open-research-protocol 0.4.7 → 0.4.8
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/README.md +9 -0
- package/cli/orp.py +668 -43
- package/docs/ORP_REASONING_KERNEL_AGENT_PILOT.md +125 -0
- package/docs/ORP_REASONING_KERNEL_AGENT_REPLICATION.md +97 -0
- package/docs/ORP_REASONING_KERNEL_CANONICAL_CONTINUATION_PILOT.md +100 -0
- package/docs/ORP_REASONING_KERNEL_COMPARISON_PILOT.md +116 -0
- package/docs/ORP_REASONING_KERNEL_CONTINUATION_PILOT.md +86 -0
- package/docs/ORP_REASONING_KERNEL_EVALUATION_PLAN.md +261 -0
- package/docs/ORP_REASONING_KERNEL_EVIDENCE_MATRIX.md +131 -0
- package/docs/ORP_REASONING_KERNEL_EVOLUTION.md +123 -0
- package/docs/ORP_REASONING_KERNEL_PICKUP_PILOT.md +107 -0
- package/docs/ORP_REASONING_KERNEL_TECHNICAL_VALIDATION.md +140 -22
- package/docs/ORP_REASONING_KERNEL_V0_1.md +11 -0
- package/docs/benchmarks/orp_reasoning_kernel_agent_pilot_v0_1.json +796 -0
- package/docs/benchmarks/orp_reasoning_kernel_agent_replication_task_smoke.json +487 -0
- package/docs/benchmarks/orp_reasoning_kernel_agent_replication_v0_1.json +1927 -0
- package/docs/benchmarks/orp_reasoning_kernel_agent_replication_v0_2.json +10217 -0
- package/docs/benchmarks/orp_reasoning_kernel_canonical_continuation_task_smoke.json +174 -0
- package/docs/benchmarks/orp_reasoning_kernel_canonical_continuation_v0_1.json +598 -0
- package/docs/benchmarks/orp_reasoning_kernel_comparison_v0_1.json +688 -0
- package/docs/benchmarks/orp_reasoning_kernel_continuation_task_smoke.json +150 -0
- package/docs/benchmarks/orp_reasoning_kernel_continuation_v0_1.json +448 -0
- package/docs/benchmarks/orp_reasoning_kernel_pickup_v0_1.json +594 -0
- package/docs/benchmarks/orp_reasoning_kernel_v0_1_validation.json +769 -41
- package/examples/README.md +2 -0
- package/examples/kernel/comparison/comparison-corpus.json +337 -0
- package/examples/kernel/comparison/next-task-continuation.json +55 -0
- package/examples/kernel/corpus/operations/habanero-routing.checkpoint.kernel.yml +12 -0
- package/examples/kernel/corpus/operations/runner-routing.policy.kernel.yml +9 -0
- package/examples/kernel/corpus/product/project-home.decision.kernel.yml +11 -0
- package/examples/kernel/corpus/research/kernel-handoff.experiment.kernel.yml +16 -0
- package/examples/kernel/corpus/research/lane-drift.hypothesis.kernel.yml +11 -0
- package/examples/kernel/corpus/software/trace-widget.task.kernel.yml +13 -0
- package/examples/kernel/corpus/writing/kernel-launch.result.kernel.yml +12 -0
- package/package.json +4 -1
- package/scripts/orp-kernel-agent-pilot.py +673 -0
- package/scripts/orp-kernel-agent-replication.py +307 -0
- package/scripts/orp-kernel-benchmark.py +471 -2
- package/scripts/orp-kernel-canonical-continuation.py +381 -0
- package/scripts/orp-kernel-ci-check.py +138 -0
- package/scripts/orp-kernel-comparison.py +592 -0
- package/scripts/orp-kernel-continuation-pilot.py +384 -0
- package/scripts/orp-kernel-pickup.py +401 -0
- package/spec/v1/kernel-extension.schema.json +96 -0
- package/spec/v1/kernel-proposal.schema.json +115 -0
- package/spec/v1/kernel.schema.json +2 -1
|
@@ -0,0 +1,688 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schema_version": "1.0.0",
|
|
3
|
+
"kind": "orp_reasoning_kernel_comparison_report",
|
|
4
|
+
"metadata": {
|
|
5
|
+
"generated_at_utc": "2026-03-23T06:06:17Z",
|
|
6
|
+
"repo_commit": "c2f7f2a52744a00fb719d37de583da1f4ae615bd",
|
|
7
|
+
"repo_branch": "main",
|
|
8
|
+
"package_version": "0.4.7",
|
|
9
|
+
"python_version": "3.9.6",
|
|
10
|
+
"node_version": "v24.10.0",
|
|
11
|
+
"platform": "macOS-26.3-arm64-arm-64bit"
|
|
12
|
+
},
|
|
13
|
+
"corpus": {
|
|
14
|
+
"source": "examples/kernel/comparison/comparison-corpus.json",
|
|
15
|
+
"cases_total": 7,
|
|
16
|
+
"domains_total": 5,
|
|
17
|
+
"domains": [
|
|
18
|
+
"operations",
|
|
19
|
+
"product",
|
|
20
|
+
"research",
|
|
21
|
+
"software",
|
|
22
|
+
"writing"
|
|
23
|
+
],
|
|
24
|
+
"artifact_classes_total": 7,
|
|
25
|
+
"artifact_classes": [
|
|
26
|
+
"checkpoint",
|
|
27
|
+
"decision",
|
|
28
|
+
"experiment",
|
|
29
|
+
"hypothesis",
|
|
30
|
+
"policy",
|
|
31
|
+
"result",
|
|
32
|
+
"task"
|
|
33
|
+
]
|
|
34
|
+
},
|
|
35
|
+
"conditions": {
|
|
36
|
+
"freeform": {
|
|
37
|
+
"condition": "freeform",
|
|
38
|
+
"cases_total": 7,
|
|
39
|
+
"rows": [
|
|
40
|
+
{
|
|
41
|
+
"id": "software_trace_widget",
|
|
42
|
+
"domain": "software",
|
|
43
|
+
"artifact_class": "task",
|
|
44
|
+
"total_score": 0.208,
|
|
45
|
+
"class_specific_completeness": 0.27,
|
|
46
|
+
"ambiguity_remaining": 0.73,
|
|
47
|
+
"present_fields": [
|
|
48
|
+
"constraints",
|
|
49
|
+
"goal",
|
|
50
|
+
"object"
|
|
51
|
+
],
|
|
52
|
+
"missing_fields": [
|
|
53
|
+
"boundary",
|
|
54
|
+
"success_criteria"
|
|
55
|
+
]
|
|
56
|
+
},
|
|
57
|
+
{
|
|
58
|
+
"id": "product_project_home",
|
|
59
|
+
"domain": "product",
|
|
60
|
+
"artifact_class": "decision",
|
|
61
|
+
"total_score": 0.323,
|
|
62
|
+
"class_specific_completeness": 0.36,
|
|
63
|
+
"ambiguity_remaining": 0.64,
|
|
64
|
+
"present_fields": [
|
|
65
|
+
"chosen_path",
|
|
66
|
+
"consequences",
|
|
67
|
+
"question",
|
|
68
|
+
"rationale"
|
|
69
|
+
],
|
|
70
|
+
"missing_fields": [
|
|
71
|
+
"rejected_alternatives"
|
|
72
|
+
]
|
|
73
|
+
},
|
|
74
|
+
{
|
|
75
|
+
"id": "research_drift_hypothesis",
|
|
76
|
+
"domain": "research",
|
|
77
|
+
"artifact_class": "hypothesis",
|
|
78
|
+
"total_score": 0.323,
|
|
79
|
+
"class_specific_completeness": 0.36,
|
|
80
|
+
"ambiguity_remaining": 0.64,
|
|
81
|
+
"present_fields": [
|
|
82
|
+
"assumptions",
|
|
83
|
+
"boundary",
|
|
84
|
+
"claim",
|
|
85
|
+
"test_path"
|
|
86
|
+
],
|
|
87
|
+
"missing_fields": [
|
|
88
|
+
"falsifiers"
|
|
89
|
+
]
|
|
90
|
+
},
|
|
91
|
+
{
|
|
92
|
+
"id": "research_handoff_experiment",
|
|
93
|
+
"domain": "research",
|
|
94
|
+
"artifact_class": "experiment",
|
|
95
|
+
"total_score": 0.2,
|
|
96
|
+
"class_specific_completeness": 0.225,
|
|
97
|
+
"ambiguity_remaining": 0.775,
|
|
98
|
+
"present_fields": [
|
|
99
|
+
"interpretation_limits",
|
|
100
|
+
"method",
|
|
101
|
+
"objective"
|
|
102
|
+
],
|
|
103
|
+
"missing_fields": [
|
|
104
|
+
"inputs",
|
|
105
|
+
"outputs",
|
|
106
|
+
"evidence_expectations"
|
|
107
|
+
]
|
|
108
|
+
},
|
|
109
|
+
{
|
|
110
|
+
"id": "operations_habanero_checkpoint",
|
|
111
|
+
"domain": "operations",
|
|
112
|
+
"artifact_class": "checkpoint",
|
|
113
|
+
"total_score": 0.285,
|
|
114
|
+
"class_specific_completeness": 0.36,
|
|
115
|
+
"ambiguity_remaining": 0.64,
|
|
116
|
+
"present_fields": [
|
|
117
|
+
"completed_unit",
|
|
118
|
+
"current_state",
|
|
119
|
+
"next_handoff_target",
|
|
120
|
+
"risks"
|
|
121
|
+
],
|
|
122
|
+
"missing_fields": [
|
|
123
|
+
"artifact_refs"
|
|
124
|
+
]
|
|
125
|
+
},
|
|
126
|
+
{
|
|
127
|
+
"id": "operations_runner_policy",
|
|
128
|
+
"domain": "operations",
|
|
129
|
+
"artifact_class": "policy",
|
|
130
|
+
"total_score": 0.298,
|
|
131
|
+
"class_specific_completeness": 0.36,
|
|
132
|
+
"ambiguity_remaining": 0.64,
|
|
133
|
+
"present_fields": [
|
|
134
|
+
"invariants",
|
|
135
|
+
"rationale",
|
|
136
|
+
"rule",
|
|
137
|
+
"scope"
|
|
138
|
+
],
|
|
139
|
+
"missing_fields": [
|
|
140
|
+
"enforcement_surface"
|
|
141
|
+
]
|
|
142
|
+
},
|
|
143
|
+
{
|
|
144
|
+
"id": "writing_kernel_launch_result",
|
|
145
|
+
"domain": "writing",
|
|
146
|
+
"artifact_class": "result",
|
|
147
|
+
"total_score": 0.285,
|
|
148
|
+
"class_specific_completeness": 0.36,
|
|
149
|
+
"ambiguity_remaining": 0.64,
|
|
150
|
+
"present_fields": [
|
|
151
|
+
"claim",
|
|
152
|
+
"evidence_paths",
|
|
153
|
+
"next_follow_up",
|
|
154
|
+
"status"
|
|
155
|
+
],
|
|
156
|
+
"missing_fields": [
|
|
157
|
+
"interpretation_limits"
|
|
158
|
+
]
|
|
159
|
+
}
|
|
160
|
+
],
|
|
161
|
+
"mean_total_score": 0.275,
|
|
162
|
+
"mean_class_specific_completeness": 0.328,
|
|
163
|
+
"mean_ambiguity_remaining": 0.672,
|
|
164
|
+
"mean_dimension_scores": {
|
|
165
|
+
"artifact_type_clarity": 0.0,
|
|
166
|
+
"objective_clarity": 0.45,
|
|
167
|
+
"limits_clarity": 0.289,
|
|
168
|
+
"evaluation_clarity": 0.193,
|
|
169
|
+
"handoff_readiness": 0.386,
|
|
170
|
+
"class_specific_completeness": 0.328
|
|
171
|
+
}
|
|
172
|
+
},
|
|
173
|
+
"generic_checklist": {
|
|
174
|
+
"condition": "generic_checklist",
|
|
175
|
+
"cases_total": 7,
|
|
176
|
+
"rows": [
|
|
177
|
+
{
|
|
178
|
+
"id": "software_trace_widget",
|
|
179
|
+
"domain": "software",
|
|
180
|
+
"artifact_class": "task",
|
|
181
|
+
"total_score": 0.697,
|
|
182
|
+
"class_specific_completeness": 0.68,
|
|
183
|
+
"ambiguity_remaining": 0.32,
|
|
184
|
+
"present_fields": [
|
|
185
|
+
"boundary",
|
|
186
|
+
"constraints",
|
|
187
|
+
"goal",
|
|
188
|
+
"object",
|
|
189
|
+
"success_criteria"
|
|
190
|
+
],
|
|
191
|
+
"missing_fields": []
|
|
192
|
+
},
|
|
193
|
+
{
|
|
194
|
+
"id": "product_project_home",
|
|
195
|
+
"domain": "product",
|
|
196
|
+
"artifact_class": "decision",
|
|
197
|
+
"total_score": 0.627,
|
|
198
|
+
"class_specific_completeness": 0.58,
|
|
199
|
+
"ambiguity_remaining": 0.42,
|
|
200
|
+
"present_fields": [
|
|
201
|
+
"chosen_path",
|
|
202
|
+
"consequences",
|
|
203
|
+
"question",
|
|
204
|
+
"rationale",
|
|
205
|
+
"rejected_alternatives"
|
|
206
|
+
],
|
|
207
|
+
"missing_fields": []
|
|
208
|
+
},
|
|
209
|
+
{
|
|
210
|
+
"id": "research_drift_hypothesis",
|
|
211
|
+
"domain": "research",
|
|
212
|
+
"artifact_class": "hypothesis",
|
|
213
|
+
"total_score": 0.675,
|
|
214
|
+
"class_specific_completeness": 0.64,
|
|
215
|
+
"ambiguity_remaining": 0.36,
|
|
216
|
+
"present_fields": [
|
|
217
|
+
"assumptions",
|
|
218
|
+
"boundary",
|
|
219
|
+
"claim",
|
|
220
|
+
"falsifiers",
|
|
221
|
+
"test_path"
|
|
222
|
+
],
|
|
223
|
+
"missing_fields": []
|
|
224
|
+
},
|
|
225
|
+
{
|
|
226
|
+
"id": "research_handoff_experiment",
|
|
227
|
+
"domain": "research",
|
|
228
|
+
"artifact_class": "experiment",
|
|
229
|
+
"total_score": 0.711,
|
|
230
|
+
"class_specific_completeness": 0.692,
|
|
231
|
+
"ambiguity_remaining": 0.308,
|
|
232
|
+
"present_fields": [
|
|
233
|
+
"evidence_expectations",
|
|
234
|
+
"inputs",
|
|
235
|
+
"interpretation_limits",
|
|
236
|
+
"method",
|
|
237
|
+
"objective",
|
|
238
|
+
"outputs"
|
|
239
|
+
],
|
|
240
|
+
"missing_fields": []
|
|
241
|
+
},
|
|
242
|
+
{
|
|
243
|
+
"id": "operations_habanero_checkpoint",
|
|
244
|
+
"domain": "operations",
|
|
245
|
+
"artifact_class": "checkpoint",
|
|
246
|
+
"total_score": 0.679,
|
|
247
|
+
"class_specific_completeness": 0.65,
|
|
248
|
+
"ambiguity_remaining": 0.35,
|
|
249
|
+
"present_fields": [
|
|
250
|
+
"artifact_refs",
|
|
251
|
+
"completed_unit",
|
|
252
|
+
"current_state",
|
|
253
|
+
"next_handoff_target",
|
|
254
|
+
"risks"
|
|
255
|
+
],
|
|
256
|
+
"missing_fields": []
|
|
257
|
+
},
|
|
258
|
+
{
|
|
259
|
+
"id": "operations_runner_policy",
|
|
260
|
+
"domain": "operations",
|
|
261
|
+
"artifact_class": "policy",
|
|
262
|
+
"total_score": 0.713,
|
|
263
|
+
"class_specific_completeness": 0.67,
|
|
264
|
+
"ambiguity_remaining": 0.33,
|
|
265
|
+
"present_fields": [
|
|
266
|
+
"enforcement_surface",
|
|
267
|
+
"invariants",
|
|
268
|
+
"rationale",
|
|
269
|
+
"rule",
|
|
270
|
+
"scope"
|
|
271
|
+
],
|
|
272
|
+
"missing_fields": []
|
|
273
|
+
},
|
|
274
|
+
{
|
|
275
|
+
"id": "writing_kernel_launch_result",
|
|
276
|
+
"domain": "writing",
|
|
277
|
+
"artifact_class": "result",
|
|
278
|
+
"total_score": 0.708,
|
|
279
|
+
"class_specific_completeness": 0.69,
|
|
280
|
+
"ambiguity_remaining": 0.31,
|
|
281
|
+
"present_fields": [
|
|
282
|
+
"claim",
|
|
283
|
+
"evidence_paths",
|
|
284
|
+
"interpretation_limits",
|
|
285
|
+
"next_follow_up",
|
|
286
|
+
"status"
|
|
287
|
+
],
|
|
288
|
+
"missing_fields": []
|
|
289
|
+
}
|
|
290
|
+
],
|
|
291
|
+
"mean_total_score": 0.687,
|
|
292
|
+
"mean_class_specific_completeness": 0.657,
|
|
293
|
+
"mean_ambiguity_remaining": 0.343,
|
|
294
|
+
"mean_dimension_scores": {
|
|
295
|
+
"artifact_type_clarity": 0.85,
|
|
296
|
+
"objective_clarity": 0.596,
|
|
297
|
+
"limits_clarity": 0.693,
|
|
298
|
+
"evaluation_clarity": 0.671,
|
|
299
|
+
"handoff_readiness": 0.655,
|
|
300
|
+
"class_specific_completeness": 0.657
|
|
301
|
+
}
|
|
302
|
+
},
|
|
303
|
+
"kernel": {
|
|
304
|
+
"condition": "kernel",
|
|
305
|
+
"cases_total": 7,
|
|
306
|
+
"rows": [
|
|
307
|
+
{
|
|
308
|
+
"id": "software_trace_widget",
|
|
309
|
+
"domain": "software",
|
|
310
|
+
"artifact_class": "task",
|
|
311
|
+
"total_score": 1.0,
|
|
312
|
+
"class_specific_completeness": 1.0,
|
|
313
|
+
"ambiguity_remaining": 0.0,
|
|
314
|
+
"present_fields": [
|
|
315
|
+
"boundary",
|
|
316
|
+
"constraints",
|
|
317
|
+
"goal",
|
|
318
|
+
"object",
|
|
319
|
+
"success_criteria"
|
|
320
|
+
],
|
|
321
|
+
"missing_fields": []
|
|
322
|
+
},
|
|
323
|
+
{
|
|
324
|
+
"id": "product_project_home",
|
|
325
|
+
"domain": "product",
|
|
326
|
+
"artifact_class": "decision",
|
|
327
|
+
"total_score": 1.0,
|
|
328
|
+
"class_specific_completeness": 1.0,
|
|
329
|
+
"ambiguity_remaining": 0.0,
|
|
330
|
+
"present_fields": [
|
|
331
|
+
"chosen_path",
|
|
332
|
+
"consequences",
|
|
333
|
+
"question",
|
|
334
|
+
"rationale",
|
|
335
|
+
"rejected_alternatives"
|
|
336
|
+
],
|
|
337
|
+
"missing_fields": []
|
|
338
|
+
},
|
|
339
|
+
{
|
|
340
|
+
"id": "research_drift_hypothesis",
|
|
341
|
+
"domain": "research",
|
|
342
|
+
"artifact_class": "hypothesis",
|
|
343
|
+
"total_score": 1.0,
|
|
344
|
+
"class_specific_completeness": 1.0,
|
|
345
|
+
"ambiguity_remaining": 0.0,
|
|
346
|
+
"present_fields": [
|
|
347
|
+
"assumptions",
|
|
348
|
+
"boundary",
|
|
349
|
+
"claim",
|
|
350
|
+
"falsifiers",
|
|
351
|
+
"test_path"
|
|
352
|
+
],
|
|
353
|
+
"missing_fields": []
|
|
354
|
+
},
|
|
355
|
+
{
|
|
356
|
+
"id": "research_handoff_experiment",
|
|
357
|
+
"domain": "research",
|
|
358
|
+
"artifact_class": "experiment",
|
|
359
|
+
"total_score": 1.0,
|
|
360
|
+
"class_specific_completeness": 1.0,
|
|
361
|
+
"ambiguity_remaining": 0.0,
|
|
362
|
+
"present_fields": [
|
|
363
|
+
"evidence_expectations",
|
|
364
|
+
"inputs",
|
|
365
|
+
"interpretation_limits",
|
|
366
|
+
"method",
|
|
367
|
+
"objective",
|
|
368
|
+
"outputs"
|
|
369
|
+
],
|
|
370
|
+
"missing_fields": []
|
|
371
|
+
},
|
|
372
|
+
{
|
|
373
|
+
"id": "operations_habanero_checkpoint",
|
|
374
|
+
"domain": "operations",
|
|
375
|
+
"artifact_class": "checkpoint",
|
|
376
|
+
"total_score": 1.0,
|
|
377
|
+
"class_specific_completeness": 1.0,
|
|
378
|
+
"ambiguity_remaining": 0.0,
|
|
379
|
+
"present_fields": [
|
|
380
|
+
"artifact_refs",
|
|
381
|
+
"completed_unit",
|
|
382
|
+
"current_state",
|
|
383
|
+
"next_handoff_target",
|
|
384
|
+
"risks"
|
|
385
|
+
],
|
|
386
|
+
"missing_fields": []
|
|
387
|
+
},
|
|
388
|
+
{
|
|
389
|
+
"id": "operations_runner_policy",
|
|
390
|
+
"domain": "operations",
|
|
391
|
+
"artifact_class": "policy",
|
|
392
|
+
"total_score": 1.0,
|
|
393
|
+
"class_specific_completeness": 1.0,
|
|
394
|
+
"ambiguity_remaining": 0.0,
|
|
395
|
+
"present_fields": [
|
|
396
|
+
"enforcement_surface",
|
|
397
|
+
"invariants",
|
|
398
|
+
"rationale",
|
|
399
|
+
"rule",
|
|
400
|
+
"scope"
|
|
401
|
+
],
|
|
402
|
+
"missing_fields": []
|
|
403
|
+
},
|
|
404
|
+
{
|
|
405
|
+
"id": "writing_kernel_launch_result",
|
|
406
|
+
"domain": "writing",
|
|
407
|
+
"artifact_class": "result",
|
|
408
|
+
"total_score": 1.0,
|
|
409
|
+
"class_specific_completeness": 1.0,
|
|
410
|
+
"ambiguity_remaining": 0.0,
|
|
411
|
+
"present_fields": [
|
|
412
|
+
"claim",
|
|
413
|
+
"evidence_paths",
|
|
414
|
+
"interpretation_limits",
|
|
415
|
+
"next_follow_up",
|
|
416
|
+
"status"
|
|
417
|
+
],
|
|
418
|
+
"missing_fields": []
|
|
419
|
+
}
|
|
420
|
+
],
|
|
421
|
+
"mean_total_score": 1.0,
|
|
422
|
+
"mean_class_specific_completeness": 1.0,
|
|
423
|
+
"mean_ambiguity_remaining": 0.0,
|
|
424
|
+
"mean_dimension_scores": {
|
|
425
|
+
"artifact_type_clarity": 1.0,
|
|
426
|
+
"objective_clarity": 1.0,
|
|
427
|
+
"limits_clarity": 1.0,
|
|
428
|
+
"evaluation_clarity": 1.0,
|
|
429
|
+
"handoff_readiness": 1.0,
|
|
430
|
+
"class_specific_completeness": 1.0
|
|
431
|
+
}
|
|
432
|
+
}
|
|
433
|
+
},
|
|
434
|
+
"pairwise": {
|
|
435
|
+
"kernel_vs_generic_checklist": {
|
|
436
|
+
"left": "kernel",
|
|
437
|
+
"right": "generic_checklist",
|
|
438
|
+
"wins": 7,
|
|
439
|
+
"ties": 0,
|
|
440
|
+
"losses": 0,
|
|
441
|
+
"mean_total_score_delta": 0.313,
|
|
442
|
+
"by_case": [
|
|
443
|
+
{
|
|
444
|
+
"id": "software_trace_widget",
|
|
445
|
+
"domain": "software",
|
|
446
|
+
"artifact_class": "task",
|
|
447
|
+
"left_score": 1.0,
|
|
448
|
+
"right_score": 0.697,
|
|
449
|
+
"delta": 0.303,
|
|
450
|
+
"outcome": "win"
|
|
451
|
+
},
|
|
452
|
+
{
|
|
453
|
+
"id": "product_project_home",
|
|
454
|
+
"domain": "product",
|
|
455
|
+
"artifact_class": "decision",
|
|
456
|
+
"left_score": 1.0,
|
|
457
|
+
"right_score": 0.627,
|
|
458
|
+
"delta": 0.373,
|
|
459
|
+
"outcome": "win"
|
|
460
|
+
},
|
|
461
|
+
{
|
|
462
|
+
"id": "research_drift_hypothesis",
|
|
463
|
+
"domain": "research",
|
|
464
|
+
"artifact_class": "hypothesis",
|
|
465
|
+
"left_score": 1.0,
|
|
466
|
+
"right_score": 0.675,
|
|
467
|
+
"delta": 0.325,
|
|
468
|
+
"outcome": "win"
|
|
469
|
+
},
|
|
470
|
+
{
|
|
471
|
+
"id": "research_handoff_experiment",
|
|
472
|
+
"domain": "research",
|
|
473
|
+
"artifact_class": "experiment",
|
|
474
|
+
"left_score": 1.0,
|
|
475
|
+
"right_score": 0.711,
|
|
476
|
+
"delta": 0.289,
|
|
477
|
+
"outcome": "win"
|
|
478
|
+
},
|
|
479
|
+
{
|
|
480
|
+
"id": "operations_habanero_checkpoint",
|
|
481
|
+
"domain": "operations",
|
|
482
|
+
"artifact_class": "checkpoint",
|
|
483
|
+
"left_score": 1.0,
|
|
484
|
+
"right_score": 0.679,
|
|
485
|
+
"delta": 0.321,
|
|
486
|
+
"outcome": "win"
|
|
487
|
+
},
|
|
488
|
+
{
|
|
489
|
+
"id": "operations_runner_policy",
|
|
490
|
+
"domain": "operations",
|
|
491
|
+
"artifact_class": "policy",
|
|
492
|
+
"left_score": 1.0,
|
|
493
|
+
"right_score": 0.713,
|
|
494
|
+
"delta": 0.287,
|
|
495
|
+
"outcome": "win"
|
|
496
|
+
},
|
|
497
|
+
{
|
|
498
|
+
"id": "writing_kernel_launch_result",
|
|
499
|
+
"domain": "writing",
|
|
500
|
+
"artifact_class": "result",
|
|
501
|
+
"left_score": 1.0,
|
|
502
|
+
"right_score": 0.708,
|
|
503
|
+
"delta": 0.292,
|
|
504
|
+
"outcome": "win"
|
|
505
|
+
}
|
|
506
|
+
]
|
|
507
|
+
},
|
|
508
|
+
"kernel_vs_freeform": {
|
|
509
|
+
"left": "kernel",
|
|
510
|
+
"right": "freeform",
|
|
511
|
+
"wins": 7,
|
|
512
|
+
"ties": 0,
|
|
513
|
+
"losses": 0,
|
|
514
|
+
"mean_total_score_delta": 0.725,
|
|
515
|
+
"by_case": [
|
|
516
|
+
{
|
|
517
|
+
"id": "software_trace_widget",
|
|
518
|
+
"domain": "software",
|
|
519
|
+
"artifact_class": "task",
|
|
520
|
+
"left_score": 1.0,
|
|
521
|
+
"right_score": 0.208,
|
|
522
|
+
"delta": 0.792,
|
|
523
|
+
"outcome": "win"
|
|
524
|
+
},
|
|
525
|
+
{
|
|
526
|
+
"id": "product_project_home",
|
|
527
|
+
"domain": "product",
|
|
528
|
+
"artifact_class": "decision",
|
|
529
|
+
"left_score": 1.0,
|
|
530
|
+
"right_score": 0.323,
|
|
531
|
+
"delta": 0.677,
|
|
532
|
+
"outcome": "win"
|
|
533
|
+
},
|
|
534
|
+
{
|
|
535
|
+
"id": "research_drift_hypothesis",
|
|
536
|
+
"domain": "research",
|
|
537
|
+
"artifact_class": "hypothesis",
|
|
538
|
+
"left_score": 1.0,
|
|
539
|
+
"right_score": 0.323,
|
|
540
|
+
"delta": 0.677,
|
|
541
|
+
"outcome": "win"
|
|
542
|
+
},
|
|
543
|
+
{
|
|
544
|
+
"id": "research_handoff_experiment",
|
|
545
|
+
"domain": "research",
|
|
546
|
+
"artifact_class": "experiment",
|
|
547
|
+
"left_score": 1.0,
|
|
548
|
+
"right_score": 0.2,
|
|
549
|
+
"delta": 0.8,
|
|
550
|
+
"outcome": "win"
|
|
551
|
+
},
|
|
552
|
+
{
|
|
553
|
+
"id": "operations_habanero_checkpoint",
|
|
554
|
+
"domain": "operations",
|
|
555
|
+
"artifact_class": "checkpoint",
|
|
556
|
+
"left_score": 1.0,
|
|
557
|
+
"right_score": 0.285,
|
|
558
|
+
"delta": 0.715,
|
|
559
|
+
"outcome": "win"
|
|
560
|
+
},
|
|
561
|
+
{
|
|
562
|
+
"id": "operations_runner_policy",
|
|
563
|
+
"domain": "operations",
|
|
564
|
+
"artifact_class": "policy",
|
|
565
|
+
"left_score": 1.0,
|
|
566
|
+
"right_score": 0.298,
|
|
567
|
+
"delta": 0.702,
|
|
568
|
+
"outcome": "win"
|
|
569
|
+
},
|
|
570
|
+
{
|
|
571
|
+
"id": "writing_kernel_launch_result",
|
|
572
|
+
"domain": "writing",
|
|
573
|
+
"artifact_class": "result",
|
|
574
|
+
"left_score": 1.0,
|
|
575
|
+
"right_score": 0.285,
|
|
576
|
+
"delta": 0.715,
|
|
577
|
+
"outcome": "win"
|
|
578
|
+
}
|
|
579
|
+
]
|
|
580
|
+
},
|
|
581
|
+
"generic_checklist_vs_freeform": {
|
|
582
|
+
"left": "generic_checklist",
|
|
583
|
+
"right": "freeform",
|
|
584
|
+
"wins": 7,
|
|
585
|
+
"ties": 0,
|
|
586
|
+
"losses": 0,
|
|
587
|
+
"mean_total_score_delta": 0.413,
|
|
588
|
+
"by_case": [
|
|
589
|
+
{
|
|
590
|
+
"id": "software_trace_widget",
|
|
591
|
+
"domain": "software",
|
|
592
|
+
"artifact_class": "task",
|
|
593
|
+
"left_score": 0.697,
|
|
594
|
+
"right_score": 0.208,
|
|
595
|
+
"delta": 0.489,
|
|
596
|
+
"outcome": "win"
|
|
597
|
+
},
|
|
598
|
+
{
|
|
599
|
+
"id": "product_project_home",
|
|
600
|
+
"domain": "product",
|
|
601
|
+
"artifact_class": "decision",
|
|
602
|
+
"left_score": 0.627,
|
|
603
|
+
"right_score": 0.323,
|
|
604
|
+
"delta": 0.304,
|
|
605
|
+
"outcome": "win"
|
|
606
|
+
},
|
|
607
|
+
{
|
|
608
|
+
"id": "research_drift_hypothesis",
|
|
609
|
+
"domain": "research",
|
|
610
|
+
"artifact_class": "hypothesis",
|
|
611
|
+
"left_score": 0.675,
|
|
612
|
+
"right_score": 0.323,
|
|
613
|
+
"delta": 0.352,
|
|
614
|
+
"outcome": "win"
|
|
615
|
+
},
|
|
616
|
+
{
|
|
617
|
+
"id": "research_handoff_experiment",
|
|
618
|
+
"domain": "research",
|
|
619
|
+
"artifact_class": "experiment",
|
|
620
|
+
"left_score": 0.711,
|
|
621
|
+
"right_score": 0.2,
|
|
622
|
+
"delta": 0.511,
|
|
623
|
+
"outcome": "win"
|
|
624
|
+
},
|
|
625
|
+
{
|
|
626
|
+
"id": "operations_habanero_checkpoint",
|
|
627
|
+
"domain": "operations",
|
|
628
|
+
"artifact_class": "checkpoint",
|
|
629
|
+
"left_score": 0.679,
|
|
630
|
+
"right_score": 0.285,
|
|
631
|
+
"delta": 0.394,
|
|
632
|
+
"outcome": "win"
|
|
633
|
+
},
|
|
634
|
+
{
|
|
635
|
+
"id": "operations_runner_policy",
|
|
636
|
+
"domain": "operations",
|
|
637
|
+
"artifact_class": "policy",
|
|
638
|
+
"left_score": 0.713,
|
|
639
|
+
"right_score": 0.298,
|
|
640
|
+
"delta": 0.415,
|
|
641
|
+
"outcome": "win"
|
|
642
|
+
},
|
|
643
|
+
{
|
|
644
|
+
"id": "writing_kernel_launch_result",
|
|
645
|
+
"domain": "writing",
|
|
646
|
+
"artifact_class": "result",
|
|
647
|
+
"left_score": 0.708,
|
|
648
|
+
"right_score": 0.285,
|
|
649
|
+
"delta": 0.423,
|
|
650
|
+
"outcome": "win"
|
|
651
|
+
}
|
|
652
|
+
]
|
|
653
|
+
}
|
|
654
|
+
},
|
|
655
|
+
"claims": [
|
|
656
|
+
{
|
|
657
|
+
"id": "matched_internal_corpus_exists",
|
|
658
|
+
"claim": "ORP has a matched internal comparison corpus spanning multiple domains and all seven kernel artifact classes.",
|
|
659
|
+
"status": "pass"
|
|
660
|
+
},
|
|
661
|
+
{
|
|
662
|
+
"id": "kernel_outscores_generic_checklist_on_matched_corpus",
|
|
663
|
+
"claim": "On the matched internal comparison corpus, kernel artifacts achieve higher mean structural scores than generic checklist artifacts.",
|
|
664
|
+
"status": "pass"
|
|
665
|
+
},
|
|
666
|
+
{
|
|
667
|
+
"id": "kernel_outscores_freeform_on_matched_corpus",
|
|
668
|
+
"claim": "On the matched internal comparison corpus, kernel artifacts achieve higher mean structural scores than free-form artifacts.",
|
|
669
|
+
"status": "pass"
|
|
670
|
+
},
|
|
671
|
+
{
|
|
672
|
+
"id": "generic_checklist_improves_on_freeform_for_structure",
|
|
673
|
+
"claim": "On the matched internal comparison corpus, a generic checklist condition improves structural scores over free-form artifacts.",
|
|
674
|
+
"status": "pass"
|
|
675
|
+
},
|
|
676
|
+
{
|
|
677
|
+
"id": "kernel_preserves_full_required_coverage",
|
|
678
|
+
"claim": "On the matched internal comparison corpus, kernel artifacts preserve full class-specific required-field coverage.",
|
|
679
|
+
"status": "pass"
|
|
680
|
+
}
|
|
681
|
+
],
|
|
682
|
+
"summary": {
|
|
683
|
+
"all_claims_pass": true,
|
|
684
|
+
"kernel_mean_total_score": 1.0,
|
|
685
|
+
"generic_checklist_mean_total_score": 0.687,
|
|
686
|
+
"freeform_mean_total_score": 0.275
|
|
687
|
+
}
|
|
688
|
+
}
|