jev-recipes 0.9.11 → 0.9.13

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.
@@ -45,29 +45,29 @@ export const recipeMetadata = [
45
45
  },
46
46
  ],
47
47
  evidence: {
48
- kind: 'earlier',
49
- label: 'Earlier-evaluator measurement',
48
+ kind: 'synthetic',
49
+ label: 'Current synthetic measurement',
50
50
  experimental: true,
51
51
  measurement: {
52
- model: 'typesafe-ai/jev',
53
- date: '2026-09-27T21:19:58.394Z',
52
+ model: 'jev-1.13.0',
53
+ date: '2026-10-04T20:42:48.259Z',
54
54
  operation: 'live',
55
55
  replayedAt: null,
56
- scoringRevision: 1,
57
- split: 'held-out',
58
- cases: 40,
59
- ready: 20,
60
- reviewRate: 0.5,
61
- readyAccuracy: 1,
56
+ scoringRevision: 2,
57
+ split: 'development',
58
+ cases: 100,
59
+ ready: 63,
60
+ reviewRate: 0.37,
61
+ readyAccuracy: 0.984,
62
62
  failed: 0,
63
63
  provenance: [
64
64
  {
65
65
  method: 'author-synthetic',
66
- source: 'AI-authored new scenario families for the September 2026 update; not independently human-reviewed. Previous inspected holdouts moved to development.',
67
- cases: 40,
66
+ source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
67
+ cases: 100,
68
68
  },
69
69
  ],
70
- acceptanceMet: true,
70
+ acceptanceMet: null,
71
71
  },
72
72
  },
73
73
  },
@@ -89,10 +89,26 @@ export const recipeMetadata = [
89
89
  },
90
90
  ],
91
91
  evidence: {
92
- kind: 'unknown',
93
- label: 'Unknown response origin',
92
+ kind: 'unspecified',
93
+ label: 'Current measurement; labels unspecified',
94
94
  experimental: true,
95
- measurement: null,
95
+ measurement: {
96
+ model: 'jev-1.13.0',
97
+ date: '2026-10-04T20:42:51.224Z',
98
+ operation: 'live',
99
+ replayedAt: null,
100
+ scoringRevision: 2,
101
+ split: 'development',
102
+ cases: 44,
103
+ ready: 20,
104
+ reviewRate: 0.545,
105
+ readyAccuracy: 1,
106
+ failed: 0,
107
+ provenance: [
108
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 44 },
109
+ ],
110
+ acceptanceMet: null,
111
+ },
96
112
  },
97
113
  },
98
114
  {
@@ -298,10 +314,26 @@ export const recipeMetadata = [
298
314
  },
299
315
  ],
300
316
  evidence: {
301
- kind: 'unknown',
302
- label: 'Unknown response origin',
317
+ kind: 'unspecified',
318
+ label: 'Current measurement; labels unspecified',
303
319
  experimental: true,
304
- measurement: null,
320
+ measurement: {
321
+ model: 'jev-1.13.0',
322
+ date: '2026-10-04T20:42:52.472Z',
323
+ operation: 'live',
324
+ replayedAt: null,
325
+ scoringRevision: 2,
326
+ split: 'development',
327
+ cases: 50,
328
+ ready: 44,
329
+ reviewRate: 0.12,
330
+ readyAccuracy: 1,
331
+ failed: 0,
332
+ provenance: [
333
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 50 },
334
+ ],
335
+ acceptanceMet: null,
336
+ },
305
337
  },
306
338
  },
307
339
  {
@@ -564,29 +596,29 @@ export const recipeMetadata = [
564
596
  },
565
597
  ],
566
598
  evidence: {
567
- kind: 'earlier',
568
- label: 'Earlier-evaluator measurement',
599
+ kind: 'synthetic',
600
+ label: 'Current synthetic measurement',
569
601
  experimental: true,
570
602
  measurement: {
571
603
  model: 'jev-1.13.0',
572
- date: '2026-09-27T07:56:26.559Z',
604
+ date: '2026-10-04T20:42:53.964Z',
573
605
  operation: 'live',
574
606
  replayedAt: null,
575
- scoringRevision: 1,
576
- split: 'held-out',
577
- cases: 25,
578
- ready: 24,
579
- reviewRate: 0.04,
580
- readyAccuracy: 0.958,
607
+ scoringRevision: 2,
608
+ split: 'development',
609
+ cases: 75,
610
+ ready: 74,
611
+ reviewRate: 0.013,
612
+ readyAccuracy: 1,
581
613
  failed: 0,
582
614
  provenance: [
583
615
  {
584
616
  method: 'author-synthetic',
585
617
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
586
- cases: 25,
618
+ cases: 75,
587
619
  },
588
620
  ],
589
- acceptanceMet: true,
621
+ acceptanceMet: null,
590
622
  },
591
623
  },
592
624
  },
@@ -632,29 +664,29 @@ export const recipeMetadata = [
632
664
  { id: 'followup-timing', reason: 'Use followup-timing for when another contact is wanted.' },
633
665
  ],
634
666
  evidence: {
635
- kind: 'earlier',
636
- label: 'Earlier-evaluator measurement',
667
+ kind: 'synthetic',
668
+ label: 'Current synthetic measurement',
637
669
  experimental: true,
638
670
  measurement: {
639
671
  model: 'jev-1.13.0',
640
- date: '2026-09-27T07:56:29.897Z',
672
+ date: '2026-10-04T20:42:56.097Z',
641
673
  operation: 'live',
642
674
  replayedAt: null,
643
- scoringRevision: 1,
644
- split: 'held-out',
645
- cases: 40,
646
- ready: 24,
647
- reviewRate: 0.4,
648
- readyAccuracy: 1,
675
+ scoringRevision: 2,
676
+ split: 'development',
677
+ cases: 60,
678
+ ready: 38,
679
+ reviewRate: 0.367,
680
+ readyAccuracy: 0.974,
649
681
  failed: 0,
650
682
  provenance: [
651
683
  {
652
684
  method: 'author-synthetic',
653
685
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
654
- cases: 40,
686
+ cases: 60,
655
687
  },
656
688
  ],
657
- acceptanceMet: true,
689
+ acceptanceMet: null,
658
690
  },
659
691
  },
660
692
  },
@@ -893,10 +925,26 @@ export const recipeMetadata = [
893
925
  },
894
926
  ],
895
927
  evidence: {
896
- kind: 'unknown',
897
- label: 'Unknown response origin',
928
+ kind: 'unspecified',
929
+ label: 'Current measurement; labels unspecified',
898
930
  experimental: true,
899
- measurement: null,
931
+ measurement: {
932
+ model: 'jev-1.13.0',
933
+ date: '2026-10-04T20:42:57.724Z',
934
+ operation: 'live',
935
+ replayedAt: null,
936
+ scoringRevision: 2,
937
+ split: 'development',
938
+ cases: 31,
939
+ ready: 0,
940
+ reviewRate: 1,
941
+ readyAccuracy: null,
942
+ failed: 0,
943
+ provenance: [
944
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 31 },
945
+ ],
946
+ acceptanceMet: null,
947
+ },
900
948
  },
901
949
  },
902
950
  {
@@ -939,10 +987,26 @@ export const recipeMetadata = [
939
987
  },
940
988
  ],
941
989
  evidence: {
942
- kind: 'unknown',
943
- label: 'Unknown response origin',
990
+ kind: 'unspecified',
991
+ label: 'Current measurement; labels unspecified',
944
992
  experimental: true,
945
- measurement: null,
993
+ measurement: {
994
+ model: 'jev-1.13.0',
995
+ date: '2026-10-04T20:42:58.654Z',
996
+ operation: 'live',
997
+ replayedAt: null,
998
+ scoringRevision: 2,
999
+ split: 'development',
1000
+ cases: 32,
1001
+ ready: 27,
1002
+ reviewRate: 0.156,
1003
+ readyAccuracy: 1,
1004
+ failed: 0,
1005
+ provenance: [
1006
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 32 },
1007
+ ],
1008
+ acceptanceMet: null,
1009
+ },
946
1010
  },
947
1011
  },
948
1012
  {
@@ -1046,10 +1110,26 @@ export const recipeMetadata = [
1046
1110
  },
1047
1111
  ],
1048
1112
  evidence: {
1049
- kind: 'unknown',
1050
- label: 'Unknown response origin',
1113
+ kind: 'unspecified',
1114
+ label: 'Current measurement; labels unspecified',
1051
1115
  experimental: true,
1052
- measurement: null,
1116
+ measurement: {
1117
+ model: 'jev-1.13.0',
1118
+ date: '2026-10-04T20:42:59.711Z',
1119
+ operation: 'live',
1120
+ replayedAt: null,
1121
+ scoringRevision: 2,
1122
+ split: 'development',
1123
+ cases: 53,
1124
+ ready: 39,
1125
+ reviewRate: 0.264,
1126
+ readyAccuracy: 1,
1127
+ failed: 0,
1128
+ provenance: [
1129
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 53 },
1130
+ ],
1131
+ acceptanceMet: null,
1132
+ },
1053
1133
  },
1054
1134
  },
1055
1135
  {
@@ -1070,10 +1150,26 @@ export const recipeMetadata = [
1070
1150
  },
1071
1151
  ],
1072
1152
  evidence: {
1073
- kind: 'unknown',
1074
- label: 'Unknown response origin',
1153
+ kind: 'unspecified',
1154
+ label: 'Current measurement; labels unspecified',
1075
1155
  experimental: true,
1076
- measurement: null,
1156
+ measurement: {
1157
+ model: 'jev-1.13.0',
1158
+ date: '2026-10-04T20:43:01.233Z',
1159
+ operation: 'live',
1160
+ replayedAt: null,
1161
+ scoringRevision: 2,
1162
+ split: 'development',
1163
+ cases: 50,
1164
+ ready: 30,
1165
+ reviewRate: 0.4,
1166
+ readyAccuracy: 1,
1167
+ failed: 0,
1168
+ provenance: [
1169
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 50 },
1170
+ ],
1171
+ acceptanceMet: null,
1172
+ },
1077
1173
  },
1078
1174
  },
1079
1175
  {
@@ -1241,29 +1337,34 @@ export const recipeMetadata = [
1241
1337
  },
1242
1338
  ],
1243
1339
  evidence: {
1244
- kind: 'earlier',
1245
- label: 'Earlier-evaluator measurement',
1340
+ kind: 'mixed',
1341
+ label: 'Current mixed-source measurement',
1246
1342
  experimental: true,
1247
1343
  measurement: {
1248
- model: 'typesafe-ai/jev',
1249
- date: '2026-09-27T21:20:01.789Z',
1344
+ model: 'jev-1.13.0',
1345
+ date: '2026-10-04T20:43:02.671Z',
1250
1346
  operation: 'live',
1251
1347
  replayedAt: null,
1252
- scoringRevision: 1,
1253
- split: 'held-out',
1254
- cases: 40,
1255
- ready: 30,
1256
- reviewRate: 0.25,
1348
+ scoringRevision: 2,
1349
+ split: 'development',
1350
+ cases: 144,
1351
+ ready: 91,
1352
+ reviewRate: 0.368,
1257
1353
  readyAccuracy: 1,
1258
1354
  failed: 0,
1259
1355
  provenance: [
1356
+ {
1357
+ method: 'unspecified',
1358
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
1359
+ cases: 44,
1360
+ },
1260
1361
  {
1261
1362
  method: 'author-synthetic',
1262
- source: 'AI-authored new scenario families for the September 2026 update; not independently human-reviewed. Previous inspected holdouts moved to development.',
1263
- cases: 40,
1363
+ source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
1364
+ cases: 100,
1264
1365
  },
1265
1366
  ],
1266
- acceptanceMet: true,
1367
+ acceptanceMet: null,
1267
1368
  },
1268
1369
  },
1269
1370
  },
@@ -1395,29 +1496,29 @@ export const recipeMetadata = [
1395
1496
  },
1396
1497
  ],
1397
1498
  evidence: {
1398
- kind: 'earlier',
1399
- label: 'Earlier-evaluator measurement',
1499
+ kind: 'synthetic',
1500
+ label: 'Current synthetic measurement',
1400
1501
  experimental: true,
1401
1502
  measurement: {
1402
1503
  model: 'jev-1.13.0',
1403
- date: '2026-09-27T07:56:27.401Z',
1504
+ date: '2026-10-04T20:43:06.812Z',
1404
1505
  operation: 'live',
1405
1506
  replayedAt: null,
1406
- scoringRevision: 1,
1407
- split: 'held-out',
1408
- cases: 40,
1409
- ready: 33,
1410
- reviewRate: 0.175,
1507
+ scoringRevision: 2,
1508
+ split: 'development',
1509
+ cases: 60,
1510
+ ready: 50,
1511
+ reviewRate: 0.167,
1411
1512
  readyAccuracy: 1,
1412
1513
  failed: 0,
1413
1514
  provenance: [
1414
1515
  {
1415
1516
  method: 'author-synthetic',
1416
1517
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
1417
- cases: 40,
1518
+ cases: 60,
1418
1519
  },
1419
1520
  ],
1420
- acceptanceMet: true,
1521
+ acceptanceMet: null,
1421
1522
  },
1422
1523
  },
1423
1524
  },
@@ -1494,29 +1595,34 @@ export const recipeMetadata = [
1494
1595
  },
1495
1596
  ],
1496
1597
  evidence: {
1497
- kind: 'earlier',
1498
- label: 'Earlier-evaluator measurement',
1598
+ kind: 'mixed',
1599
+ label: 'Current mixed-source measurement',
1499
1600
  experimental: true,
1500
1601
  measurement: {
1501
- model: 'typesafe-ai/jev',
1502
- date: '2026-09-27T21:21:18.930Z',
1602
+ model: 'jev-1.13.0',
1603
+ date: '2026-10-04T20:43:08.518Z',
1503
1604
  operation: 'live',
1504
1605
  replayedAt: null,
1505
- scoringRevision: 1,
1506
- split: 'held-out',
1507
- cases: 20,
1508
- ready: 11,
1509
- reviewRate: 0.45,
1606
+ scoringRevision: 2,
1607
+ split: 'development',
1608
+ cases: 136,
1609
+ ready: 81,
1610
+ reviewRate: 0.404,
1510
1611
  readyAccuracy: 1,
1511
1612
  failed: 0,
1512
1613
  provenance: [
1614
+ {
1615
+ method: 'unspecified',
1616
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
1617
+ cases: 36,
1618
+ },
1513
1619
  {
1514
1620
  method: 'author-synthetic',
1515
- source: 'AI-authored new scenario families for the September 2026 update; not independently human-reviewed. Previous inspected holdouts moved to development.',
1516
- cases: 20,
1621
+ source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
1622
+ cases: 100,
1517
1623
  },
1518
1624
  ],
1519
- acceptanceMet: false,
1625
+ acceptanceMet: null,
1520
1626
  },
1521
1627
  },
1522
1628
  },
@@ -1730,10 +1836,26 @@ export const recipeMetadata = [
1730
1836
  },
1731
1837
  ],
1732
1838
  evidence: {
1733
- kind: 'unknown',
1734
- label: 'Unknown response origin',
1839
+ kind: 'unspecified',
1840
+ label: 'Current measurement; labels unspecified',
1735
1841
  experimental: true,
1736
- measurement: null,
1842
+ measurement: {
1843
+ model: 'jev-1.13.0',
1844
+ date: '2026-10-04T20:43:12.519Z',
1845
+ operation: 'live',
1846
+ replayedAt: null,
1847
+ scoringRevision: 2,
1848
+ split: 'development',
1849
+ cases: 43,
1850
+ ready: 35,
1851
+ reviewRate: 0.186,
1852
+ readyAccuracy: 0.971,
1853
+ failed: 0,
1854
+ provenance: [
1855
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 43 },
1856
+ ],
1857
+ acceptanceMet: null,
1858
+ },
1737
1859
  },
1738
1860
  },
1739
1861
  {
@@ -1913,18 +2035,18 @@ export const recipeMetadata = [
1913
2035
  },
1914
2036
  ],
1915
2037
  evidence: {
1916
- kind: 'earlier',
1917
- label: 'Earlier-evaluator measurement',
2038
+ kind: 'synthetic',
2039
+ label: 'Current synthetic measurement',
1918
2040
  experimental: true,
1919
2041
  measurement: {
1920
2042
  model: 'jev-1.13.0',
1921
- date: '2026-09-27T07:56:33.689Z',
2043
+ date: '2026-10-04T20:43:14.302Z',
1922
2044
  operation: 'live',
1923
2045
  replayedAt: null,
1924
- scoringRevision: 1,
1925
- split: 'held-out',
1926
- cases: 40,
1927
- ready: 30,
2046
+ scoringRevision: 2,
2047
+ split: 'development',
2048
+ cases: 60,
2049
+ ready: 45,
1928
2050
  reviewRate: 0.25,
1929
2051
  readyAccuracy: 1,
1930
2052
  failed: 0,
@@ -1932,10 +2054,10 @@ export const recipeMetadata = [
1932
2054
  {
1933
2055
  method: 'author-synthetic',
1934
2056
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
1935
- cases: 40,
2057
+ cases: 60,
1936
2058
  },
1937
2059
  ],
1938
- acceptanceMet: true,
2060
+ acceptanceMet: null,
1939
2061
  },
1940
2062
  },
1941
2063
  },
@@ -2212,18 +2334,18 @@ export const recipeMetadata = [
2212
2334
  },
2213
2335
  ],
2214
2336
  evidence: {
2215
- kind: 'earlier',
2216
- label: 'Earlier-evaluator measurement',
2337
+ kind: 'synthetic',
2338
+ label: 'Current synthetic measurement',
2217
2339
  experimental: true,
2218
2340
  measurement: {
2219
2341
  model: 'jev-1.13.0',
2220
- date: '2026-09-27T07:56:32.456Z',
2342
+ date: '2026-10-04T20:43:16.085Z',
2221
2343
  operation: 'live',
2222
2344
  replayedAt: null,
2223
- scoringRevision: 1,
2224
- split: 'held-out',
2225
- cases: 40,
2226
- ready: 40,
2345
+ scoringRevision: 2,
2346
+ split: 'development',
2347
+ cases: 60,
2348
+ ready: 60,
2227
2349
  reviewRate: 0,
2228
2350
  readyAccuracy: 1,
2229
2351
  failed: 0,
@@ -2231,10 +2353,10 @@ export const recipeMetadata = [
2231
2353
  {
2232
2354
  method: 'author-synthetic',
2233
2355
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
2234
- cases: 40,
2356
+ cases: 60,
2235
2357
  },
2236
2358
  ],
2237
- acceptanceMet: true,
2359
+ acceptanceMet: null,
2238
2360
  },
2239
2361
  },
2240
2362
  },
@@ -2359,29 +2481,29 @@ export const recipeMetadata = [
2359
2481
  },
2360
2482
  ],
2361
2483
  evidence: {
2362
- kind: 'earlier',
2363
- label: 'Earlier-evaluator measurement',
2484
+ kind: 'synthetic',
2485
+ label: 'Current synthetic measurement',
2364
2486
  experimental: true,
2365
2487
  measurement: {
2366
2488
  model: 'jev-1.13.0',
2367
- date: '2026-09-27T07:56:31.169Z',
2489
+ date: '2026-10-04T20:43:17.929Z',
2368
2490
  operation: 'live',
2369
2491
  replayedAt: null,
2370
- scoringRevision: 1,
2371
- split: 'held-out',
2372
- cases: 40,
2373
- ready: 30,
2374
- reviewRate: 0.25,
2492
+ scoringRevision: 2,
2493
+ split: 'development',
2494
+ cases: 60,
2495
+ ready: 44,
2496
+ reviewRate: 0.267,
2375
2497
  readyAccuracy: 1,
2376
2498
  failed: 0,
2377
2499
  provenance: [
2378
2500
  {
2379
2501
  method: 'author-synthetic',
2380
2502
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
2381
- cases: 40,
2503
+ cases: 60,
2382
2504
  },
2383
2505
  ],
2384
- acceptanceMet: true,
2506
+ acceptanceMet: null,
2385
2507
  },
2386
2508
  },
2387
2509
  },
@@ -2451,18 +2573,18 @@ export const recipeMetadata = [
2451
2573
  },
2452
2574
  ],
2453
2575
  evidence: {
2454
- kind: 'earlier',
2455
- label: 'Earlier-evaluator measurement',
2576
+ kind: 'synthetic',
2577
+ label: 'Current synthetic measurement',
2456
2578
  experimental: true,
2457
2579
  measurement: {
2458
2580
  model: 'jev-1.13.0',
2459
- date: '2026-09-27T07:56:24.179Z',
2581
+ date: '2026-10-04T20:43:19.595Z',
2460
2582
  operation: 'live',
2461
2583
  replayedAt: null,
2462
- scoringRevision: 1,
2463
- split: 'held-out',
2464
- cases: 40,
2465
- ready: 30,
2584
+ scoringRevision: 2,
2585
+ split: 'development',
2586
+ cases: 60,
2587
+ ready: 45,
2466
2588
  reviewRate: 0.25,
2467
2589
  readyAccuracy: 1,
2468
2590
  failed: 0,
@@ -2470,10 +2592,10 @@ export const recipeMetadata = [
2470
2592
  {
2471
2593
  method: 'author-synthetic',
2472
2594
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
2473
- cases: 40,
2595
+ cases: 60,
2474
2596
  },
2475
2597
  ],
2476
- acceptanceMet: true,
2598
+ acceptanceMet: null,
2477
2599
  },
2478
2600
  },
2479
2601
  },
@@ -2504,29 +2626,29 @@ export const recipeMetadata = [
2504
2626
  },
2505
2627
  ],
2506
2628
  evidence: {
2507
- kind: 'earlier',
2508
- label: 'Earlier-evaluator measurement',
2629
+ kind: 'synthetic',
2630
+ label: 'Current synthetic measurement',
2509
2631
  experimental: true,
2510
2632
  measurement: {
2511
2633
  model: 'jev-1.13.0',
2512
- date: '2026-09-27T07:56:28.731Z',
2634
+ date: '2026-10-04T20:43:21.226Z',
2513
2635
  operation: 'live',
2514
2636
  replayedAt: null,
2515
- scoringRevision: 1,
2516
- split: 'held-out',
2517
- cases: 40,
2518
- ready: 37,
2519
- reviewRate: 0.075,
2637
+ scoringRevision: 2,
2638
+ split: 'development',
2639
+ cases: 60,
2640
+ ready: 51,
2641
+ reviewRate: 0.15,
2520
2642
  readyAccuracy: 1,
2521
2643
  failed: 0,
2522
2644
  provenance: [
2523
2645
  {
2524
2646
  method: 'author-synthetic',
2525
2647
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
2526
- cases: 40,
2648
+ cases: 60,
2527
2649
  },
2528
2650
  ],
2529
- acceptanceMet: true,
2651
+ acceptanceMet: null,
2530
2652
  },
2531
2653
  },
2532
2654
  },
@@ -2706,10 +2828,26 @@ export const recipeMetadata = [
2706
2828
  },
2707
2829
  ],
2708
2830
  evidence: {
2709
- kind: 'unknown',
2710
- label: 'Unknown response origin',
2831
+ kind: 'unspecified',
2832
+ label: 'Current measurement; labels unspecified',
2711
2833
  experimental: true,
2712
- measurement: null,
2834
+ measurement: {
2835
+ model: 'jev-1.13.0',
2836
+ date: '2026-10-04T20:43:22.858Z',
2837
+ operation: 'live',
2838
+ replayedAt: null,
2839
+ scoringRevision: 2,
2840
+ split: 'development',
2841
+ cases: 44,
2842
+ ready: 0,
2843
+ reviewRate: 1,
2844
+ readyAccuracy: null,
2845
+ failed: 0,
2846
+ provenance: [
2847
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 44 },
2848
+ ],
2849
+ acceptanceMet: null,
2850
+ },
2713
2851
  },
2714
2852
  },
2715
2853
  {
@@ -2773,29 +2911,29 @@ export const recipeMetadata = [
2773
2911
  { id: 'route', reason: 'Use route to choose a handler when escalation is not the decision.' },
2774
2912
  ],
2775
2913
  evidence: {
2776
- kind: 'earlier',
2777
- label: 'Earlier-evaluator measurement',
2914
+ kind: 'synthetic',
2915
+ label: 'Current synthetic measurement',
2778
2916
  experimental: true,
2779
2917
  measurement: {
2780
2918
  model: 'jev-1.13.0',
2781
- date: '2026-09-27T07:56:21.635Z',
2919
+ date: '2026-10-04T20:43:24.162Z',
2782
2920
  operation: 'live',
2783
2921
  replayedAt: null,
2784
- scoringRevision: 1,
2785
- split: 'held-out',
2786
- cases: 40,
2787
- ready: 28,
2788
- reviewRate: 0.3,
2789
- readyAccuracy: 1,
2922
+ scoringRevision: 2,
2923
+ split: 'development',
2924
+ cases: 60,
2925
+ ready: 47,
2926
+ reviewRate: 0.217,
2927
+ readyAccuracy: 0.957,
2790
2928
  failed: 0,
2791
2929
  provenance: [
2792
2930
  {
2793
2931
  method: 'author-synthetic',
2794
2932
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
2795
- cases: 40,
2933
+ cases: 60,
2796
2934
  },
2797
2935
  ],
2798
- acceptanceMet: true,
2936
+ acceptanceMet: null,
2799
2937
  },
2800
2938
  },
2801
2939
  },
@@ -2906,10 +3044,26 @@ export const recipeMetadata = [
2906
3044
  },
2907
3045
  ],
2908
3046
  evidence: {
2909
- kind: 'unknown',
2910
- label: 'Unknown response origin',
3047
+ kind: 'unspecified',
3048
+ label: 'Current measurement; labels unspecified',
2911
3049
  experimental: true,
2912
- measurement: null,
3050
+ measurement: {
3051
+ model: 'jev-1.13.0',
3052
+ date: '2026-10-04T20:43:25.936Z',
3053
+ operation: 'live',
3054
+ replayedAt: null,
3055
+ scoringRevision: 2,
3056
+ split: 'development',
3057
+ cases: 44,
3058
+ ready: 42,
3059
+ reviewRate: 0.045,
3060
+ readyAccuracy: 1,
3061
+ failed: 0,
3062
+ provenance: [
3063
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 44 },
3064
+ ],
3065
+ acceptanceMet: null,
3066
+ },
2913
3067
  },
2914
3068
  },
2915
3069
  {
@@ -3623,29 +3777,34 @@ export const recipeMetadata = [
3623
3777
  },
3624
3778
  ],
3625
3779
  evidence: {
3626
- kind: 'earlier',
3627
- label: 'Earlier-evaluator measurement',
3780
+ kind: 'mixed',
3781
+ label: 'Current mixed-source measurement',
3628
3782
  experimental: true,
3629
3783
  measurement: {
3630
3784
  model: 'jev-1.13.0',
3631
- date: '2026-09-27T07:56:10.135Z',
3785
+ date: '2026-10-04T20:43:27.171Z',
3632
3786
  operation: 'live',
3633
3787
  replayedAt: null,
3634
- scoringRevision: 1,
3635
- split: 'held-out',
3636
- cases: 40,
3637
- ready: 26,
3638
- reviewRate: 0.35,
3639
- readyAccuracy: 1,
3788
+ scoringRevision: 2,
3789
+ split: 'development',
3790
+ cases: 104,
3791
+ ready: 64,
3792
+ reviewRate: 0.385,
3793
+ readyAccuracy: 0.984,
3640
3794
  failed: 0,
3641
3795
  provenance: [
3796
+ {
3797
+ method: 'unspecified',
3798
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
3799
+ cases: 44,
3800
+ },
3642
3801
  {
3643
3802
  method: 'author-synthetic',
3644
3803
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
3645
- cases: 40,
3804
+ cases: 60,
3646
3805
  },
3647
3806
  ],
3648
- acceptanceMet: true,
3807
+ acceptanceMet: null,
3649
3808
  },
3650
3809
  },
3651
3810
  },
@@ -4058,129 +4217,94 @@ export const recipeMetadata = [
4058
4217
  },
4059
4218
  ],
4060
4219
  evidence: {
4061
- kind: 'earlier',
4062
- label: 'Earlier-evaluator measurement',
4220
+ kind: 'public-dataset',
4221
+ label: 'Current public-dataset measurement',
4063
4222
  experimental: true,
4064
4223
  measurement: {
4065
- model: 'typesafe-ai/jev',
4066
- date: '2026-09-27T21:21:33.447Z',
4224
+ model: 'jev-1.13.0',
4225
+ date: '2026-10-04T20:43:30.092Z',
4067
4226
  operation: 'live',
4068
4227
  replayedAt: null,
4069
- scoringRevision: 1,
4070
- split: 'held-out',
4071
- cases: 167,
4072
- ready: 82,
4228
+ scoringRevision: 2,
4229
+ split: 'development',
4230
+ cases: 112,
4231
+ ready: 55,
4073
4232
  reviewRate: 0.509,
4074
- readyAccuracy: 0.988,
4233
+ readyAccuracy: 1,
4075
4234
  failed: 0,
4076
4235
  provenance: [
4077
4236
  {
4078
4237
  method: 'public-dataset',
4079
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/async_context.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4238
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/assert.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4080
4239
  cases: 8,
4081
4240
  },
4082
4241
  {
4083
4242
  method: 'public-dataset',
4084
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/buffer.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4243
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/child_process.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4085
4244
  cases: 8,
4086
4245
  },
4087
4246
  {
4088
4247
  method: 'public-dataset',
4089
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/cli.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4248
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/crypto.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4090
4249
  cases: 8,
4091
4250
  },
4092
4251
  {
4093
4252
  method: 'public-dataset',
4094
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/console.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4253
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/diagnostics_channel.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4095
4254
  cases: 8,
4096
4255
  },
4097
4256
  {
4098
4257
  method: 'public-dataset',
4099
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/events.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4258
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/dns.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4100
4259
  cases: 8,
4101
4260
  },
4102
4261
  {
4103
4262
  method: 'public-dataset',
4104
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/fs.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4263
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/errors.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4105
4264
  cases: 8,
4106
4265
  },
4107
4266
  {
4108
4267
  method: 'public-dataset',
4109
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/http.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4268
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/process.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4110
4269
  cases: 8,
4111
4270
  },
4112
4271
  {
4113
4272
  method: 'public-dataset',
4114
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/https.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4273
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/repl.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4115
4274
  cases: 8,
4116
4275
  },
4117
4276
  {
4118
4277
  method: 'public-dataset',
4119
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/net.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4278
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/timers.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4120
4279
  cases: 8,
4121
4280
  },
4122
4281
  {
4123
4282
  method: 'public-dataset',
4124
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/path.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4283
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/tls.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4125
4284
  cases: 8,
4126
4285
  },
4127
4286
  {
4128
4287
  method: 'public-dataset',
4129
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/perf_hooks.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4288
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/util.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4130
4289
  cases: 8,
4131
4290
  },
4132
4291
  {
4133
4292
  method: 'public-dataset',
4134
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/querystring.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4293
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/vm.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4135
4294
  cases: 8,
4136
4295
  },
4137
4296
  {
4138
4297
  method: 'public-dataset',
4139
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/readline.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4298
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/wasi.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4140
4299
  cases: 8,
4141
4300
  },
4142
4301
  {
4143
4302
  method: 'public-dataset',
4144
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/stream.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4145
- cases: 8,
4146
- },
4147
- {
4148
- method: 'public-dataset',
4149
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/string_decoder.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4150
- cases: 7,
4151
- },
4152
- {
4153
- method: 'public-dataset',
4154
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/test.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4155
- cases: 8,
4156
- },
4157
- {
4158
- method: 'public-dataset',
4159
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/tty.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4160
- cases: 8,
4161
- },
4162
- {
4163
- method: 'public-dataset',
4164
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/url.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4165
- cases: 8,
4166
- },
4167
- {
4168
- method: 'public-dataset',
4169
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/v8.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4170
- cases: 8,
4171
- },
4172
- {
4173
- method: 'public-dataset',
4174
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/worker_threads.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4175
- cases: 8,
4176
- },
4177
- {
4178
- method: 'public-dataset',
4179
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/zlib.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4303
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/webcrypto.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
4180
4304
  cases: 8,
4181
4305
  },
4182
4306
  ],
4183
- acceptanceMet: false,
4307
+ acceptanceMet: null,
4184
4308
  },
4185
4309
  },
4186
4310
  },
@@ -4389,10 +4513,26 @@ export const recipeMetadata = [
4389
4513
  },
4390
4514
  ],
4391
4515
  evidence: {
4392
- kind: 'unknown',
4393
- label: 'Unknown response origin',
4516
+ kind: 'unspecified',
4517
+ label: 'Current measurement; labels unspecified',
4394
4518
  experimental: true,
4395
- measurement: null,
4519
+ measurement: {
4520
+ model: 'jev-1.13.0',
4521
+ date: '2026-10-04T20:43:33.245Z',
4522
+ operation: 'live',
4523
+ replayedAt: null,
4524
+ scoringRevision: 2,
4525
+ split: 'development',
4526
+ cases: 43,
4527
+ ready: 36,
4528
+ reviewRate: 0.163,
4529
+ readyAccuracy: 1,
4530
+ failed: 0,
4531
+ provenance: [
4532
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 43 },
4533
+ ],
4534
+ acceptanceMet: null,
4535
+ },
4396
4536
  },
4397
4537
  },
4398
4538
  {
@@ -4625,29 +4765,34 @@ export const recipeMetadata = [
4625
4765
  },
4626
4766
  ],
4627
4767
  evidence: {
4628
- kind: 'earlier',
4629
- label: 'Earlier-evaluator measurement',
4768
+ kind: 'mixed',
4769
+ label: 'Current mixed-source measurement',
4630
4770
  experimental: true,
4631
4771
  measurement: {
4632
4772
  model: 'jev-1.13.0',
4633
- date: '2026-09-27T07:56:14.022Z',
4773
+ date: '2026-10-04T20:43:34.650Z',
4634
4774
  operation: 'live',
4635
4775
  replayedAt: null,
4636
- scoringRevision: 1,
4637
- split: 'held-out',
4638
- cases: 40,
4639
- ready: 40,
4776
+ scoringRevision: 2,
4777
+ split: 'development',
4778
+ cases: 102,
4779
+ ready: 102,
4640
4780
  reviewRate: 0,
4641
4781
  readyAccuracy: 1,
4642
4782
  failed: 0,
4643
4783
  provenance: [
4784
+ {
4785
+ method: 'unspecified',
4786
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
4787
+ cases: 42,
4788
+ },
4644
4789
  {
4645
4790
  method: 'author-synthetic',
4646
4791
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
4647
- cases: 40,
4792
+ cases: 60,
4648
4793
  },
4649
4794
  ],
4650
- acceptanceMet: true,
4795
+ acceptanceMet: null,
4651
4796
  },
4652
4797
  },
4653
4798
  },
@@ -4668,29 +4813,34 @@ export const recipeMetadata = [
4668
4813
  },
4669
4814
  ],
4670
4815
  evidence: {
4671
- kind: 'earlier',
4672
- label: 'Earlier-evaluator measurement',
4816
+ kind: 'mixed',
4817
+ label: 'Current mixed-source measurement',
4673
4818
  experimental: true,
4674
4819
  measurement: {
4675
4820
  model: 'jev-1.13.0',
4676
- date: '2026-09-27T07:56:25.450Z',
4821
+ date: '2026-10-04T20:43:37.293Z',
4677
4822
  operation: 'live',
4678
4823
  replayedAt: null,
4679
- scoringRevision: 1,
4680
- split: 'held-out',
4681
- cases: 40,
4682
- ready: 40,
4683
- reviewRate: 0,
4824
+ scoringRevision: 2,
4825
+ split: 'development',
4826
+ cases: 102,
4827
+ ready: 93,
4828
+ reviewRate: 0.088,
4684
4829
  readyAccuracy: 1,
4685
4830
  failed: 0,
4686
4831
  provenance: [
4832
+ {
4833
+ method: 'unspecified',
4834
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
4835
+ cases: 42,
4836
+ },
4687
4837
  {
4688
4838
  method: 'author-synthetic',
4689
4839
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
4690
- cases: 40,
4840
+ cases: 60,
4691
4841
  },
4692
4842
  ],
4693
- acceptanceMet: true,
4843
+ acceptanceMet: null,
4694
4844
  },
4695
4845
  },
4696
4846
  },
@@ -5047,7 +5197,7 @@ export const recipeMetadata = [
5047
5197
  'Does not retrieve documents or establish source truth.',
5048
5198
  'Provide 1 to 100 items with unique IDs.',
5049
5199
  ],
5050
- useWhen: 'You have retrieved passages and want the most relevant evidence for a question.',
5200
+ useWhen: 'You have retrieved passages and want to find the most useful evidence for a question.',
5051
5201
  related: [
5052
5202
  {
5053
5203
  id: 'answerability',
@@ -5055,10 +5205,26 @@ export const recipeMetadata = [
5055
5205
  },
5056
5206
  ],
5057
5207
  evidence: {
5058
- kind: 'unknown',
5059
- label: 'Unknown response origin',
5208
+ kind: 'unspecified',
5209
+ label: 'Current measurement; labels unspecified',
5060
5210
  experimental: true,
5061
- measurement: null,
5211
+ measurement: {
5212
+ model: 'jev-1.13.0',
5213
+ date: '2026-10-04T20:43:40.159Z',
5214
+ operation: 'live',
5215
+ replayedAt: null,
5216
+ scoringRevision: 2,
5217
+ split: 'development',
5218
+ cases: 57,
5219
+ ready: 57,
5220
+ reviewRate: 0,
5221
+ readyAccuracy: 1,
5222
+ failed: 0,
5223
+ provenance: [
5224
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 57 },
5225
+ ],
5226
+ acceptanceMet: null,
5227
+ },
5062
5228
  },
5063
5229
  },
5064
5230
  {
@@ -5100,18 +5266,18 @@ export const recipeMetadata = [
5100
5266
  },
5101
5267
  ],
5102
5268
  evidence: {
5103
- kind: 'earlier',
5104
- label: 'Earlier-evaluator measurement',
5269
+ kind: 'synthetic',
5270
+ label: 'Current synthetic measurement',
5105
5271
  experimental: true,
5106
5272
  measurement: {
5107
5273
  model: 'jev-1.13.0',
5108
- date: '2026-09-27T07:56:22.956Z',
5274
+ date: '2026-10-04T20:43:41.765Z',
5109
5275
  operation: 'live',
5110
5276
  replayedAt: null,
5111
- scoringRevision: 1,
5112
- split: 'held-out',
5113
- cases: 40,
5114
- ready: 40,
5277
+ scoringRevision: 2,
5278
+ split: 'development',
5279
+ cases: 60,
5280
+ ready: 60,
5115
5281
  reviewRate: 0,
5116
5282
  readyAccuracy: 1,
5117
5283
  failed: 0,
@@ -5119,10 +5285,10 @@ export const recipeMetadata = [
5119
5285
  {
5120
5286
  method: 'author-synthetic',
5121
5287
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
5122
- cases: 40,
5288
+ cases: 60,
5123
5289
  },
5124
5290
  ],
5125
- acceptanceMet: true,
5291
+ acceptanceMet: null,
5126
5292
  },
5127
5293
  },
5128
5294
  },
@@ -5161,10 +5327,26 @@ export const recipeMetadata = [
5161
5327
  },
5162
5328
  ],
5163
5329
  evidence: {
5164
- kind: 'unknown',
5165
- label: 'Unknown response origin',
5330
+ kind: 'unspecified',
5331
+ label: 'Current measurement; labels unspecified',
5166
5332
  experimental: true,
5167
- measurement: null,
5333
+ measurement: {
5334
+ model: 'jev-1.13.0',
5335
+ date: '2026-10-04T20:43:43.607Z',
5336
+ operation: 'live',
5337
+ replayedAt: null,
5338
+ scoringRevision: 2,
5339
+ split: 'development',
5340
+ cases: 44,
5341
+ ready: 39,
5342
+ reviewRate: 0.114,
5343
+ readyAccuracy: 1,
5344
+ failed: 0,
5345
+ provenance: [
5346
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 44 },
5347
+ ],
5348
+ acceptanceMet: null,
5349
+ },
5168
5350
  },
5169
5351
  },
5170
5352
  {
@@ -5266,29 +5448,29 @@ export const recipeMetadata = [
5266
5448
  },
5267
5449
  ],
5268
5450
  evidence: {
5269
- kind: 'earlier',
5270
- label: 'Earlier-evaluator measurement',
5451
+ kind: 'synthetic',
5452
+ label: 'Current synthetic measurement',
5271
5453
  experimental: true,
5272
5454
  measurement: {
5273
5455
  model: 'jev-1.13.0',
5274
- date: '2026-09-27T07:56:15.276Z',
5456
+ date: '2026-10-04T20:43:44.923Z',
5275
5457
  operation: 'live',
5276
5458
  replayedAt: null,
5277
- scoringRevision: 1,
5278
- split: 'held-out',
5279
- cases: 40,
5280
- ready: 39,
5281
- reviewRate: 0.025,
5459
+ scoringRevision: 2,
5460
+ split: 'development',
5461
+ cases: 60,
5462
+ ready: 60,
5463
+ reviewRate: 0,
5282
5464
  readyAccuracy: 1,
5283
5465
  failed: 0,
5284
5466
  provenance: [
5285
5467
  {
5286
5468
  method: 'author-synthetic',
5287
5469
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
5288
- cases: 40,
5470
+ cases: 60,
5289
5471
  },
5290
5472
  ],
5291
- acceptanceMet: true,
5473
+ acceptanceMet: null,
5292
5474
  },
5293
5475
  },
5294
5476
  },
@@ -5457,29 +5639,34 @@ export const recipeMetadata = [
5457
5639
  },
5458
5640
  ],
5459
5641
  evidence: {
5460
- kind: 'earlier',
5461
- label: 'Earlier-evaluator measurement',
5642
+ kind: 'mixed',
5643
+ label: 'Current mixed-source measurement',
5462
5644
  experimental: true,
5463
5645
  measurement: {
5464
5646
  model: 'jev-1.13.0',
5465
- date: '2026-09-27T07:56:19.034Z',
5647
+ date: '2026-10-04T20:43:46.575Z',
5466
5648
  operation: 'live',
5467
5649
  replayedAt: null,
5468
- scoringRevision: 1,
5469
- split: 'held-out',
5470
- cases: 40,
5471
- ready: 24,
5472
- reviewRate: 0.4,
5650
+ scoringRevision: 2,
5651
+ split: 'development',
5652
+ cases: 110,
5653
+ ready: 73,
5654
+ reviewRate: 0.336,
5473
5655
  readyAccuracy: 1,
5474
5656
  failed: 0,
5475
5657
  provenance: [
5658
+ {
5659
+ method: 'unspecified',
5660
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
5661
+ cases: 50,
5662
+ },
5476
5663
  {
5477
5664
  method: 'author-synthetic',
5478
5665
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
5479
- cases: 40,
5666
+ cases: 60,
5480
5667
  },
5481
5668
  ],
5482
- acceptanceMet: true,
5669
+ acceptanceMet: null,
5483
5670
  },
5484
5671
  },
5485
5672
  },
@@ -5505,29 +5692,34 @@ export const recipeMetadata = [
5505
5692
  },
5506
5693
  ],
5507
5694
  evidence: {
5508
- kind: 'earlier',
5509
- label: 'Earlier-version measurement',
5695
+ kind: 'mixed',
5696
+ label: 'Current mixed-source measurement',
5510
5697
  experimental: true,
5511
5698
  measurement: {
5512
5699
  model: 'jev-1.13.0',
5513
- date: '2026-09-27T07:56:20.238Z',
5700
+ date: '2026-10-04T20:43:49.817Z',
5514
5701
  operation: 'live',
5515
5702
  replayedAt: null,
5516
- scoringRevision: 1,
5517
- split: 'held-out',
5518
- cases: 40,
5519
- ready: 25,
5520
- reviewRate: 0.375,
5521
- readyAccuracy: 1,
5703
+ scoringRevision: 2,
5704
+ split: 'development',
5705
+ cases: 87,
5706
+ ready: 47,
5707
+ reviewRate: 0.46,
5708
+ readyAccuracy: 0.979,
5522
5709
  failed: 0,
5523
5710
  provenance: [
5711
+ {
5712
+ method: 'unspecified',
5713
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
5714
+ cases: 27,
5715
+ },
5524
5716
  {
5525
5717
  method: 'author-synthetic',
5526
5718
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
5527
- cases: 40,
5719
+ cases: 60,
5528
5720
  },
5529
5721
  ],
5530
- acceptanceMet: true,
5722
+ acceptanceMet: null,
5531
5723
  },
5532
5724
  },
5533
5725
  },
@@ -6132,129 +6324,94 @@ export const recipeMetadata = [
6132
6324
  },
6133
6325
  ],
6134
6326
  evidence: {
6135
- kind: 'earlier',
6136
- label: 'Earlier-evaluator measurement',
6327
+ kind: 'public-dataset',
6328
+ label: 'Current public-dataset measurement',
6137
6329
  experimental: true,
6138
6330
  measurement: {
6139
- model: 'typesafe-ai/jev',
6140
- date: '2026-09-27T21:21:20.879Z',
6331
+ model: 'jev-1.13.0',
6332
+ date: '2026-10-04T20:43:52.490Z',
6141
6333
  operation: 'live',
6142
6334
  replayedAt: null,
6143
- scoringRevision: 1,
6144
- split: 'held-out',
6145
- cases: 189,
6146
- ready: 136,
6147
- reviewRate: 0.28,
6148
- readyAccuracy: 0.904,
6335
+ scoringRevision: 2,
6336
+ split: 'development',
6337
+ cases: 126,
6338
+ ready: 92,
6339
+ reviewRate: 0.27,
6340
+ readyAccuracy: 0.967,
6149
6341
  failed: 0,
6150
6342
  provenance: [
6151
6343
  {
6152
6344
  method: 'public-dataset',
6153
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/async_context.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6345
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/assert.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6154
6346
  cases: 9,
6155
6347
  },
6156
6348
  {
6157
6349
  method: 'public-dataset',
6158
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/buffer.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6350
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/child_process.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6159
6351
  cases: 9,
6160
6352
  },
6161
6353
  {
6162
6354
  method: 'public-dataset',
6163
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/cli.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6355
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/crypto.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6164
6356
  cases: 9,
6165
6357
  },
6166
6358
  {
6167
6359
  method: 'public-dataset',
6168
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/console.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6360
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/diagnostics_channel.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6169
6361
  cases: 9,
6170
6362
  },
6171
6363
  {
6172
6364
  method: 'public-dataset',
6173
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/events.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6365
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/dns.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6174
6366
  cases: 9,
6175
6367
  },
6176
6368
  {
6177
6369
  method: 'public-dataset',
6178
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/fs.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6370
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/errors.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6179
6371
  cases: 9,
6180
6372
  },
6181
6373
  {
6182
6374
  method: 'public-dataset',
6183
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/http.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6375
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/process.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6184
6376
  cases: 9,
6185
6377
  },
6186
6378
  {
6187
6379
  method: 'public-dataset',
6188
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/https.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6380
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/repl.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6189
6381
  cases: 9,
6190
6382
  },
6191
6383
  {
6192
6384
  method: 'public-dataset',
6193
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/net.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6385
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/timers.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6194
6386
  cases: 9,
6195
6387
  },
6196
6388
  {
6197
6389
  method: 'public-dataset',
6198
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/path.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6390
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/tls.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6199
6391
  cases: 9,
6200
6392
  },
6201
6393
  {
6202
6394
  method: 'public-dataset',
6203
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/perf_hooks.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6395
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/util.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6204
6396
  cases: 9,
6205
6397
  },
6206
6398
  {
6207
6399
  method: 'public-dataset',
6208
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/querystring.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6400
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/vm.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6209
6401
  cases: 9,
6210
6402
  },
6211
6403
  {
6212
6404
  method: 'public-dataset',
6213
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/readline.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6405
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/wasi.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6214
6406
  cases: 9,
6215
6407
  },
6216
6408
  {
6217
6409
  method: 'public-dataset',
6218
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/stream.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6219
- cases: 9,
6220
- },
6221
- {
6222
- method: 'public-dataset',
6223
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/string_decoder.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6224
- cases: 9,
6225
- },
6226
- {
6227
- method: 'public-dataset',
6228
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/test.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6229
- cases: 9,
6230
- },
6231
- {
6232
- method: 'public-dataset',
6233
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/tty.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6234
- cases: 9,
6235
- },
6236
- {
6237
- method: 'public-dataset',
6238
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/url.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6239
- cases: 9,
6240
- },
6241
- {
6242
- method: 'public-dataset',
6243
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/v8.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6244
- cases: 9,
6245
- },
6246
- {
6247
- method: 'public-dataset',
6248
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/worker_threads.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6249
- cases: 9,
6250
- },
6251
- {
6252
- method: 'public-dataset',
6253
- source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/zlib.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6410
+ source: 'https://raw.githubusercontent.com/nodejs/node/v22.20.0/doc/api/webcrypto.md; source Markdown labels, transformed by build-ingestion-cases.mjs; no independent human review; Node.js MIT license.',
6254
6411
  cases: 9,
6255
6412
  },
6256
6413
  ],
6257
- acceptanceMet: false,
6414
+ acceptanceMet: null,
6258
6415
  },
6259
6416
  },
6260
6417
  },
@@ -6315,7 +6472,7 @@ export const recipeMetadata = [
6315
6472
  'Deterministic rules such as allowlists, path restrictions, and rate limits are cheaper and more reliable for what they cover; run them first and use this recipe for the cases they leave open.',
6316
6473
  'A review or ask outcome means a person should look; it does not itself block execution. The application must enforce the action.',
6317
6474
  ],
6318
- useWhen: 'A coding or browsing agent is about to execute a tool call and you need an allow, ask, or deny decision plus the specific risks, in one fast request.',
6475
+ useWhen: 'A coding or browsing agent proposes a tool call and you need to review that action: an allow, ask, or deny decision plus the specific risks, in one fast request.',
6319
6476
  related: [
6320
6477
  {
6321
6478
  id: 'action-scope',
@@ -6335,29 +6492,34 @@ export const recipeMetadata = [
6335
6492
  },
6336
6493
  ],
6337
6494
  evidence: {
6338
- kind: 'earlier',
6339
- label: 'Earlier-evaluator measurement',
6495
+ kind: 'mixed',
6496
+ label: 'Current mixed-source measurement',
6340
6497
  experimental: true,
6341
6498
  measurement: {
6342
6499
  model: 'jev-1.13.0',
6343
- date: '2026-09-27T07:56:16.449Z',
6500
+ date: '2026-10-04T20:43:55.923Z',
6344
6501
  operation: 'live',
6345
6502
  replayedAt: null,
6346
- scoringRevision: 1,
6347
- split: 'held-out',
6348
- cases: 40,
6349
- ready: 31,
6350
- reviewRate: 0.225,
6503
+ scoringRevision: 2,
6504
+ split: 'development',
6505
+ cases: 104,
6506
+ ready: 76,
6507
+ reviewRate: 0.269,
6351
6508
  readyAccuracy: 1,
6352
6509
  failed: 0,
6353
6510
  provenance: [
6511
+ {
6512
+ method: 'unspecified',
6513
+ source: 'Existing repository labels; independent review and original authorship were not recorded.',
6514
+ cases: 44,
6515
+ },
6354
6516
  {
6355
6517
  method: 'author-synthetic',
6356
6518
  source: 'AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.',
6357
- cases: 40,
6519
+ cases: 60,
6358
6520
  },
6359
6521
  ],
6360
- acceptanceMet: true,
6522
+ acceptanceMet: null,
6361
6523
  },
6362
6524
  },
6363
6525
  },
@@ -6552,10 +6714,26 @@ export const recipeMetadata = [
6552
6714
  },
6553
6715
  ],
6554
6716
  evidence: {
6555
- kind: 'unknown',
6556
- label: 'Unknown response origin',
6717
+ kind: 'unspecified',
6718
+ label: 'Current measurement; labels unspecified',
6557
6719
  experimental: true,
6558
- measurement: null,
6720
+ measurement: {
6721
+ model: 'jev-1.13.0',
6722
+ date: '2026-10-04T20:43:58.961Z',
6723
+ operation: 'live',
6724
+ replayedAt: null,
6725
+ scoringRevision: 2,
6726
+ split: 'development',
6727
+ cases: 50,
6728
+ ready: 43,
6729
+ reviewRate: 0.14,
6730
+ readyAccuracy: 1,
6731
+ failed: 0,
6732
+ provenance: [
6733
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 50 },
6734
+ ],
6735
+ acceptanceMet: null,
6736
+ },
6559
6737
  },
6560
6738
  },
6561
6739
  {
@@ -6580,10 +6758,26 @@ export const recipeMetadata = [
6580
6758
  },
6581
6759
  ],
6582
6760
  evidence: {
6583
- kind: 'unknown',
6584
- label: 'Unknown response origin',
6761
+ kind: 'unspecified',
6762
+ label: 'Current measurement; labels unspecified',
6585
6763
  experimental: true,
6586
- measurement: null,
6764
+ measurement: {
6765
+ model: 'jev-1.13.0',
6766
+ date: '2026-10-04T20:44:00.345Z',
6767
+ operation: 'live',
6768
+ replayedAt: null,
6769
+ scoringRevision: 2,
6770
+ split: 'development',
6771
+ cases: 44,
6772
+ ready: 33,
6773
+ reviewRate: 0.25,
6774
+ readyAccuracy: 0.97,
6775
+ failed: 0,
6776
+ provenance: [
6777
+ { method: 'unspecified', source: 'Not recorded by the dataset author.', cases: 44 },
6778
+ ],
6779
+ acceptanceMet: null,
6780
+ },
6587
6781
  },
6588
6782
  },
6589
6783
  {