@skill-harness/core 0.16.0 → 0.18.0
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 +1 -2
- package/dist/adapters/types.d.ts +0 -12
- package/dist/capture-trace-types.d.ts +0 -96
- package/dist/capture-trace-types.js +0 -4
- package/dist/command-cost.d.ts +1 -1
- package/dist/command-cost.js +0 -5
- package/dist/execution-trace.js +1 -1
- package/dist/index.d.ts +3 -23
- package/dist/index.js +3 -23
- package/dist/instruction-coverage.d.ts +0 -7
- package/dist/instruction-coverage.js +5 -7
- package/dist/journal.d.ts +0 -8
- package/dist/metrics.d.ts +18 -0
- package/dist/metrics.js +28 -0
- package/dist/redaction.d.ts +6 -0
- package/dist/redaction.js +55 -0
- package/dist/regate.js +6 -71
- package/dist/regrade.d.ts +1 -3
- package/dist/regrade.js +1 -21
- package/dist/report.d.ts +1 -15
- package/dist/report.js +1 -13
- package/dist/reps.d.ts +1 -2
- package/dist/reps.js +0 -2
- package/dist/results.d.ts +0 -9
- package/dist/results.js +2 -19
- package/dist/run.js +11 -72
- package/dist/seeded.d.ts +1 -7
- package/dist/seeded.js +5 -11
- package/dist/sources.js +4 -9
- package/dist/spec.d.ts +0 -13
- package/dist/spec.js +6 -48
- package/dist/trajectory-events.d.ts +58 -0
- package/dist/trajectory-events.js +119 -0
- package/package.json +1 -1
- package/schemas/results-v2.schema.json +376 -75
- package/schemas/specification-v1.schema.json +128 -174
- package/dist/adjudication.d.ts +0 -219
- package/dist/adjudication.js +0 -422
- package/dist/adoption.d.ts +0 -128
- package/dist/adoption.js +0 -102
- package/dist/affected.d.ts +0 -88
- package/dist/affected.js +0 -222
- package/dist/capture.d.ts +0 -193
- package/dist/capture.js +0 -344
- package/dist/comparison.d.ts +0 -97
- package/dist/comparison.js +0 -189
- package/dist/factory-calibration.d.ts +0 -63
- package/dist/factory-calibration.js +0 -115
- package/dist/intervention-results.d.ts +0 -6
- package/dist/intervention-results.js +0 -52
- package/dist/intervention.d.ts +0 -123
- package/dist/intervention.js +0 -218
- package/dist/investigation.d.ts +0 -94
- package/dist/investigation.js +0 -148
- package/dist/judge-agreement.d.ts +0 -19
- package/dist/judge-agreement.js +0 -31
- package/dist/qualification-capture.d.ts +0 -44
- package/dist/qualification-capture.js +0 -163
- package/dist/qualification-config.d.ts +0 -233
- package/dist/qualification-config.js +0 -694
- package/dist/qualification-lock.d.ts +0 -15
- package/dist/qualification-lock.js +0 -87
- package/dist/qualification-oauth-directory.d.ts +0 -79
- package/dist/qualification-oauth-directory.js +0 -444
- package/dist/qualification-panel-store.d.ts +0 -14
- package/dist/qualification-panel-store.js +0 -188
- package/dist/qualification-panels.d.ts +0 -99
- package/dist/qualification-panels.js +0 -91
- package/dist/qualification-process.d.ts +0 -13
- package/dist/qualification-process.js +0 -190
- package/dist/qualification-runner.d.ts +0 -203
- package/dist/qualification-runner.js +0 -2037
- package/dist/qualification-store.d.ts +0 -187
- package/dist/qualification-store.js +0 -833
- package/dist/qualification-validation.d.ts +0 -12
- package/dist/qualification-validation.js +0 -8
- package/dist/screen.d.ts +0 -31
- package/dist/screen.js +0 -105
- package/dist/trajectory-gates.d.ts +0 -137
- package/dist/trajectory-gates.js +0 -576
- package/dist/trusted-host-supervision.d.ts +0 -25
- package/dist/trusted-host-supervision.js +0 -94
- package/dist/work-capture-schema.d.ts +0 -38
- package/dist/work-capture-schema.js +0 -33
- package/dist/work-capture.d.ts +0 -43
- package/dist/work-capture.js +0 -81
- package/dist/work-signals.d.ts +0 -55
- package/dist/work-signals.js +0 -71
- package/schemas/qualification-cell-result-v1.schema.json +0 -32
- package/schemas/qualification-config-v1.schema.json +0 -244
- package/schemas/qualification-invocation-request-v1.schema.json +0 -59
- package/schemas/qualification-judge-panel-result-v1.schema.json +0 -42
- package/schemas/trajectory-event-v1.schema.json +0 -80
|
@@ -3,128 +3,429 @@
|
|
|
3
3
|
"$id": "https://skill-harness.dev/schemas/results-v2.schema.json",
|
|
4
4
|
"title": "skill-harness results v2",
|
|
5
5
|
"type": "object",
|
|
6
|
-
"required": [
|
|
6
|
+
"required": [
|
|
7
|
+
"schema",
|
|
8
|
+
"skill",
|
|
9
|
+
"harness",
|
|
10
|
+
"model",
|
|
11
|
+
"judge",
|
|
12
|
+
"timestamp",
|
|
13
|
+
"label",
|
|
14
|
+
"mode",
|
|
15
|
+
"effective_grade",
|
|
16
|
+
"scenarios"
|
|
17
|
+
],
|
|
7
18
|
"properties": {
|
|
8
|
-
"schema": {
|
|
9
|
-
|
|
10
|
-
|
|
11
|
-
"
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
"
|
|
19
|
+
"schema": {
|
|
20
|
+
"const": 2
|
|
21
|
+
},
|
|
22
|
+
"harness_version": {
|
|
23
|
+
"type": "string"
|
|
24
|
+
},
|
|
25
|
+
"harness_cli_version": {
|
|
26
|
+
"type": "string"
|
|
27
|
+
},
|
|
28
|
+
"delivery_canary": {
|
|
29
|
+
"enum": [
|
|
30
|
+
"pass",
|
|
31
|
+
"skipped"
|
|
32
|
+
]
|
|
33
|
+
},
|
|
34
|
+
"skill": {
|
|
35
|
+
"type": "string"
|
|
36
|
+
},
|
|
37
|
+
"harness": {
|
|
38
|
+
"type": "string"
|
|
39
|
+
},
|
|
40
|
+
"model": {
|
|
41
|
+
"type": "string"
|
|
42
|
+
},
|
|
15
43
|
"judge": {
|
|
16
44
|
"type": "object",
|
|
17
|
-
"required": [
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
45
|
+
"required": [
|
|
46
|
+
"provider",
|
|
47
|
+
"model"
|
|
48
|
+
],
|
|
49
|
+
"properties": {
|
|
50
|
+
"provider": {
|
|
51
|
+
"type": "string"
|
|
52
|
+
},
|
|
53
|
+
"model": {
|
|
54
|
+
"type": "string"
|
|
55
|
+
}
|
|
56
|
+
}
|
|
57
|
+
},
|
|
58
|
+
"timestamp": {
|
|
59
|
+
"type": "string"
|
|
60
|
+
},
|
|
61
|
+
"label": {
|
|
62
|
+
"type": [
|
|
63
|
+
"string",
|
|
64
|
+
"null"
|
|
65
|
+
]
|
|
66
|
+
},
|
|
67
|
+
"mode": {
|
|
68
|
+
"enum": [
|
|
69
|
+
"red",
|
|
70
|
+
"green",
|
|
71
|
+
"force"
|
|
72
|
+
]
|
|
73
|
+
},
|
|
74
|
+
"partial": {
|
|
75
|
+
"type": "boolean"
|
|
76
|
+
},
|
|
77
|
+
"source_hashes": {
|
|
78
|
+
"type": "object",
|
|
79
|
+
"additionalProperties": {
|
|
80
|
+
"type": "string"
|
|
81
|
+
}
|
|
82
|
+
},
|
|
83
|
+
"effective_grade": {
|
|
84
|
+
"$ref": "#/$defs/grade"
|
|
85
|
+
},
|
|
86
|
+
"scenarios": {
|
|
87
|
+
"type": "array",
|
|
88
|
+
"items": {
|
|
89
|
+
"$ref": "#/$defs/scenario"
|
|
90
|
+
}
|
|
91
|
+
}
|
|
27
92
|
},
|
|
28
93
|
"$defs": {
|
|
29
|
-
"verdict": {
|
|
94
|
+
"verdict": {
|
|
95
|
+
"enum": [
|
|
96
|
+
"PASS",
|
|
97
|
+
"FAIL",
|
|
98
|
+
"ERROR",
|
|
99
|
+
"JUDGE-AMBIGUOUS"
|
|
100
|
+
]
|
|
101
|
+
},
|
|
30
102
|
"grade": {
|
|
31
103
|
"type": "object",
|
|
32
|
-
"required": [
|
|
104
|
+
"required": [
|
|
105
|
+
"passed",
|
|
106
|
+
"total",
|
|
107
|
+
"pct",
|
|
108
|
+
"letter",
|
|
109
|
+
"ship",
|
|
110
|
+
"note"
|
|
111
|
+
],
|
|
33
112
|
"properties": {
|
|
34
|
-
"passed": {
|
|
35
|
-
|
|
113
|
+
"passed": {
|
|
114
|
+
"type": "integer"
|
|
115
|
+
},
|
|
116
|
+
"total": {
|
|
117
|
+
"type": "integer"
|
|
118
|
+
},
|
|
119
|
+
"pct": {
|
|
120
|
+
"type": "number"
|
|
121
|
+
},
|
|
122
|
+
"letter": {
|
|
123
|
+
"type": "string"
|
|
124
|
+
},
|
|
125
|
+
"ship": {
|
|
126
|
+
"type": "boolean"
|
|
127
|
+
},
|
|
128
|
+
"note": {
|
|
129
|
+
"type": "string"
|
|
130
|
+
}
|
|
36
131
|
}
|
|
37
132
|
},
|
|
38
133
|
"metrics": {
|
|
39
134
|
"type": "object",
|
|
40
|
-
"required": [
|
|
135
|
+
"required": [
|
|
136
|
+
"wall_time_ms",
|
|
137
|
+
"judge_calls",
|
|
138
|
+
"judge_rejudge_calls",
|
|
139
|
+
"subject_metrics_reps",
|
|
140
|
+
"total_reps"
|
|
141
|
+
],
|
|
41
142
|
"properties": {
|
|
42
|
-
"wall_time_ms": {
|
|
43
|
-
|
|
44
|
-
|
|
45
|
-
|
|
46
|
-
"
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
"
|
|
51
|
-
|
|
52
|
-
|
|
53
|
-
|
|
54
|
-
"
|
|
55
|
-
|
|
143
|
+
"wall_time_ms": {
|
|
144
|
+
"type": "number",
|
|
145
|
+
"minimum": 0
|
|
146
|
+
},
|
|
147
|
+
"judge_calls": {
|
|
148
|
+
"type": "integer",
|
|
149
|
+
"minimum": 0
|
|
150
|
+
},
|
|
151
|
+
"judge_rejudge_calls": {
|
|
152
|
+
"type": "integer",
|
|
153
|
+
"minimum": 0
|
|
154
|
+
},
|
|
155
|
+
"subject_metrics_reps": {
|
|
156
|
+
"type": "integer",
|
|
157
|
+
"minimum": 0
|
|
158
|
+
},
|
|
159
|
+
"total_reps": {
|
|
160
|
+
"type": "integer",
|
|
161
|
+
"minimum": 1
|
|
162
|
+
},
|
|
163
|
+
"input_tokens": {
|
|
164
|
+
"type": "integer",
|
|
165
|
+
"minimum": 0
|
|
166
|
+
},
|
|
167
|
+
"output_tokens": {
|
|
168
|
+
"type": "integer",
|
|
169
|
+
"minimum": 0
|
|
170
|
+
},
|
|
171
|
+
"cache_read_tokens": {
|
|
172
|
+
"type": "integer",
|
|
173
|
+
"minimum": 0
|
|
174
|
+
},
|
|
175
|
+
"cache_write_tokens": {
|
|
176
|
+
"type": "integer",
|
|
177
|
+
"minimum": 0
|
|
178
|
+
},
|
|
179
|
+
"subject_cost_usd": {
|
|
180
|
+
"type": "number",
|
|
181
|
+
"minimum": 0
|
|
182
|
+
},
|
|
183
|
+
"cost_source": {
|
|
184
|
+
"enum": [
|
|
185
|
+
"provider-reported",
|
|
186
|
+
"subscription",
|
|
187
|
+
"unreported"
|
|
188
|
+
]
|
|
189
|
+
},
|
|
190
|
+
"tool_calls": {
|
|
191
|
+
"type": "integer",
|
|
192
|
+
"minimum": 0
|
|
193
|
+
},
|
|
194
|
+
"delegated_children": {
|
|
195
|
+
"type": "integer",
|
|
196
|
+
"minimum": 0
|
|
197
|
+
},
|
|
198
|
+
"max_concurrency": {
|
|
199
|
+
"type": "integer",
|
|
200
|
+
"minimum": 0
|
|
201
|
+
}
|
|
56
202
|
}
|
|
57
203
|
},
|
|
58
204
|
"judgment": {
|
|
59
205
|
"type": "object",
|
|
60
206
|
"additionalProperties": false,
|
|
61
|
-
"required": [
|
|
207
|
+
"required": [
|
|
208
|
+
"ordinal",
|
|
209
|
+
"judge",
|
|
210
|
+
"verdict",
|
|
211
|
+
"reason",
|
|
212
|
+
"suspect"
|
|
213
|
+
],
|
|
62
214
|
"properties": {
|
|
63
|
-
"ordinal": {
|
|
215
|
+
"ordinal": {
|
|
216
|
+
"type": "integer",
|
|
217
|
+
"minimum": 1,
|
|
218
|
+
"maximum": 3
|
|
219
|
+
},
|
|
64
220
|
"judge": {
|
|
65
|
-
"type": "object",
|
|
66
|
-
"
|
|
221
|
+
"type": "object",
|
|
222
|
+
"additionalProperties": false,
|
|
223
|
+
"required": [
|
|
224
|
+
"provider",
|
|
225
|
+
"model"
|
|
226
|
+
],
|
|
227
|
+
"properties": {
|
|
228
|
+
"provider": {
|
|
229
|
+
"type": "string"
|
|
230
|
+
},
|
|
231
|
+
"model": {
|
|
232
|
+
"type": "string"
|
|
233
|
+
}
|
|
234
|
+
}
|
|
235
|
+
},
|
|
236
|
+
"verdict": {
|
|
237
|
+
"$ref": "#/$defs/verdict"
|
|
238
|
+
},
|
|
239
|
+
"reason": {
|
|
240
|
+
"type": "string"
|
|
67
241
|
},
|
|
68
|
-
"
|
|
69
|
-
|
|
70
|
-
|
|
242
|
+
"suspect": {
|
|
243
|
+
"type": "boolean"
|
|
244
|
+
}
|
|
71
245
|
}
|
|
72
246
|
},
|
|
73
247
|
"scenario": {
|
|
74
248
|
"type": "object",
|
|
75
|
-
"required": [
|
|
249
|
+
"required": [
|
|
250
|
+
"id",
|
|
251
|
+
"judge_verdict",
|
|
252
|
+
"judge_reason",
|
|
253
|
+
"suspect",
|
|
254
|
+
"override",
|
|
255
|
+
"note"
|
|
256
|
+
],
|
|
76
257
|
"properties": {
|
|
77
|
-
"id": {
|
|
78
|
-
|
|
79
|
-
|
|
80
|
-
"
|
|
81
|
-
|
|
82
|
-
|
|
83
|
-
"
|
|
84
|
-
|
|
85
|
-
|
|
86
|
-
"
|
|
87
|
-
|
|
88
|
-
|
|
258
|
+
"id": {
|
|
259
|
+
"type": "string"
|
|
260
|
+
},
|
|
261
|
+
"judge_verdict": {
|
|
262
|
+
"$ref": "#/$defs/verdict"
|
|
263
|
+
},
|
|
264
|
+
"judge_reason": {
|
|
265
|
+
"type": "string"
|
|
266
|
+
},
|
|
267
|
+
"suspect": {
|
|
268
|
+
"type": "boolean"
|
|
269
|
+
},
|
|
270
|
+
"override": {
|
|
271
|
+
"oneOf": [
|
|
272
|
+
{
|
|
273
|
+
"const": null
|
|
274
|
+
},
|
|
275
|
+
{
|
|
276
|
+
"enum": [
|
|
277
|
+
"PASS",
|
|
278
|
+
"FAIL"
|
|
279
|
+
]
|
|
280
|
+
}
|
|
281
|
+
]
|
|
282
|
+
},
|
|
283
|
+
"note": {
|
|
284
|
+
"type": "string"
|
|
285
|
+
},
|
|
286
|
+
"reps": {
|
|
287
|
+
"type": "integer",
|
|
288
|
+
"minimum": 1
|
|
289
|
+
},
|
|
290
|
+
"passes": {
|
|
291
|
+
"type": "integer",
|
|
292
|
+
"minimum": 0
|
|
293
|
+
},
|
|
294
|
+
"clean": {
|
|
295
|
+
"type": "integer",
|
|
296
|
+
"minimum": 0
|
|
297
|
+
},
|
|
298
|
+
"flakiness": {
|
|
299
|
+
"type": "number",
|
|
300
|
+
"minimum": 0,
|
|
301
|
+
"maximum": 1
|
|
302
|
+
},
|
|
303
|
+
"pass_threshold": {
|
|
304
|
+
"type": "number",
|
|
305
|
+
"minimum": 0,
|
|
306
|
+
"maximum": 1
|
|
307
|
+
},
|
|
308
|
+
"metrics": {
|
|
309
|
+
"$ref": "#/$defs/metrics"
|
|
310
|
+
},
|
|
89
311
|
"objective": {
|
|
90
312
|
"type": "object",
|
|
91
313
|
"additionalProperties": false,
|
|
92
|
-
"required": [
|
|
314
|
+
"required": [
|
|
315
|
+
"status",
|
|
316
|
+
"assertions"
|
|
317
|
+
],
|
|
93
318
|
"properties": {
|
|
94
|
-
"status": {
|
|
95
|
-
|
|
96
|
-
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
|
|
319
|
+
"status": {
|
|
320
|
+
"enum": [
|
|
321
|
+
"PASS",
|
|
322
|
+
"FAIL",
|
|
323
|
+
"ERROR"
|
|
324
|
+
]
|
|
325
|
+
},
|
|
326
|
+
"trace_version": {
|
|
327
|
+
"type": "integer"
|
|
328
|
+
},
|
|
329
|
+
"trace_sha256": {
|
|
330
|
+
"type": "string"
|
|
331
|
+
},
|
|
332
|
+
"rep_trace_sha256": {
|
|
333
|
+
"type": "array",
|
|
334
|
+
"items": {
|
|
335
|
+
"type": "string",
|
|
336
|
+
"pattern": "^[a-fA-F0-9]{64}$"
|
|
337
|
+
}
|
|
338
|
+
},
|
|
101
339
|
"assertions": {
|
|
102
340
|
"type": "array",
|
|
103
341
|
"items": {
|
|
104
342
|
"type": "object",
|
|
105
343
|
"additionalProperties": false,
|
|
106
|
-
"required": [
|
|
344
|
+
"required": [
|
|
345
|
+
"kind",
|
|
346
|
+
"status",
|
|
347
|
+
"detail"
|
|
348
|
+
],
|
|
107
349
|
"properties": {
|
|
108
|
-
"kind": {
|
|
109
|
-
|
|
110
|
-
|
|
350
|
+
"kind": {
|
|
351
|
+
"type": "string",
|
|
352
|
+
"minLength": 1
|
|
353
|
+
},
|
|
354
|
+
"status": {
|
|
355
|
+
"enum": [
|
|
356
|
+
"PASS",
|
|
357
|
+
"FAIL",
|
|
358
|
+
"ERROR"
|
|
359
|
+
]
|
|
360
|
+
},
|
|
361
|
+
"detail": {
|
|
362
|
+
"type": "string"
|
|
363
|
+
}
|
|
111
364
|
}
|
|
112
365
|
}
|
|
366
|
+
},
|
|
367
|
+
"trajectory_version": {
|
|
368
|
+
"type": "string"
|
|
369
|
+
},
|
|
370
|
+
"events_sha256": {
|
|
371
|
+
"type": "string",
|
|
372
|
+
"pattern": "^[a-fA-F0-9]{64}$"
|
|
373
|
+
},
|
|
374
|
+
"rep_events_sha256": {
|
|
375
|
+
"type": "array",
|
|
376
|
+
"items": {
|
|
377
|
+
"type": "string",
|
|
378
|
+
"pattern": "^[a-fA-F0-9]{64}$"
|
|
379
|
+
}
|
|
113
380
|
}
|
|
114
381
|
}
|
|
115
382
|
},
|
|
116
383
|
"adjudication": {
|
|
117
384
|
"type": "object",
|
|
118
385
|
"additionalProperties": false,
|
|
119
|
-
"required": [
|
|
386
|
+
"required": [
|
|
387
|
+
"state",
|
|
388
|
+
"trigger",
|
|
389
|
+
"judgments"
|
|
390
|
+
],
|
|
120
391
|
"properties": {
|
|
121
|
-
"state": {
|
|
122
|
-
|
|
123
|
-
|
|
124
|
-
|
|
392
|
+
"state": {
|
|
393
|
+
"enum": [
|
|
394
|
+
"confirmed",
|
|
395
|
+
"tie_broken",
|
|
396
|
+
"unresolved"
|
|
397
|
+
]
|
|
398
|
+
},
|
|
399
|
+
"trigger": {
|
|
400
|
+
"enum": [
|
|
401
|
+
"ambiguous",
|
|
402
|
+
"contradictory",
|
|
403
|
+
"non_unanimous",
|
|
404
|
+
"ship_deciding"
|
|
405
|
+
]
|
|
406
|
+
},
|
|
407
|
+
"verdict": {
|
|
408
|
+
"enum": [
|
|
409
|
+
"PASS",
|
|
410
|
+
"FAIL"
|
|
411
|
+
]
|
|
412
|
+
},
|
|
413
|
+
"judgments": {
|
|
414
|
+
"type": "array",
|
|
415
|
+
"items": {
|
|
416
|
+
"$ref": "#/$defs/judgment"
|
|
417
|
+
}
|
|
418
|
+
}
|
|
125
419
|
}
|
|
126
420
|
},
|
|
127
|
-
"judge_history": {
|
|
421
|
+
"judge_history": {
|
|
422
|
+
"type": "array",
|
|
423
|
+
"minItems": 2,
|
|
424
|
+
"maxItems": 3,
|
|
425
|
+
"items": {
|
|
426
|
+
"$ref": "#/$defs/judgment"
|
|
427
|
+
}
|
|
428
|
+
}
|
|
128
429
|
}
|
|
129
430
|
}
|
|
130
431
|
}
|