dead-check guard, since a
|
|
104
|
+
lint that has never objected is output-identical to a lint that cannot object. One asserts
|
|
105
|
+
the escape decodes to exactly the raw byte, so the behaviour-preserving claim is checked
|
|
106
|
+
rather than stated. One scans the tree. One asserts the scan examined more than 500 files,
|
|
107
|
+
so a scan that walked the wrong roots cannot pass forever by finding nothing.
|
|
108
|
+
|
|
109
|
+
Verified by reintroducing a raw NUL into `FeatureMetricsLedger.ts`: the lint exits 1,
|
|
110
|
+
names the file, and states the consequence. `npx tsc --noEmit` clean; the full unit suite
|
|
111
|
+
runs 39,046 tests with no failure attributable to this change.
|
|
112
|
+
|
|
113
|
+
**Provenance note.** This was found only because a *different* investigation produced a
|
|
114
|
+
suspiciously empty search result that was about to be written up as "no such mechanism
|
|
115
|
+
exists in the codebase." That search had skipped twenty-two files. The finding was
|
|
116
|
+
manufactured by the defect, and has been retracted and superseded rather than deleted.
|