betterwithage commited on
Commit
3d503cd
·
verified ·
1 Parent(s): b4c281f

FIX: honest proven count + Live System Map redesign (service-health dashboard)

Browse files
packages/a11oy-knowledge/src/knowledge.json CHANGED
@@ -1,10 +1,10 @@
1
  {
2
- "version": "1.0.0",
3
  "byline": "Lutar, Stephen P.",
4
  "orcid": "0009-0001-0110-4173",
5
  "email": "stephen@szlholdings.com",
6
  "org": "SZL Holdings",
7
- "generated_at": "2026-05-15T16:30:00Z",
8
  "axioms": [
9
  {
10
  "id": "A1",
@@ -92,9 +92,9 @@
92
  {
93
  "id": "TH_L1",
94
  "name": "\u039b_uniqueness",
95
- "statement": "The Lutar Invariant \u039b_k as weighted geometric mean with Egyptian unit-fraction weights is unique given the four axioms",
96
- "source_file": "lutar-lean/Lutar/Uniqueness.lean",
97
- "maturity": "proven",
98
  "citation": "https://doi.org/10.5281/zenodo.20053148"
99
  },
100
  {
@@ -129,7 +129,8 @@
129
  "source_line": 27,
130
  "latex": "\\mathcal{S} = \\langle R, A, E, \\Lambda, \\rho, W \\rangle",
131
  "context": "ith a doctrine-locked runtime** as a category-defining primitive for verifiable agency. We define the system as a tuple \\( \\mathcal{S} = \\langle R, A, E, \\Lambda, \\rho, W \\rangle \\) over an eight-regi",
132
- "source_id": "thesis_session"
 
133
  },
134
  {
135
  "id": "F0002",
@@ -137,7 +138,8 @@
137
  "source_line": 229,
138
  "latex": "\\mathtt{szl\\text{-}trust}",
139
  "context": "d system. - \\(A\\) \u2014 the set of **named actors**. Every actor in \\(A\\) carries a stable identity resolvable to a key in \\(\\mathtt{szl\\text{-}trust}\\). No edge in \\(E\\) may originate from or terminate ",
140
- "source_id": "thesis_session"
 
141
  },
142
  {
143
  "id": "F0003",
@@ -145,7 +147,8 @@
145
  "source_line": 231,
146
  "latex": "e \\in E",
147
  "context": "and resolvable \u2014 unidentified actors are structurally excluded. - \\(E\\) \u2014 the set of **receipt-bound edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\",
148
- "source_id": "thesis_session"
 
149
  },
150
  {
151
  "id": "F0004",
@@ -153,7 +156,8 @@
153
  "source_line": 231,
154
  "latex": "(a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)",
155
  "context": "ntified actors are structurally excluded. - \\(E\\) \u2014 the set of **receipt-bound edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(",
156
- "source_id": "thesis_session"
 
157
  },
158
  {
159
  "id": "F0005",
@@ -161,7 +165,8 @@
161
  "source_line": 231,
162
  "latex": "a_{\\text{src}} \\in A",
163
  "context": "d edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varep",
164
- "source_id": "thesis_session"
 
165
  },
166
  {
167
  "id": "F0006",
@@ -169,7 +174,8 @@
169
  "source_line": 231,
170
  "latex": "r_{\\text{src}}, r_{\\text{dst}} \\in R",
171
  "context": "E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varepsilon\\) is the receipt enve",
172
- "source_id": "thesis_session"
 
173
  },
174
  {
175
  "id": "F0007",
@@ -177,7 +183,8 @@
177
  "source_line": 231,
178
  "latex": "\\varepsilon",
179
  "context": "src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varepsilon\\) is the receipt envelope defined in \u00a73.3. No message may traverse",
180
- "source_id": "thesis_session"
 
181
  },
182
  {
183
  "id": "F0008",
@@ -185,7 +192,8 @@
185
  "source_line": 231,
186
  "latex": "\\varepsilon",
187
  "context": "repsilon\\) is the receipt envelope defined in \u00a73.3. No message may traverse a region boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Forma",
188
- "source_id": "thesis_session"
 
189
  },
190
  {
191
  "id": "F0009",
@@ -193,7 +201,8 @@
193
  "source_line": 233,
194
  "latex": "\\Lambda",
195
  "context": "ceipt envelope defined in \u00a73.3. No message may traverse a region boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,",
196
- "source_id": "thesis_session"
 
197
  },
198
  {
199
  "id": "F0010",
@@ -201,7 +210,8 @@
201
  "source_line": 233,
202
  "latex": "\\Lambda : [0,1]^k \\to \\{0,1\\}",
203
  "context": "boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,1]^k \\to \\{0,1\\}\\) for \\(k \\geq 9\\), defined as the conjunctive A",
204
- "source_id": "thesis_session"
 
205
  },
206
  {
207
  "id": "F0011",
@@ -209,7 +219,8 @@
209
  "source_line": 233,
210
  "latex": "k \\geq 9",
211
  "context": "varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,1]^k \\to \\{0,1\\}\\) for \\(k \\geq 9\\), defined as the conjunctive AND: \\[ \\Lambda(\\mathbf{x}) = 1 \\iff \\",
212
- "source_id": "thesis_session"
 
213
  },
214
  {
215
  "id": "F0012",
@@ -217,7 +228,8 @@
217
  "source_line": 239,
218
  "latex": "\\mathbf{x}",
219
  "context": "bilityHonesty}} \\geq 0.95 \\] The composability property states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wed",
220
- "source_id": "thesis_session"
 
221
  },
222
  {
223
  "id": "F0013",
@@ -225,7 +237,8 @@
225
  "source_line": 239,
226
  "latex": "\\mathbf{y}",
227
  "context": "q 0.95 \\] The composability property states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is",
228
- "source_id": "thesis_session"
 
229
  },
230
  {
231
  "id": "F0014",
@@ -233,7 +246,8 @@
233
  "source_line": 239,
234
  "latex": "\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})",
235
  "context": "rty states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is equivalent to \\(\\Lambda(\\mathbf{x})",
236
- "source_id": "thesis_session"
 
237
  },
238
  {
239
  "id": "F0015",
@@ -241,7 +255,8 @@
241
  "source_line": 239,
242
  "latex": "\\Lambda(\\mathbf{x}) \\wedge \\Lambda(\\mathbf{y})",
243
  "context": "ctors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is equivalent to \\(\\Lambda(\\mathbf{x}) \\wedge \\Lambda(\\mathbf{y})\\) \u2014 gate composition does not w",
244
- "source_id": "thesis_session"
 
245
  },
246
  {
247
  "id": "F0016",
@@ -249,7 +264,8 @@
249
  "source_line": 239,
250
  "latex": "\\Lambda",
251
  "context": "\u2014 gate composition does not weaken the invariant. The `lutar-lean` skeleton repository contains the Lean 4 statement of \\(\\Lambda\\) uniqueness: given the four axioms (A1 monotonicity, A2 homogeneity, ",
252
- "source_id": "thesis_session"
 
253
  },
254
  {
255
  "id": "F0017",
@@ -257,7 +273,8 @@
257
  "source_line": 239,
258
  "latex": "\\Lambda",
259
  "context": "ment of \\(\\Lambda\\) uniqueness: given the four axioms (A1 monotonicity, A2 homogeneity, A3 Egyptian-exact, A4 bounded), \\(\\Lambda\\) is the *unique* function satisfying them. The uniqueness theorem and",
260
- "source_id": "thesis_session"
 
261
  },
262
  {
263
  "id": "F0018",
@@ -265,7 +282,8 @@
265
  "source_line": 241,
266
  "latex": "\\rho(e)",
267
  "context": "arget is zero. - \\(\\rho\\) \u2014 the **dual-witness closure relation**. For any edge \\(e\\) carrying execution result \\(v\\), \\(\\rho(e)\\) holds iff two independent witnesses \\(w_1, w_2 \\in W\\) each produce ",
268
- "source_id": "thesis_session"
 
269
  },
270
  {
271
  "id": "F0019",
@@ -273,7 +291,8 @@
273
  "source_line": 241,
274
  "latex": "w_1, w_2 \\in W",
275
  "context": "closure relation**. For any edge \\(e\\) carrying execution result \\(v\\), \\(\\rho(e)\\) holds iff two independent witnesses \\(w_1, w_2 \\in W\\) each produce byte-identical output on the same input, and the",
276
- "source_id": "thesis_session"
 
277
  },
278
  {
279
  "id": "F0020",
@@ -281,7 +300,8 @@
281
  "source_line": 251,
282
  "latex": "\\mathcal{S}",
283
  "context": "uroboros` core + 4 `a11oy` covenant), while the full upstream runtime suite registers 218/218 passing tests. The tuple \\(\\mathcal{S}\\) is **doctrine-locked**: any runtime configuration in which (a) a",
284
- "source_id": "thesis_session"
 
285
  },
286
  {
287
  "id": "F0021",
@@ -289,7 +309,8 @@
289
  "source_line": 251,
290
  "latex": "\\Lambda",
291
  "context": "which (a) a region is unnamed, (b) an actor is not in \\(A\\), (c) an edge is produced without a receipt envelope, or (d) \\(\\Lambda\\) is evaluated below threshold does not constitute a valid instantiati",
292
- "source_id": "thesis_session"
 
293
  },
294
  {
295
  "id": "F0022",
@@ -297,7 +318,8 @@
297
  "source_line": 251,
298
  "latex": "\\mathcal{S}",
299
  "context": "ithout a receipt envelope, or (d) \\(\\Lambda\\) is evaluated below threshold does not constitute a valid instantiation of \\(\\mathcal{S}\\). --- ## The 8-Region Anatomy The eight canonical regions of \\",
300
- "source_id": "thesis_session"
 
301
  },
302
  {
303
  "id": "F0023",
@@ -305,7 +327,8 @@
305
  "source_line": 257,
306
  "latex": "\\mathcal{S}",
307
  "context": "of \\(R\\) are enumerated below. For each region the presentation gives: the repository identifier, its role in the tuple \\(\\mathcal{S}\\), its public interfaces, and its dependency relations within \\(E\\",
308
- "source_id": "thesis_session"
 
309
  },
310
  {
311
  "id": "F0024",
@@ -313,7 +336,8 @@
313
  "source_line": 265,
314
  "latex": "\\mathcal{S}",
315
  "context": "released 2026-05-13; concept DOI `10.5281/zenodo.19944926`, v11 paper DOI `10.5281/zenodo.20119582`) **Formal role in \\(\\mathcal{S}\\):** The Brain Stem is the runtime kernel that evaluates \\(\\Lambda\\",
316
- "source_id": "thesis_session"
 
317
  },
318
  {
319
  "id": "F0025",
@@ -321,7 +345,8 @@
321
  "source_line": 265,
322
  "latex": "\\Lambda",
323
  "context": "DOI `10.5281/zenodo.20119582`) **Formal role in \\(\\mathcal{S}\\):** The Brain Stem is the runtime kernel that evaluates \\(\\Lambda\\) and emits receipts. Every edge in \\(E\\) that crosses a region bounda",
324
- "source_id": "thesis_session"
 
325
  },
326
  {
327
  "id": "F0026",
@@ -329,7 +354,8 @@
329
  "source_line": 268,
330
  "latex": "\\Lambda",
331
  "context": "bda(axes: number[9|10]) \u2192 Receipt` \u2014 evaluates the conjunctive AND gate and returns a signed receipt with the composite \\(\\Lambda\\) score, Bekenstein budget, and dual-witness closure status. - `build_",
332
- "source_id": "thesis_session"
 
333
  },
334
  {
335
  "id": "F0027",
@@ -337,7 +363,8 @@
337
  "source_line": 274,
338
  "latex": "\\Lambda",
339
  "context": "chain root for third-party verification. **Dependencies:** - Depends on: `lutar-lean` (Skeleton) \u2014 the axiom set that \\(\\Lambda\\) is required to satisfy is formally stated there; the Brain Stem is th",
340
- "source_id": "thesis_session"
 
341
  },
342
  {
343
  "id": "F0028",
@@ -345,7 +372,8 @@
345
  "source_line": 277,
346
  "latex": "\\Lambda_9",
347
  "context": "utbound edge must call `evaluate_lambda` before the edge enters \\(E\\). The gate composition benchmark for v6.3.0 shows \\(\\Lambda_9\\) base p50 = 3.12 \u00b5s and composed p50 = 3.29 \u00b5s; with the Platform v",
348
- "source_id": "thesis_session"
 
349
  },
350
  {
351
  "id": "F0029",
@@ -353,7 +381,8 @@
353
  "source_line": 285,
354
  "latex": "\\mathcal{S}",
355
  "context": "a continuous supply-chain security posture. --- ### Heart \u2014 `a11oy` **Repo:** `szl-holdings/a11oy` **Formal role in \\(\\mathcal{S}\\):** The Heart is the covenant policy engine and the agent approva",
356
- "source_id": "thesis_session"
 
357
  },
358
  {
359
  "id": "F0030",
@@ -361,7 +390,8 @@
361
  "source_line": 285,
362
  "latex": "\\mathcal{S}",
363
  "context": "\\):** The Heart is the covenant policy engine and the agent approval queue. It governs the *authorization* dimension of \\(\\mathcal{S}\\): while the Brain Stem answers \"does this action score above \\(\\L",
364
- "source_id": "thesis_session"
 
365
  },
366
  {
367
  "id": "F0031",
@@ -369,7 +399,8 @@
369
  "source_line": 285,
370
  "latex": "\\Lambda",
371
  "context": "It governs the *authorization* dimension of \\(\\mathcal{S}\\): while the Brain Stem answers \"does this action score above \\(\\Lambda\\)?\", the Heart answers \"is this action permitted under the active cove",
372
- "source_id": "thesis_session"
 
373
  },
374
  {
375
  "id": "F0032",
@@ -377,7 +408,8 @@
377
  "source_line": 285,
378
  "latex": "r_{\\text{dst}} \\notin R",
379
  "context": "this action permitted under the active covenant?\". No action may exit the body graph \u2014 i.e., no edge in \\(E\\) may have \\(r_{\\text{dst}} \\notin R\\) \u2014 without a Heart pulse. The covenant is a named, ver",
380
- "source_id": "thesis_session"
 
381
  },
382
  {
383
  "id": "F0033",
@@ -385,39 +417,44 @@
385
  "source_line": 293,
386
  "latex": "\\Lambda",
387
  "context": "Stem's chain. **Dependencies:** - Depends on: `ouroboros` (Brain Stem) \u2014 covenant evaluation results are sealed with a \\(\\Lambda\\)-gated receipt; a covenant check that fails \\(\\Lambda\\) is itself a g",
388
- "source_id": "thesis_session"
 
389
  },
390
  {
391
  "id": "F0034",
392
  "source_file": "thesis.md",
393
  "source_line": 293,
394
  "latex": "\\Lambda",
395
- "context": "os` (Brain Stem) \u2014 covenant evaluation results are sealed with a \\(\\Lambda\\)-gated receipt; a covenant check that fails \\(\\Lambda\\) is itself a gate-level violation. - Depends on: `sentra` (Wires) \u2014 t",
396
- "source_id": "thesis_session"
 
397
  },
398
  {
399
  "id": "F0035",
400
  "source_file": "thesis.md",
401
  "source_line": 305,
402
  "latex": "\\mathcal{S}",
403
- "context": "but a verifiable, chain-linked artifact. --- ### Wires \u2014 `sentra` **Repo:** `szl-holdings/sentra` **Formal role in \\(\\mathcal{S}\\):** The Wires are the attribution trail \u2014 the afferent channel tha",
404
- "source_id": "thesis_session"
 
405
  },
406
  {
407
  "id": "F0036",
408
  "source_file": "thesis.md",
409
  "source_line": 305,
410
  "latex": "\\text{attr}: E \\to A",
411
- "context": "rent channel that carries signals inward and records *who observed what and when*. Formally, Wires maintain the mapping \\(\\text{attr}: E \\to A\\), ensuring that every edge in \\(E\\) is attributable to a",
412
- "source_id": "thesis_session"
 
413
  },
414
  {
415
  "id": "F0037",
416
  "source_file": "thesis.md",
417
  "source_line": 305,
418
  "latex": "\\mathcal{S}",
419
- "context": "he mapping \\(\\text{attr}: E \\to A\\), ensuring that every edge in \\(E\\) is attributable to a named actor. Without Wires, \\(\\mathcal{S}\\) degrades: edges carry receipts but not attributions, making the ",
420
- "source_id": "thesis_session"
 
421
  },
422
  {
423
  "id": "F0038",
@@ -425,31 +462,35 @@
425
  "source_line": 308,
426
  "latex": "a \\in A",
427
  "context": "egal-accountability sense. **Public interfaces:** - `observe(edge, actor_id) \u2192 AttributionRecord` \u2014 records that actor \\(a \\in A\\) produced or consumed edge \\(e\\). - `attribution_trail(region, time_r",
428
- "source_id": "thesis_session"
 
429
  },
430
  {
431
  "id": "F0039",
432
  "source_file": "thesis.md",
433
  "source_line": 325,
434
  "latex": "\\mathcal{S}",
435
- "context": "aft-morrow-sogomonian-exec-outcome-attest`. --- ### Spine \u2014 `amaru` **Repo:** `szl-holdings/amaru` **Formal role in \\(\\mathcal{S}\\):** The Spine is the append-only coordination and protocol bridge",
436
- "source_id": "thesis_session"
 
437
  },
438
  {
439
  "id": "F0040",
440
  "source_file": "thesis.md",
441
  "source_line": 325,
442
  "latex": "\\langle e_1, e_2, \\ldots, e_n \\rangle \\subseteq E",
443
- "context": "ordered, hash-verified record of every state transition across the body graph. Formally, `amaru` maintains the sequence \\(\\langle e_1, e_2, \\ldots, e_n \\rangle \\subseteq E\\) ordered by timestamp, with",
444
- "source_id": "thesis_session"
 
445
  },
446
  {
447
  "id": "F0041",
448
  "source_file": "thesis.md",
449
  "source_line": 339,
450
  "latex": "O(\\log n)",
451
- "context": "(identified in the runtime roadmap) would upgrade the Spine's linear hash-chain to a directed acyclic graph supporting \\(O(\\log n)\\) subset inclusion proofs \u2014 enabling privacy-preserving audits for re",
452
- "source_id": "thesis_session"
 
453
  },
454
  {
455
  "id": "F0042",
@@ -457,7 +498,8 @@
457
  "source_line": 347,
458
  "latex": "\\mathcal{S}",
459
  "context": "nce in the enterprise segment. --- ### Skeleton \u2014 `lutar-lean` **Repo:** `szl-holdings/lutar-lean` **Formal role in \\(\\mathcal{S}\\):** The Skeleton is the formal scaffold \u2014 the Lean 4 axioms and M",
460
- "source_id": "thesis_session"
 
461
  },
462
  {
463
  "id": "F0043",
@@ -465,7 +507,8 @@
465
  "source_line": 347,
466
  "latex": "\\{A1, A2, A3, A4\\}",
467
  "context": "es not execute at runtime; it is the *proof that the runtime is correct*. Formally, `lutar-lean` provides the axiom set \\(\\{A1, A2, A3, A4\\}\\) and the derived theorems (\u039b uniqueness, Bound theorem) th",
468
- "source_id": "thesis_session"
 
469
  },
470
  {
471
  "id": "F0044",
@@ -473,7 +516,8 @@
473
  "source_line": 347,
474
  "latex": "\\Lambda",
475
  "context": "A2, A3, A4\\}\\) and the derived theorems (\u039b uniqueness, Bound theorem) that constitute a machine-checked certificate for \\(\\Lambda\\). If the Skeleton's `sorry` count is zero, the gate the Brain Stem en",
476
- "source_id": "thesis_session"
 
477
  },
478
  {
479
  "id": "F0045",
@@ -481,7 +525,8 @@
481
  "source_line": 351,
482
  "latex": "\\Lambda",
483
  "context": "statements of A1 (monotonicity), A2 (homogeneity), A3 (Egyptian-exact), A4 (bounded). - `Uniqueness.lean` \u2014 Theorem 1: \\(\\Lambda\\) is the unique function satisfying A1\u2013A4; proof scaffold with tracked ",
484
- "source_id": "thesis_session"
 
485
  },
486
  {
487
  "id": "F0046",
@@ -489,7 +534,8 @@
489
  "source_line": 367,
490
  "latex": "\\mathcal{S}",
491
  "context": "*Repos:** `szl-holdings/counsel` (governance UI), `szl-holdings/terra` (dashboards and visualization) **Formal role in \\(\\mathcal{S}\\):** The Hands are the tooling and visualization surfaces \u2014 the co",
492
- "source_id": "thesis_session"
 
493
  },
494
  {
495
  "id": "F0047",
@@ -497,7 +543,8 @@
497
  "source_line": 371,
498
  "latex": "\\Lambda",
499
  "context": "as an interactive SVG, streaming live receipt counts via SSE from `/api/chain/stream`; node colors reflect the current \\(\\Lambda\\) score band (green \u2265 0.95, amber 0.90\u20130.95, red < 0.90). The planned \"",
500
- "source_id": "thesis_session"
 
501
  },
502
  {
503
  "id": "F0048",
@@ -505,7 +552,8 @@
505
  "source_line": 386,
506
  "latex": "\\mathcal{S}",
507
  "context": "*is* the system. --- ### Full Body \u2014 `ouroboros-thesis` **Repo:** `szl-holdings/ouroboros-thesis` **Formal role in \\(\\mathcal{S}\\):** The Full Body is the public-record thesis \u2014 the DOI-pinned, ve",
508
- "source_id": "thesis_session"
 
509
  },
510
  {
511
  "id": "F0049",
@@ -513,7 +561,8 @@
513
  "source_line": 386,
514
  "latex": "\\mathcal{S}",
515
  "context": "l Body is the public-record thesis \u2014 the DOI-pinned, versioned document that constitutes the canonical specification of \\(\\mathcal{S}\\). Formally, `ouroboros-thesis` defines the normative description ",
516
- "source_id": "thesis_session"
 
517
  },
518
  {
519
  "id": "F0050",
@@ -521,7 +570,8 @@
521
  "source_line": 405,
522
  "latex": "\\mathcal{S}",
523
  "context": "d identity anchoring), `szl-holdings/szl-cookbook` (reference implementations / developer onboarding) **Formal role in \\(\\mathcal{S}\\):** The Vessels and Chakras collectively form the trust mesh and ",
524
- "source_id": "thesis_session"
 
525
  },
526
  {
527
  "id": "F0051",
@@ -529,7 +579,8 @@
529
  "source_line": 421,
530
  "latex": "\\varepsilon",
531
  "context": "eue under the covenant pack schema. --- ## Cross-Region Contracts Every edge in \\(E\\) carries a **receipt envelope** \\(\\varepsilon\\). The envelope is a typed, signed, content-addressed record that ",
532
- "source_id": "thesis_session"
 
533
  },
534
  {
535
  "id": "F0052",
@@ -537,7 +588,8 @@
537
  "source_line": 421,
538
  "latex": "\\Lambda",
539
  "context": "es a **receipt envelope** \\(\\varepsilon\\). The envelope is a typed, signed, content-addressed record that provides: the \\(\\Lambda\\) score vector, the dual-witness closure status (\\(\\rho\\)), the actor ",
540
- "source_id": "thesis_session"
 
541
  },
542
  {
543
  "id": "F0053",
@@ -545,7 +597,8 @@
545
  "source_line": 479,
546
  "latex": "\\Lambda",
547
  "context": "_lambda(axes) \u2192 Receipt` \u2014 any MCP-compatible client (Claude Desktop, Cursor, enterprise agent frameworks) can call the \\(\\Lambda\\) gate as a typed tool and receive a signed receipt in the tool respon",
548
- "source_id": "thesis_session"
 
549
  },
550
  {
551
  "id": "F0054",
@@ -553,7 +606,8 @@
553
  "source_line": 507,
554
  "latex": "\\mathcal{S}",
555
  "context": "the 8-Region Model Structurally Surpasses the Leaders Each leading framework or protocol is a partial instantiation of \\(\\mathcal{S}\\). The gap is structural: the missing region is not a feature that",
556
- "source_id": "thesis_session"
 
557
  },
558
  {
559
  "id": "F0055",
@@ -561,7 +615,8 @@
561
  "source_line": 515,
562
  "latex": "\\Lambda_9",
563
  "context": "l engineering pattern, but skills are *files*, not services with receipts. A Brain Stem can issue a decision that fails \\(\\Lambda_9\\) moralGrounding; in the Managed Agents architecture there is no mec",
564
- "source_id": "thesis_session"
 
565
  },
566
  {
567
  "id": "F0056",
@@ -569,7 +624,8 @@
569
  "source_line": 515,
570
  "latex": "\\mathcal{S}",
571
  "context": "fails \\(\\Lambda_9\\) moralGrounding; in the Managed Agents architecture there is no mechanism to detect or block it. In \\(\\mathcal{S}\\), that decision never exits the Brain Stem. **Mastra** (22K+ GitH",
572
- "source_id": "thesis_session"
 
573
  },
574
  {
575
  "id": "F0057",
@@ -577,7 +633,8 @@
577
  "source_line": 517,
578
  "latex": "\\Lambda",
579
  "context": "ource agent framework in the TypeScript ecosystem. Mastra has no Skeleton: there are no Lean 4 proofs. It has no formal \\(\\Lambda\\) gate \u2014 behavioral constraints are implemented as runtime checks with",
580
- "source_id": "thesis_session"
 
581
  },
582
  {
583
  "id": "F0058",
@@ -585,7 +642,8 @@
585
  "source_line": 567,
586
  "latex": "\\lambda_1",
587
  "context": "l(\\lambda_1(c),\\, \\lambda_2(c),\\, \\ldots,\\, \\lambda_9(c)\\bigr) \\in [0,1]^9 \\] The nine axes are defined as follows. **\\(\\lambda_1\\): moralGrounding.** Measures the degree to which a proposed action ",
588
- "source_id": "thesis_session"
 
589
  },
590
  {
591
  "id": "F0059",
@@ -593,7 +651,8 @@
593
  "source_line": 567,
594
  "latex": "\\lambda_1",
595
  "context": "nce policies, and principal hierarchies that the operator has encoded in the agent's governing covenant. Operationally, \\(\\lambda_1\\) is the normalized cosine similarity between the action's intent em",
596
- "source_id": "thesis_session"
 
597
  },
598
  {
599
  "id": "F0060",
@@ -601,7 +660,8 @@
601
  "source_line": 567,
602
  "latex": "[0,1]",
603
  "context": "mbedding and a reference \"moral anchor\" embedding, averaged over the operator's registered covenant clauses, clamped to \\([0,1]\\). The floor constraint \\(\\lambda_1 \\geq 0.95\\) is a hard asymptote: an ",
604
- "source_id": "thesis_session"
 
605
  },
606
  {
607
  "id": "F0061",
@@ -609,7 +669,8 @@
609
  "source_line": 567,
610
  "latex": "\\lambda_1 \\geq 0.95",
611
  "context": "anchor\" embedding, averaged over the operator's registered covenant clauses, clamped to \\([0,1]\\). The floor constraint \\(\\lambda_1 \\geq 0.95\\) is a hard asymptote: an agent that is even marginally mo",
612
- "source_id": "thesis_session"
 
613
  },
614
  {
615
  "id": "F0062",
@@ -617,7 +678,8 @@
617
  "source_line": 569,
618
  "latex": "\\lambda_2",
619
  "context": "even marginally morally misaligned fails the gate irrespective of how perfectly calibrated the other eight axes are. **\\(\\lambda_2\\): measurabilityHonesty.** Measures whether an action's declared eff",
620
- "source_id": "thesis_session"
 
621
  },
622
  {
623
  "id": "F0063",
@@ -625,7 +687,8 @@
625
  "source_line": 571,
626
  "latex": "\\lambda_3",
627
  "context": "ine clause \"no hallucinations no bandaids; test test test\" by making measurement-honesty a prerequisite for passage. **\\(\\lambda_3\\): epistemicHumility.** Scores the agent's acknowledgment of its own",
628
- "source_id": "thesis_session"
 
629
  },
630
  {
631
  "id": "F0064",
@@ -633,7 +696,8 @@
633
  "source_line": 571,
634
  "latex": "\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]",
635
  "context": "sparse scores low on this axis. The scoring function penalizes unjustified confidence using a calibration-error analog: \\(\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{c",
636
- "source_id": "thesis_session"
 
637
  },
638
  {
639
  "id": "F0065",
@@ -641,7 +705,8 @@
641
  "source_line": 571,
642
  "latex": "\\text{conf}(c)",
643
  "context": "ied confidence using a calibration-error analog: \\(\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{conf}(c)\\) is the agent's stated confidence and \\(\\text{acc}(c)\\) is the",
644
- "source_id": "thesis_session"
 
645
  },
646
  {
647
  "id": "F0066",
@@ -649,7 +714,8 @@
649
  "source_line": 571,
650
  "latex": "\\text{acc}(c)",
651
  "context": "da_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{conf}(c)\\) is the agent's stated confidence and \\(\\text{acc}(c)\\) is the empirically measured accuracy over a calibration set. ",
652
- "source_id": "thesis_session"
 
653
  },
654
  {
655
  "id": "F0067",
@@ -657,7 +723,8 @@
657
  "source_line": 573,
658
  "latex": "\\lambda_4",
659
  "context": "is the agent's stated confidence and \\(\\text{acc}(c)\\) is the empirically measured accuracy over a calibration set. **\\(\\lambda_4\\): counterfactualAwareness.** Measures whether the agent has consider",
660
- "source_id": "thesis_session"
 
661
  },
662
  {
663
  "id": "F0068",
@@ -665,7 +732,8 @@
665
  "source_line": 575,
666
  "latex": "\\lambda_5",
667
  "context": "res 0.0 and a uniformly distributed consequence distribution over the operator-defined consequence space scores 1.0. **\\(\\lambda_5\\): temporalConsistency.** Measures the stability of the gate verdict",
668
- "source_id": "thesis_session"
 
669
  },
670
  {
671
  "id": "F0069",
@@ -673,7 +741,8 @@
673
  "source_line": 575,
674
  "latex": "t + \\Delta",
675
  "context": "Measures the stability of the gate verdict under repeated evaluation on the same input at two different times \\(t\\) and \\(t + \\Delta\\). Let \\(v_t\\) and \\(v_{t+\\Delta}\\) denote the \u039b\u2089 composite scores ",
676
- "source_id": "thesis_session"
 
677
  },
678
  {
679
  "id": "F0070",
@@ -681,7 +750,8 @@
681
  "source_line": 575,
682
  "latex": "v_{t+\\Delta}",
683
  "context": "te verdict under repeated evaluation on the same input at two different times \\(t\\) and \\(t + \\Delta\\). Let \\(v_t\\) and \\(v_{t+\\Delta}\\) denote the \u039b\u2089 composite scores at the two evaluation times. The",
684
- "source_id": "thesis_session"
 
685
  },
686
  {
687
  "id": "F0071",
@@ -689,7 +759,8 @@
689
  "source_line": 581,
690
  "latex": "\\lambda_5 = 1.0",
691
  "context": "Then: \\[ \\lambda_5 = \\max\\!\\Bigl(0,\\; 1 - 4\\,\\bigl(v_t - v_{t+\\Delta}\\bigr)^2\\Bigr) \\] A zero-drift evaluation scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 =",
692
- "source_id": "thesis_session"
 
693
  },
694
  {
695
  "id": "F0072",
@@ -697,7 +768,8 @@
697
  "source_line": 581,
698
  "latex": "\\lambda_5 = 0.99",
699
  "context": "ta}\\bigr)^2\\Bigr) \\] A zero-drift evaluation scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 = 0.99\\). A drift of 0.25 yields \\(\\lambda_5 = 0.75\\), below the \u2265 0",
700
- "source_id": "thesis_session"
 
701
  },
702
  {
703
  "id": "F0073",
@@ -705,7 +777,8 @@
705
  "source_line": 581,
706
  "latex": "\\lambda_5 = 0.75",
707
  "context": "scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 = 0.99\\). A drift of 0.25 yields \\(\\lambda_5 = 0.75\\), below the \u2265 0.90 conjunctive floor. This axis operationaliz",
708
- "source_id": "thesis_session"
 
709
  },
710
  {
711
  "id": "F0074",
@@ -713,7 +786,10 @@
713
  "source_line": 583,
714
  "latex": "\\lambda_6",
715
  "context": "-identical replay guarantee: a system that cannot reproduce its own gate verdict is not operating deterministically. **\\(\\lambda_6\\): evidenceProvenance.** Measures whether every empirical claim embe",
716
- "source_id": "thesis_session"
 
 
 
717
  },
718
  {
719
  "id": "F0075",
@@ -721,7 +797,8 @@
721
  "source_line": 585,
722
  "latex": "\\lambda_7",
723
  "context": "ertions score at most 0.50. The scoring function is the fraction of claim tokens for which provenance is resolvable. **\\(\\lambda_7\\): actorIdentity.** Measures the definiteness of the acting agent's ",
724
- "source_id": "thesis_session"
 
725
  },
726
  {
727
  "id": "F0076",
@@ -729,7 +806,8 @@
729
  "source_line": 587,
730
  "latex": "\\lambda_8",
731
  "context": "ting under delegated authority \u2014 the score decays as a function of delegation depth to penalize opaque proxy chains. **\\(\\lambda_8\\): axiomConsistency.** Measures whether the proposed action is inter",
732
- "source_id": "thesis_session"
 
733
  },
734
  {
735
  "id": "F0077",
@@ -737,7 +815,8 @@
737
  "source_line": 589,
738
  "latex": "\\lambda_9",
739
  "context": "Lean 4 formalization: it enforces, at runtime, the constraints that are statically verified at theorem-proving time. **\\(\\lambda_9\\): coherence.** Measures the multi-step logical coherence of the age",
740
- "source_id": "thesis_session"
 
741
  },
742
  {
743
  "id": "F0078",
@@ -745,7 +824,8 @@
745
  "source_line": 589,
746
  "latex": "A_1, A_2, \\ldots, A_k",
747
  "context": "-step logical coherence of the agent's plan across the action sequence, not just for the current step in isolation. Let \\(A_1, A_2, \\ldots, A_k\\) denote the \\(k\\) preceding actions in the current sess",
748
- "source_id": "thesis_session"
 
749
  },
750
  {
751
  "id": "F0079",
@@ -753,7 +833,8 @@
753
  "source_line": 589,
754
  "latex": "(A_i, A_{i+1})",
755
  "context": "e the \\(k\\) preceding actions in the current session. The coherence score is the proportion of consecutive action-pairs \\((A_i, A_{i+1})\\) for which the precondition of \\(A_{i+1}\\) is satisfied by the",
756
- "source_id": "thesis_session"
 
757
  },
758
  {
759
  "id": "F0080",
@@ -761,7 +842,8 @@
761
  "source_line": 589,
762
  "latex": "A_{i+1}",
763
  "context": "ion. The coherence score is the proportion of consecutive action-pairs \\((A_i, A_{i+1})\\) for which the precondition of \\(A_{i+1}\\) is satisfied by the postcondition of \\(A_i\\), under the operator's p",
764
- "source_id": "thesis_session"
 
765
  },
766
  {
767
  "id": "F0081",
@@ -769,7 +851,8 @@
769
  "source_line": 589,
770
  "latex": "\\lambda_9 = 1.0",
771
  "context": "ied by the postcondition of \\(A_i\\), under the operator's precondition/postcondition schema. For the base case \\(k=0\\), \\(\\lambda_9 = 1.0\\). ### The conjunctive gate condition The \u039b\u2089 gate passes if ",
772
- "source_id": "thesis_session"
 
773
  },
774
  {
775
  "id": "F0082",
@@ -777,7 +860,8 @@
777
  "source_line": 599,
778
  "latex": "\\lambda_1 = 0.50",
779
  "context": "for the following reason. A single composite score \u2014 even a geometric mean \u2014 can mask localized failures. An agent with \\(\\lambda_1 = 0.50\\) (severely morally misaligned) and all remaining axes at \\(1",
780
- "source_id": "thesis_session"
 
781
  },
782
  {
783
  "id": "F0083",
@@ -785,7 +869,8 @@
785
  "source_line": 599,
786
  "latex": "\\prod_{i}^{1/9} = 0.50^{1/9} \\approx 0.926",
787
  "context": "with \\(\\lambda_1 = 0.50\\) (severely morally misaligned) and all remaining axes at \\(1.0\\) achieves a geometric mean of \\(\\prod_{i}^{1/9} = 0.50^{1/9} \\approx 0.926\\), which would pass a \u2265 0.90 single-",
788
- "source_id": "thesis_session"
 
789
  },
790
  {
791
  "id": "F0084",
@@ -793,7 +878,8 @@
793
  "source_line": 599,
794
  "latex": "\\lambda_1",
795
  "context": "gle-score gate. The conjunctive AND structure prevents this: every axis is a blocking veto. The two elevated floors for \\(\\lambda_1\\) and \\(\\lambda_2\\) add a second layer of asymmetry \u2014 these are the ",
796
- "source_id": "thesis_session"
 
797
  },
798
  {
799
  "id": "F0085",
@@ -801,7 +887,8 @@
801
  "source_line": 599,
802
  "latex": "\\lambda_2",
803
  "context": "e conjunctive AND structure prevents this: every axis is a blocking veto. The two elevated floors for \\(\\lambda_1\\) and \\(\\lambda_2\\) add a second layer of asymmetry \u2014 these are the axes most directly",
804
- "source_id": "thesis_session"
 
805
  },
806
  {
807
  "id": "F0086",
@@ -809,7 +896,8 @@
809
  "source_line": 605,
810
  "latex": "m \\in \\{0,1\\}^9",
811
  "context": "to the receipt structure. Rather than publishing the raw nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evalua",
812
- "source_id": "thesis_session"
 
813
  },
814
  {
815
  "id": "F0087",
@@ -817,7 +905,8 @@
817
  "source_line": 605,
818
  "latex": "m_i = 1",
819
  "context": "her than publishing the raw nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evaluated and passed its floor. The",
820
- "source_id": "thesis_session"
 
821
  },
822
  {
823
  "id": "F0088",
@@ -825,7 +914,8 @@
825
  "source_line": 605,
826
  "latex": "\\lambda_i",
827
  "context": "nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evaluated and passed its floor. The raw scores are withheld fro",
828
- "source_id": "thesis_session"
 
829
  },
830
  {
831
  "id": "F0089",
@@ -833,7 +923,8 @@
833
  "source_line": 611,
834
  "latex": "\\theta_i = 0.95",
835
  "context": "ity profile of the agent. Formally, the mask is computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The",
836
- "source_id": "thesis_session"
 
837
  },
838
  {
839
  "id": "F0090",
@@ -841,7 +932,8 @@
841
  "source_line": 611,
842
  "latex": "i \\in \\{1,2\\}",
843
  "context": ". Formally, the mask is computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_",
844
- "source_id": "thesis_session"
 
845
  },
846
  {
847
  "id": "F0091",
@@ -849,7 +941,8 @@
849
  "source_line": 611,
850
  "latex": "\\theta_i = 0.90",
851
  "context": "s computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_i m_i = 9\\) (or 10 und",
852
- "source_id": "thesis_session"
 
853
  },
854
  {
855
  "id": "F0092",
@@ -857,7 +950,8 @@
857
  "source_line": 611,
858
  "latex": "\\sum_i m_i = 9",
859
  "context": "eq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_i m_i = 9\\) (or 10 under \u039b\u2081\u2080). The mask is committed via SHA-256 and incl",
860
- "source_id": "thesis_session"
 
861
  },
862
  {
863
  "id": "F0093",
@@ -865,7 +959,8 @@
865
  "source_line": 627,
866
  "latex": "\\textit{parent\\_hash}",
867
  "context": "\\textit{timestamp},\\; \\vec{\\lambda},\\; \\rho\\_\\textit{witness\\_set},\\; \\textit{signature}\\bigr) \\] The fields are: - **\\(\\textit{parent\\_hash}\\)**: The SHA-256 digest of receipt \\(r_{i-1}\\). For the ",
868
- "source_id": "thesis_session"
 
869
  },
870
  {
871
  "id": "F0094",
@@ -873,7 +968,8 @@
873
  "source_line": 627,
874
  "latex": "r_{i-1}",
875
  "context": "s\\_set},\\; \\textit{signature}\\bigr) \\] The fields are: - **\\(\\textit{parent\\_hash}\\)**: The SHA-256 digest of receipt \\(r_{i-1}\\). For the genesis receipt, this is the SHA-256 of a protocol-specifie",
876
- "source_id": "thesis_session"
 
877
  },
878
  {
879
  "id": "F0095",
@@ -881,7 +977,8 @@
881
  "source_line": 628,
882
  "latex": "\\textit{content\\_digest}",
883
  "context": "this is the SHA-256 of a protocol-specified null seed. This field creates the backward-pointing link of the chain. - **\\(\\textit{content\\_digest}\\)**: The SHA-256 of the canonical JSON serialization o",
884
- "source_id": "thesis_session"
 
885
  },
886
  {
887
  "id": "F0096",
@@ -889,7 +986,8 @@
889
  "source_line": 629,
890
  "latex": "\\textit{actor}",
891
  "context": "fore any side-effectful execution. This binds the gate verdict irrevocably to the specific input that triggered it. - **\\(\\textit{actor}\\)**: The identifier of the acting agent as registered in the pr",
892
- "source_id": "thesis_session"
 
893
  },
894
  {
895
  "id": "F0097",
@@ -897,7 +995,8 @@
897
  "source_line": 629,
898
  "latex": "\\lambda_7",
899
  "context": "t. - **\\(\\textit{actor}\\)**: The identifier of the acting agent as registered in the principal registry. Corresponds to \\(\\lambda_7\\) (actorIdentity). - **\\(\\textit{timestamp}\\)**: A monotonic timesta",
900
- "source_id": "thesis_session"
 
901
  },
902
  {
903
  "id": "F0098",
@@ -905,7 +1004,8 @@
905
  "source_line": 630,
906
  "latex": "\\textit{timestamp}",
907
  "context": "entifier of the acting agent as registered in the principal registry. Corresponds to \\(\\lambda_7\\) (actorIdentity). - **\\(\\textit{timestamp}\\)**: A monotonic timestamp in milliseconds since the Unix e",
908
- "source_id": "thesis_session"
 
909
  },
910
  {
911
  "id": "F0099",
@@ -913,7 +1013,8 @@
913
  "source_line": 631,
914
  "latex": "\\vec{\\lambda}",
915
  "context": ": A monotonic timestamp in milliseconds since the Unix epoch, drawn from a pinned, non-forgeable source (see \u00a74.6). - **\\(\\vec{\\lambda}\\)**: The full nine-dimensional \u039b vector, or the `lambda9_mask` b",
916
- "source_id": "thesis_session"
 
917
  },
918
  {
919
  "id": "F0100",
@@ -921,7 +1022,8 @@
921
  "source_line": 632,
922
  "latex": "\\rho\\_\\textit{witness\\_set}",
923
  "context": "vec{\\lambda}\\)**: The full nine-dimensional \u039b vector, or the `lambda9_mask` bitfield under the \u039b\u2081\u2080 privacy variant. - **\\(\\rho\\_\\textit{witness\\_set}\\)**: The set of co-witnesses whose signatures are ",
924
- "source_id": "thesis_session"
 
925
  }
926
  ],
927
  "definitions": [],
@@ -1583,5 +1685,1067 @@
1583
  "10.5281/zenodo.20053163",
1584
  "10.5281/zenodo.20119582",
1585
  "10.5281/zenodo.20162352"
1586
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1587
  }
 
1
  {
2
+ "version": "5.1.0",
3
  "byline": "Lutar, Stephen P.",
4
  "orcid": "0009-0001-0110-4173",
5
  "email": "stephen@szlholdings.com",
6
  "org": "SZL Holdings",
7
+ "generated_at": "2026-06-06T06:00:00Z",
8
  "axioms": [
9
  {
10
  "id": "A1",
 
92
  {
93
  "id": "TH_L1",
94
  "name": "\u039b_uniqueness",
95
+ "statement": "Conjecture 1: the Lutar Invariant \u039b_k (weighted geometric mean with Egyptian unit-fraction weights) is the unique aggregator satisfying axioms A1-A5. NOT a theorem: unconditional uniqueness is FALSE under A1-A5 (machine-checked counterexample maxAgg_ne_Lambda; max-aggregator satisfies A1-A5 yet differs from \u039b at (4,1)). The conditional theorem lambda_unique_of_factors (uniqueness GIVEN factorization \u03a6 x = \u220f x_i^\u03b1_i) IS fully proved; unconditional uniqueness closes only under a declared bisymmetry axiom A6 (Kolmogorov-Nagumo-Aczel).",
96
+ "source_file": "lutar-lean/Lutar/Round13/Lambda_Uniqueness.lean",
97
+ "maturity": "conjectured",
98
  "citation": "https://doi.org/10.5281/zenodo.20053148"
99
  },
100
  {
 
129
  "source_line": 27,
130
  "latex": "\\mathcal{S} = \\langle R, A, E, \\Lambda, \\rho, W \\rangle",
131
  "context": "ith a doctrine-locked runtime** as a category-defining primitive for verifiable agency. We define the system as a tuple \\( \\mathcal{S} = \\langle R, A, E, \\Lambda, \\rho, W \\rangle \\) over an eight-regi",
132
+ "source_id": "thesis_session",
133
+ "maturity": "defined"
134
  },
135
  {
136
  "id": "F0002",
 
138
  "source_line": 229,
139
  "latex": "\\mathtt{szl\\text{-}trust}",
140
  "context": "d system. - \\(A\\) \u2014 the set of **named actors**. Every actor in \\(A\\) carries a stable identity resolvable to a key in \\(\\mathtt{szl\\text{-}trust}\\). No edge in \\(E\\) may originate from or terminate ",
141
+ "source_id": "thesis_session",
142
+ "maturity": "defined"
143
  },
144
  {
145
  "id": "F0003",
 
147
  "source_line": 231,
148
  "latex": "e \\in E",
149
  "context": "and resolvable \u2014 unidentified actors are structurally excluded. - \\(E\\) \u2014 the set of **receipt-bound edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\",
150
+ "source_id": "thesis_session",
151
+ "maturity": "defined"
152
  },
153
  {
154
  "id": "F0004",
 
156
  "source_line": 231,
157
  "latex": "(a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)",
158
  "context": "ntified actors are structurally excluded. - \\(E\\) \u2014 the set of **receipt-bound edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(",
159
+ "source_id": "thesis_session",
160
+ "maturity": "defined"
161
  },
162
  {
163
  "id": "F0005",
 
165
  "source_line": 231,
166
  "latex": "a_{\\text{src}} \\in A",
167
  "context": "d edges**. An edge \\(e \\in E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varep",
168
+ "source_id": "thesis_session",
169
+ "maturity": "defined"
170
  },
171
  {
172
  "id": "F0006",
 
174
  "source_line": 231,
175
  "latex": "r_{\\text{src}}, r_{\\text{dst}} \\in R",
176
  "context": "E\\) is a tuple \\((a_{\\text{src}},\\; r_{\\text{src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varepsilon\\) is the receipt enve",
177
+ "source_id": "thesis_session",
178
+ "maturity": "defined"
179
  },
180
  {
181
  "id": "F0007",
 
183
  "source_line": 231,
184
  "latex": "\\varepsilon",
185
  "context": "src}},\\; r_{\\text{dst}},\\; \\varepsilon)\\) where \\(a_{\\text{src}} \\in A\\), \\(r_{\\text{src}}, r_{\\text{dst}} \\in R\\), and \\(\\varepsilon\\) is the receipt envelope defined in \u00a73.3. No message may traverse",
186
+ "source_id": "thesis_session",
187
+ "maturity": "defined"
188
  },
189
  {
190
  "id": "F0008",
 
192
  "source_line": 231,
193
  "latex": "\\varepsilon",
194
  "context": "repsilon\\) is the receipt envelope defined in \u00a73.3. No message may traverse a region boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Forma",
195
+ "source_id": "thesis_session",
196
+ "maturity": "defined"
197
  },
198
  {
199
  "id": "F0009",
 
201
  "source_line": 233,
202
  "latex": "\\Lambda",
203
  "context": "ceipt envelope defined in \u00a73.3. No message may traverse a region boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,",
204
+ "source_id": "thesis_session",
205
+ "maturity": "defined"
206
  },
207
  {
208
  "id": "F0010",
 
210
  "source_line": 233,
211
  "latex": "\\Lambda : [0,1]^k \\to \\{0,1\\}",
212
  "context": "boundary unless it carries a valid \\(\\varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,1]^k \\to \\{0,1\\}\\) for \\(k \\geq 9\\), defined as the conjunctive A",
213
+ "source_id": "thesis_session",
214
+ "maturity": "defined"
215
  },
216
  {
217
  "id": "F0011",
 
219
  "source_line": 233,
220
  "latex": "k \\geq 9",
221
  "context": "varepsilon\\). - \\(\\Lambda\\) \u2014 the **composable axis-gating function**. Formally, \\(\\Lambda : [0,1]^k \\to \\{0,1\\}\\) for \\(k \\geq 9\\), defined as the conjunctive AND: \\[ \\Lambda(\\mathbf{x}) = 1 \\iff \\",
222
+ "source_id": "thesis_session",
223
+ "maturity": "defined"
224
  },
225
  {
226
  "id": "F0012",
 
228
  "source_line": 239,
229
  "latex": "\\mathbf{x}",
230
  "context": "bilityHonesty}} \\geq 0.95 \\] The composability property states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wed",
231
+ "source_id": "thesis_session",
232
+ "maturity": "defined"
233
  },
234
  {
235
  "id": "F0013",
 
237
  "source_line": 239,
238
  "latex": "\\mathbf{y}",
239
  "context": "q 0.95 \\] The composability property states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is",
240
+ "source_id": "thesis_session",
241
+ "maturity": "defined"
242
  },
243
  {
244
  "id": "F0014",
 
246
  "source_line": 239,
247
  "latex": "\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})",
248
  "context": "rty states that for any two independently evaluated axis vectors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is equivalent to \\(\\Lambda(\\mathbf{x})",
249
+ "source_id": "thesis_session",
250
+ "maturity": "defined"
251
  },
252
  {
253
  "id": "F0015",
 
255
  "source_line": 239,
256
  "latex": "\\Lambda(\\mathbf{x}) \\wedge \\Lambda(\\mathbf{y})",
257
  "context": "ctors \\(\\mathbf{x}\\) and \\(\\mathbf{y}\\), their composed gate \\(\\Lambda(\\mathbf{x} \\wedge \\mathbf{y})\\) is equivalent to \\(\\Lambda(\\mathbf{x}) \\wedge \\Lambda(\\mathbf{y})\\) \u2014 gate composition does not w",
258
+ "source_id": "thesis_session",
259
+ "maturity": "defined"
260
  },
261
  {
262
  "id": "F0016",
 
264
  "source_line": 239,
265
  "latex": "\\Lambda",
266
  "context": "\u2014 gate composition does not weaken the invariant. The `lutar-lean` skeleton repository contains the Lean 4 statement of \\(\\Lambda\\) uniqueness: given the four axioms (A1 monotonicity, A2 homogeneity, ",
267
+ "source_id": "thesis_session",
268
+ "maturity": "conjectured"
269
  },
270
  {
271
  "id": "F0017",
 
273
  "source_line": 239,
274
  "latex": "\\Lambda",
275
  "context": "ment of \\(\\Lambda\\) uniqueness: given the four axioms (A1 monotonicity, A2 homogeneity, A3 Egyptian-exact, A4 bounded), \\(\\Lambda\\) is the *unique* function satisfying them. The uniqueness theorem and",
276
+ "source_id": "thesis_session",
277
+ "maturity": "conjectured"
278
  },
279
  {
280
  "id": "F0018",
 
282
  "source_line": 241,
283
  "latex": "\\rho(e)",
284
  "context": "arget is zero. - \\(\\rho\\) \u2014 the **dual-witness closure relation**. For any edge \\(e\\) carrying execution result \\(v\\), \\(\\rho(e)\\) holds iff two independent witnesses \\(w_1, w_2 \\in W\\) each produce ",
285
+ "source_id": "thesis_session",
286
+ "maturity": "defined"
287
  },
288
  {
289
  "id": "F0019",
 
291
  "source_line": 241,
292
  "latex": "w_1, w_2 \\in W",
293
  "context": "closure relation**. For any edge \\(e\\) carrying execution result \\(v\\), \\(\\rho(e)\\) holds iff two independent witnesses \\(w_1, w_2 \\in W\\) each produce byte-identical output on the same input, and the",
294
+ "source_id": "thesis_session",
295
+ "maturity": "defined"
296
  },
297
  {
298
  "id": "F0020",
 
300
  "source_line": 251,
301
  "latex": "\\mathcal{S}",
302
  "context": "uroboros` core + 4 `a11oy` covenant), while the full upstream runtime suite registers 218/218 passing tests. The tuple \\(\\mathcal{S}\\) is **doctrine-locked**: any runtime configuration in which (a) a",
303
+ "source_id": "thesis_session",
304
+ "maturity": "defined"
305
  },
306
  {
307
  "id": "F0021",
 
309
  "source_line": 251,
310
  "latex": "\\Lambda",
311
  "context": "which (a) a region is unnamed, (b) an actor is not in \\(A\\), (c) an edge is produced without a receipt envelope, or (d) \\(\\Lambda\\) is evaluated below threshold does not constitute a valid instantiati",
312
+ "source_id": "thesis_session",
313
+ "maturity": "defined"
314
  },
315
  {
316
  "id": "F0022",
 
318
  "source_line": 251,
319
  "latex": "\\mathcal{S}",
320
  "context": "ithout a receipt envelope, or (d) \\(\\Lambda\\) is evaluated below threshold does not constitute a valid instantiation of \\(\\mathcal{S}\\). --- ## The 8-Region Anatomy The eight canonical regions of \\",
321
+ "source_id": "thesis_session",
322
+ "maturity": "defined"
323
  },
324
  {
325
  "id": "F0023",
 
327
  "source_line": 257,
328
  "latex": "\\mathcal{S}",
329
  "context": "of \\(R\\) are enumerated below. For each region the presentation gives: the repository identifier, its role in the tuple \\(\\mathcal{S}\\), its public interfaces, and its dependency relations within \\(E\\",
330
+ "source_id": "thesis_session",
331
+ "maturity": "defined"
332
  },
333
  {
334
  "id": "F0024",
 
336
  "source_line": 265,
337
  "latex": "\\mathcal{S}",
338
  "context": "released 2026-05-13; concept DOI `10.5281/zenodo.19944926`, v11 paper DOI `10.5281/zenodo.20119582`) **Formal role in \\(\\mathcal{S}\\):** The Brain Stem is the runtime kernel that evaluates \\(\\Lambda\\",
339
+ "source_id": "thesis_session",
340
+ "maturity": "defined"
341
  },
342
  {
343
  "id": "F0025",
 
345
  "source_line": 265,
346
  "latex": "\\Lambda",
347
  "context": "DOI `10.5281/zenodo.20119582`) **Formal role in \\(\\mathcal{S}\\):** The Brain Stem is the runtime kernel that evaluates \\(\\Lambda\\) and emits receipts. Every edge in \\(E\\) that crosses a region bounda",
348
+ "source_id": "thesis_session",
349
+ "maturity": "defined"
350
  },
351
  {
352
  "id": "F0026",
 
354
  "source_line": 268,
355
  "latex": "\\Lambda",
356
  "context": "bda(axes: number[9|10]) \u2192 Receipt` \u2014 evaluates the conjunctive AND gate and returns a signed receipt with the composite \\(\\Lambda\\) score, Bekenstein budget, and dual-witness closure status. - `build_",
357
+ "source_id": "thesis_session",
358
+ "maturity": "conjectured"
359
  },
360
  {
361
  "id": "F0027",
 
363
  "source_line": 274,
364
  "latex": "\\Lambda",
365
  "context": "chain root for third-party verification. **Dependencies:** - Depends on: `lutar-lean` (Skeleton) \u2014 the axiom set that \\(\\Lambda\\) is required to satisfy is formally stated there; the Brain Stem is th",
366
+ "source_id": "thesis_session",
367
+ "maturity": "defined"
368
  },
369
  {
370
  "id": "F0028",
 
372
  "source_line": 277,
373
  "latex": "\\Lambda_9",
374
  "context": "utbound edge must call `evaluate_lambda` before the edge enters \\(E\\). The gate composition benchmark for v6.3.0 shows \\(\\Lambda_9\\) base p50 = 3.12 \u00b5s and composed p50 = 3.29 \u00b5s; with the Platform v",
375
+ "source_id": "thesis_session",
376
+ "maturity": "defined"
377
  },
378
  {
379
  "id": "F0029",
 
381
  "source_line": 285,
382
  "latex": "\\mathcal{S}",
383
  "context": "a continuous supply-chain security posture. --- ### Heart \u2014 `a11oy` **Repo:** `szl-holdings/a11oy` **Formal role in \\(\\mathcal{S}\\):** The Heart is the covenant policy engine and the agent approva",
384
+ "source_id": "thesis_session",
385
+ "maturity": "defined"
386
  },
387
  {
388
  "id": "F0030",
 
390
  "source_line": 285,
391
  "latex": "\\mathcal{S}",
392
  "context": "\\):** The Heart is the covenant policy engine and the agent approval queue. It governs the *authorization* dimension of \\(\\mathcal{S}\\): while the Brain Stem answers \"does this action score above \\(\\L",
393
+ "source_id": "thesis_session",
394
+ "maturity": "defined"
395
  },
396
  {
397
  "id": "F0031",
 
399
  "source_line": 285,
400
  "latex": "\\Lambda",
401
  "context": "It governs the *authorization* dimension of \\(\\mathcal{S}\\): while the Brain Stem answers \"does this action score above \\(\\Lambda\\)?\", the Heart answers \"is this action permitted under the active cove",
402
+ "source_id": "thesis_session",
403
+ "maturity": "defined"
404
  },
405
  {
406
  "id": "F0032",
 
408
  "source_line": 285,
409
  "latex": "r_{\\text{dst}} \\notin R",
410
  "context": "this action permitted under the active covenant?\". No action may exit the body graph \u2014 i.e., no edge in \\(E\\) may have \\(r_{\\text{dst}} \\notin R\\) \u2014 without a Heart pulse. The covenant is a named, ver",
411
+ "source_id": "thesis_session",
412
+ "maturity": "defined"
413
  },
414
  {
415
  "id": "F0033",
 
417
  "source_line": 293,
418
  "latex": "\\Lambda",
419
  "context": "Stem's chain. **Dependencies:** - Depends on: `ouroboros` (Brain Stem) \u2014 covenant evaluation results are sealed with a \\(\\Lambda\\)-gated receipt; a covenant check that fails \\(\\Lambda\\) is itself a g",
420
+ "source_id": "thesis_session",
421
+ "maturity": "defined"
422
  },
423
  {
424
  "id": "F0034",
425
  "source_file": "thesis.md",
426
  "source_line": 293,
427
  "latex": "\\Lambda",
428
+ "context": "os` (Brain Stem) \u2014 covenant evaluation results are sealed with a \\(\\Lambda\\)-gated receipt; a covenant check that fails \\(\\Lambda\\) is itself a gate-level violation. - Depends on: `safety-gate layer` (safety wires) \u2014 t",
429
+ "source_id": "thesis_session",
430
+ "maturity": "defined"
431
  },
432
  {
433
  "id": "F0035",
434
  "source_file": "thesis.md",
435
  "source_line": 305,
436
  "latex": "\\mathcal{S}",
437
+ "context": "but a verifiable, chain-linked artifact. --- ### safety wires \u2014 `safety-gate layer` **Repo:** `szl-holdings/safety-gate layer` **Formal role in \\(\\mathcal{S}\\):** The safety wires are the attribution trail \u2014 the afferent channel tha",
438
+ "source_id": "thesis_session",
439
+ "maturity": "defined"
440
  },
441
  {
442
  "id": "F0036",
443
  "source_file": "thesis.md",
444
  "source_line": 305,
445
  "latex": "\\text{attr}: E \\to A",
446
+ "context": "rent channel that carries signals inward and records *who observed what and when*. Formally, safety wires maintain the mapping \\(\\text{attr}: E \\to A\\), ensuring that every edge in \\(E\\) is attributable to a",
447
+ "source_id": "thesis_session",
448
+ "maturity": "defined"
449
  },
450
  {
451
  "id": "F0037",
452
  "source_file": "thesis.md",
453
  "source_line": 305,
454
  "latex": "\\mathcal{S}",
455
+ "context": "he mapping \\(\\text{attr}: E \\to A\\), ensuring that every edge in \\(E\\) is attributable to a named actor. Without safety wires, \\(\\mathcal{S}\\) degrades: edges carry receipts but not attributions, making the ",
456
+ "source_id": "thesis_session",
457
+ "maturity": "defined"
458
  },
459
  {
460
  "id": "F0038",
 
462
  "source_line": 308,
463
  "latex": "a \\in A",
464
  "context": "egal-accountability sense. **Public interfaces:** - `observe(edge, actor_id) \u2192 AttributionRecord` \u2014 records that actor \\(a \\in A\\) produced or consumed edge \\(e\\). - `attribution_trail(region, time_r",
465
+ "source_id": "thesis_session",
466
+ "maturity": "defined"
467
  },
468
  {
469
  "id": "F0039",
470
  "source_file": "thesis.md",
471
  "source_line": 325,
472
  "latex": "\\mathcal{S}",
473
+ "context": "aft-morrow-sogomonian-exec-outcome-attest`. --- ### reasoning spine \u2014 `reasoning core` **Repo:** `szl-holdings/reasoning core` **Formal role in \\(\\mathcal{S}\\):** The reasoning spine is the append-only coordination and protocol bridge",
474
+ "source_id": "thesis_session",
475
+ "maturity": "defined"
476
  },
477
  {
478
  "id": "F0040",
479
  "source_file": "thesis.md",
480
  "source_line": 325,
481
  "latex": "\\langle e_1, e_2, \\ldots, e_n \\rangle \\subseteq E",
482
+ "context": "ordered, hash-verified record of every state transition across the body graph. Formally, `reasoning core` maintains the sequence \\(\\langle e_1, e_2, \\ldots, e_n \\rangle \\subseteq E\\) ordered by timestamp, with",
483
+ "source_id": "thesis_session",
484
+ "maturity": "defined"
485
  },
486
  {
487
  "id": "F0041",
488
  "source_file": "thesis.md",
489
  "source_line": 339,
490
  "latex": "O(\\log n)",
491
+ "context": "(identified in the runtime roadmap) would upgrade the reasoning spine's linear hash-chain to a directed acyclic graph supporting \\(O(\\log n)\\) subset inclusion proofs \u2014 enabling privacy-preserving audits for re",
492
+ "source_id": "thesis_session",
493
+ "maturity": "defined"
494
  },
495
  {
496
  "id": "F0042",
 
498
  "source_line": 347,
499
  "latex": "\\mathcal{S}",
500
  "context": "nce in the enterprise segment. --- ### Skeleton \u2014 `lutar-lean` **Repo:** `szl-holdings/lutar-lean` **Formal role in \\(\\mathcal{S}\\):** The Skeleton is the formal scaffold \u2014 the Lean 4 axioms and M",
501
+ "source_id": "thesis_session",
502
+ "maturity": "defined"
503
  },
504
  {
505
  "id": "F0043",
 
507
  "source_line": 347,
508
  "latex": "\\{A1, A2, A3, A4\\}",
509
  "context": "es not execute at runtime; it is the *proof that the runtime is correct*. Formally, `lutar-lean` provides the axiom set \\(\\{A1, A2, A3, A4\\}\\) and the derived theorems (\u039b uniqueness, Bound theorem) th",
510
+ "source_id": "thesis_session",
511
+ "maturity": "conjectured"
512
  },
513
  {
514
  "id": "F0044",
 
516
  "source_line": 347,
517
  "latex": "\\Lambda",
518
  "context": "A2, A3, A4\\}\\) and the derived theorems (\u039b uniqueness, Bound theorem) that constitute a machine-checked certificate for \\(\\Lambda\\). If the Skeleton's `sorry` count is zero, the gate the Brain Stem en",
519
+ "source_id": "thesis_session",
520
+ "maturity": "conjectured"
521
  },
522
  {
523
  "id": "F0045",
 
525
  "source_line": 351,
526
  "latex": "\\Lambda",
527
  "context": "statements of A1 (monotonicity), A2 (homogeneity), A3 (Egyptian-exact), A4 (bounded). - `Uniqueness.lean` \u2014 Theorem 1: \\(\\Lambda\\) is the unique function satisfying A1\u2013A4; proof scaffold with tracked ",
528
+ "source_id": "thesis_session",
529
+ "maturity": "conjectured"
530
  },
531
  {
532
  "id": "F0046",
 
534
  "source_line": 367,
535
  "latex": "\\mathcal{S}",
536
  "context": "*Repos:** `szl-holdings/counsel` (governance UI), `szl-holdings/terra` (dashboards and visualization) **Formal role in \\(\\mathcal{S}\\):** The Hands are the tooling and visualization surfaces \u2014 the co",
537
+ "source_id": "thesis_session",
538
+ "maturity": "defined"
539
  },
540
  {
541
  "id": "F0047",
 
543
  "source_line": 371,
544
  "latex": "\\Lambda",
545
  "context": "as an interactive SVG, streaming live receipt counts via SSE from `/api/chain/stream`; node colors reflect the current \\(\\Lambda\\) score band (green \u2265 0.95, amber 0.90\u20130.95, red < 0.90). The planned \"",
546
+ "source_id": "thesis_session",
547
+ "maturity": "defined"
548
  },
549
  {
550
  "id": "F0048",
 
552
  "source_line": 386,
553
  "latex": "\\mathcal{S}",
554
  "context": "*is* the system. --- ### Full Body \u2014 `ouroboros-thesis` **Repo:** `szl-holdings/ouroboros-thesis` **Formal role in \\(\\mathcal{S}\\):** The Full Body is the public-record thesis \u2014 the DOI-pinned, ve",
555
+ "source_id": "thesis_session",
556
+ "maturity": "defined"
557
  },
558
  {
559
  "id": "F0049",
 
561
  "source_line": 386,
562
  "latex": "\\mathcal{S}",
563
  "context": "l Body is the public-record thesis \u2014 the DOI-pinned, versioned document that constitutes the canonical specification of \\(\\mathcal{S}\\). Formally, `ouroboros-thesis` defines the normative description ",
564
+ "source_id": "thesis_session",
565
+ "maturity": "defined"
566
  },
567
  {
568
  "id": "F0050",
 
570
  "source_line": 405,
571
  "latex": "\\mathcal{S}",
572
  "context": "d identity anchoring), `szl-holdings/szl-cookbook` (reference implementations / developer onboarding) **Formal role in \\(\\mathcal{S}\\):** The Vessels and Chakras collectively form the trust mesh and ",
573
+ "source_id": "thesis_session",
574
+ "maturity": "defined"
575
  },
576
  {
577
  "id": "F0051",
 
579
  "source_line": 421,
580
  "latex": "\\varepsilon",
581
  "context": "eue under the covenant pack schema. --- ## Cross-Region Contracts Every edge in \\(E\\) carries a **receipt envelope** \\(\\varepsilon\\). The envelope is a typed, signed, content-addressed record that ",
582
+ "source_id": "thesis_session",
583
+ "maturity": "defined"
584
  },
585
  {
586
  "id": "F0052",
 
588
  "source_line": 421,
589
  "latex": "\\Lambda",
590
  "context": "es a **receipt envelope** \\(\\varepsilon\\). The envelope is a typed, signed, content-addressed record that provides: the \\(\\Lambda\\) score vector, the dual-witness closure status (\\(\\rho\\)), the actor ",
591
+ "source_id": "thesis_session",
592
+ "maturity": "defined"
593
  },
594
  {
595
  "id": "F0053",
 
597
  "source_line": 479,
598
  "latex": "\\Lambda",
599
  "context": "_lambda(axes) \u2192 Receipt` \u2014 any MCP-compatible client (Claude Desktop, Cursor, enterprise agent frameworks) can call the \\(\\Lambda\\) gate as a typed tool and receive a signed receipt in the tool respon",
600
+ "source_id": "thesis_session",
601
+ "maturity": "defined"
602
  },
603
  {
604
  "id": "F0054",
 
606
  "source_line": 507,
607
  "latex": "\\mathcal{S}",
608
  "context": "the 8-Region Model Structurally Surpasses the Leaders Each leading framework or protocol is a partial instantiation of \\(\\mathcal{S}\\). The gap is structural: the missing region is not a feature that",
609
+ "source_id": "thesis_session",
610
+ "maturity": "defined"
611
  },
612
  {
613
  "id": "F0055",
 
615
  "source_line": 515,
616
  "latex": "\\Lambda_9",
617
  "context": "l engineering pattern, but skills are *files*, not services with receipts. A Brain Stem can issue a decision that fails \\(\\Lambda_9\\) moralGrounding; in the Managed Agents architecture there is no mec",
618
+ "source_id": "thesis_session",
619
+ "maturity": "defined"
620
  },
621
  {
622
  "id": "F0056",
 
624
  "source_line": 515,
625
  "latex": "\\mathcal{S}",
626
  "context": "fails \\(\\Lambda_9\\) moralGrounding; in the Managed Agents architecture there is no mechanism to detect or block it. In \\(\\mathcal{S}\\), that decision never exits the Brain Stem. **Mastra** (22K+ GitH",
627
+ "source_id": "thesis_session",
628
+ "maturity": "defined"
629
  },
630
  {
631
  "id": "F0057",
 
633
  "source_line": 517,
634
  "latex": "\\Lambda",
635
  "context": "ource agent framework in the TypeScript ecosystem. Mastra has no Skeleton: there are no Lean 4 proofs. It has no formal \\(\\Lambda\\) gate \u2014 behavioral constraints are implemented as runtime checks with",
636
+ "source_id": "thesis_session",
637
+ "maturity": "defined"
638
  },
639
  {
640
  "id": "F0058",
 
642
  "source_line": 567,
643
  "latex": "\\lambda_1",
644
  "context": "l(\\lambda_1(c),\\, \\lambda_2(c),\\, \\ldots,\\, \\lambda_9(c)\\bigr) \\in [0,1]^9 \\] The nine axes are defined as follows. **\\(\\lambda_1\\): moralGrounding.** Measures the degree to which a proposed action ",
645
+ "source_id": "thesis_session",
646
+ "maturity": "defined"
647
  },
648
  {
649
  "id": "F0059",
 
651
  "source_line": 567,
652
  "latex": "\\lambda_1",
653
  "context": "nce policies, and principal hierarchies that the operator has encoded in the agent's governing covenant. Operationally, \\(\\lambda_1\\) is the normalized cosine similarity between the action's intent em",
654
+ "source_id": "thesis_session",
655
+ "maturity": "defined"
656
  },
657
  {
658
  "id": "F0060",
 
660
  "source_line": 567,
661
  "latex": "[0,1]",
662
  "context": "mbedding and a reference \"moral anchor\" embedding, averaged over the operator's registered covenant clauses, clamped to \\([0,1]\\). The floor constraint \\(\\lambda_1 \\geq 0.95\\) is a hard asymptote: an ",
663
+ "source_id": "thesis_session",
664
+ "maturity": "defined"
665
  },
666
  {
667
  "id": "F0061",
 
669
  "source_line": 567,
670
  "latex": "\\lambda_1 \\geq 0.95",
671
  "context": "anchor\" embedding, averaged over the operator's registered covenant clauses, clamped to \\([0,1]\\). The floor constraint \\(\\lambda_1 \\geq 0.95\\) is a hard asymptote: an agent that is even marginally mo",
672
+ "source_id": "thesis_session",
673
+ "maturity": "defined"
674
  },
675
  {
676
  "id": "F0062",
 
678
  "source_line": 569,
679
  "latex": "\\lambda_2",
680
  "context": "even marginally morally misaligned fails the gate irrespective of how perfectly calibrated the other eight axes are. **\\(\\lambda_2\\): measurabilityHonesty.** Measures whether an action's declared eff",
681
+ "source_id": "thesis_session",
682
+ "maturity": "defined"
683
  },
684
  {
685
  "id": "F0063",
 
687
  "source_line": 571,
688
  "latex": "\\lambda_3",
689
  "context": "ine clause \"no hallucinations no bandaids; test test test\" by making measurement-honesty a prerequisite for passage. **\\(\\lambda_3\\): epistemicHumility.** Scores the agent's acknowledgment of its own",
690
+ "source_id": "thesis_session",
691
+ "maturity": "defined"
692
  },
693
  {
694
  "id": "F0064",
 
696
  "source_line": 571,
697
  "latex": "\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]",
698
  "context": "sparse scores low on this axis. The scoring function penalizes unjustified confidence using a calibration-error analog: \\(\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{c",
699
+ "source_id": "thesis_session",
700
+ "maturity": "defined"
701
  },
702
  {
703
  "id": "F0065",
 
705
  "source_line": 571,
706
  "latex": "\\text{conf}(c)",
707
  "context": "ied confidence using a calibration-error analog: \\(\\lambda_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{conf}(c)\\) is the agent's stated confidence and \\(\\text{acc}(c)\\) is the",
708
+ "source_id": "thesis_session",
709
+ "maturity": "defined"
710
  },
711
  {
712
  "id": "F0066",
 
714
  "source_line": 571,
715
  "latex": "\\text{acc}(c)",
716
  "context": "da_3 = 1 - \\mathbb{E}[|\\text{conf}(c) - \\text{acc}(c)|]\\) where \\(\\text{conf}(c)\\) is the agent's stated confidence and \\(\\text{acc}(c)\\) is the empirically measured accuracy over a calibration set. ",
717
+ "source_id": "thesis_session",
718
+ "maturity": "defined"
719
  },
720
  {
721
  "id": "F0067",
 
723
  "source_line": 573,
724
  "latex": "\\lambda_4",
725
  "context": "is the agent's stated confidence and \\(\\text{acc}(c)\\) is the empirically measured accuracy over a calibration set. **\\(\\lambda_4\\): counterfactualAwareness.** Measures whether the agent has consider",
726
+ "source_id": "thesis_session",
727
+ "maturity": "defined"
728
  },
729
  {
730
  "id": "F0068",
 
732
  "source_line": 575,
733
  "latex": "\\lambda_5",
734
  "context": "res 0.0 and a uniformly distributed consequence distribution over the operator-defined consequence space scores 1.0. **\\(\\lambda_5\\): temporalConsistency.** Measures the stability of the gate verdict",
735
+ "source_id": "thesis_session",
736
+ "maturity": "defined"
737
  },
738
  {
739
  "id": "F0069",
 
741
  "source_line": 575,
742
  "latex": "t + \\Delta",
743
  "context": "Measures the stability of the gate verdict under repeated evaluation on the same input at two different times \\(t\\) and \\(t + \\Delta\\). Let \\(v_t\\) and \\(v_{t+\\Delta}\\) denote the \u039b\u2089 composite scores ",
744
+ "source_id": "thesis_session",
745
+ "maturity": "defined"
746
  },
747
  {
748
  "id": "F0070",
 
750
  "source_line": 575,
751
  "latex": "v_{t+\\Delta}",
752
  "context": "te verdict under repeated evaluation on the same input at two different times \\(t\\) and \\(t + \\Delta\\). Let \\(v_t\\) and \\(v_{t+\\Delta}\\) denote the \u039b\u2089 composite scores at the two evaluation times. The",
753
+ "source_id": "thesis_session",
754
+ "maturity": "defined"
755
  },
756
  {
757
  "id": "F0071",
 
759
  "source_line": 581,
760
  "latex": "\\lambda_5 = 1.0",
761
  "context": "Then: \\[ \\lambda_5 = \\max\\!\\Bigl(0,\\; 1 - 4\\,\\bigl(v_t - v_{t+\\Delta}\\bigr)^2\\Bigr) \\] A zero-drift evaluation scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 =",
762
+ "source_id": "thesis_session",
763
+ "maturity": "defined"
764
  },
765
  {
766
  "id": "F0072",
 
768
  "source_line": 581,
769
  "latex": "\\lambda_5 = 0.99",
770
  "context": "ta}\\bigr)^2\\Bigr) \\] A zero-drift evaluation scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 = 0.99\\). A drift of 0.25 yields \\(\\lambda_5 = 0.75\\), below the \u2265 0",
771
+ "source_id": "thesis_session",
772
+ "maturity": "defined"
773
  },
774
  {
775
  "id": "F0073",
 
777
  "source_line": 581,
778
  "latex": "\\lambda_5 = 0.75",
779
  "context": "scores \\(\\lambda_5 = 1.0\\). A drift of 0.05 in the composite score yields \\(\\lambda_5 = 0.99\\). A drift of 0.25 yields \\(\\lambda_5 = 0.75\\), below the \u2265 0.90 conjunctive floor. This axis operationaliz",
780
+ "source_id": "thesis_session",
781
+ "maturity": "defined"
782
  },
783
  {
784
  "id": "F0074",
 
786
  "source_line": 583,
787
  "latex": "\\lambda_6",
788
  "context": "-identical replay guarantee: a system that cannot reproduce its own gate verdict is not operating deterministically. **\\(\\lambda_6\\): evidenceProvenance.** Measures whether every empirical claim embe",
789
+ "source_id": "thesis_session",
790
+ "maturity": "defined",
791
+ "puriq_ref": "F1",
792
+ "lean_ref": "f1_replay_fold_deterministic"
793
  },
794
  {
795
  "id": "F0075",
 
797
  "source_line": 585,
798
  "latex": "\\lambda_7",
799
  "context": "ertions score at most 0.50. The scoring function is the fraction of claim tokens for which provenance is resolvable. **\\(\\lambda_7\\): actorIdentity.** Measures the definiteness of the acting agent's ",
800
+ "source_id": "thesis_session",
801
+ "maturity": "defined"
802
  },
803
  {
804
  "id": "F0076",
 
806
  "source_line": 587,
807
  "latex": "\\lambda_8",
808
  "context": "ting under delegated authority \u2014 the score decays as a function of delegation depth to penalize opaque proxy chains. **\\(\\lambda_8\\): axiomConsistency.** Measures whether the proposed action is inter",
809
+ "source_id": "thesis_session",
810
+ "maturity": "defined"
811
  },
812
  {
813
  "id": "F0077",
 
815
  "source_line": 589,
816
  "latex": "\\lambda_9",
817
  "context": "Lean 4 formalization: it enforces, at runtime, the constraints that are statically verified at theorem-proving time. **\\(\\lambda_9\\): coherence.** Measures the multi-step logical coherence of the age",
818
+ "source_id": "thesis_session",
819
+ "maturity": "defined"
820
  },
821
  {
822
  "id": "F0078",
 
824
  "source_line": 589,
825
  "latex": "A_1, A_2, \\ldots, A_k",
826
  "context": "-step logical coherence of the agent's plan across the action sequence, not just for the current step in isolation. Let \\(A_1, A_2, \\ldots, A_k\\) denote the \\(k\\) preceding actions in the current sess",
827
+ "source_id": "thesis_session",
828
+ "maturity": "defined"
829
  },
830
  {
831
  "id": "F0079",
 
833
  "source_line": 589,
834
  "latex": "(A_i, A_{i+1})",
835
  "context": "e the \\(k\\) preceding actions in the current session. The coherence score is the proportion of consecutive action-pairs \\((A_i, A_{i+1})\\) for which the precondition of \\(A_{i+1}\\) is satisfied by the",
836
+ "source_id": "thesis_session",
837
+ "maturity": "defined"
838
  },
839
  {
840
  "id": "F0080",
 
842
  "source_line": 589,
843
  "latex": "A_{i+1}",
844
  "context": "ion. The coherence score is the proportion of consecutive action-pairs \\((A_i, A_{i+1})\\) for which the precondition of \\(A_{i+1}\\) is satisfied by the postcondition of \\(A_i\\), under the operator's p",
845
+ "source_id": "thesis_session",
846
+ "maturity": "defined"
847
  },
848
  {
849
  "id": "F0081",
 
851
  "source_line": 589,
852
  "latex": "\\lambda_9 = 1.0",
853
  "context": "ied by the postcondition of \\(A_i\\), under the operator's precondition/postcondition schema. For the base case \\(k=0\\), \\(\\lambda_9 = 1.0\\). ### The conjunctive gate condition The \u039b\u2089 gate passes if ",
854
+ "source_id": "thesis_session",
855
+ "maturity": "defined"
856
  },
857
  {
858
  "id": "F0082",
 
860
  "source_line": 599,
861
  "latex": "\\lambda_1 = 0.50",
862
  "context": "for the following reason. A single composite score \u2014 even a geometric mean \u2014 can mask localized failures. An agent with \\(\\lambda_1 = 0.50\\) (severely morally misaligned) and all remaining axes at \\(1",
863
+ "source_id": "thesis_session",
864
+ "maturity": "defined"
865
  },
866
  {
867
  "id": "F0083",
 
869
  "source_line": 599,
870
  "latex": "\\prod_{i}^{1/9} = 0.50^{1/9} \\approx 0.926",
871
  "context": "with \\(\\lambda_1 = 0.50\\) (severely morally misaligned) and all remaining axes at \\(1.0\\) achieves a geometric mean of \\(\\prod_{i}^{1/9} = 0.50^{1/9} \\approx 0.926\\), which would pass a \u2265 0.90 single-",
872
+ "source_id": "thesis_session",
873
+ "maturity": "defined"
874
  },
875
  {
876
  "id": "F0084",
 
878
  "source_line": 599,
879
  "latex": "\\lambda_1",
880
  "context": "gle-score gate. The conjunctive AND structure prevents this: every axis is a blocking veto. The two elevated floors for \\(\\lambda_1\\) and \\(\\lambda_2\\) add a second layer of asymmetry \u2014 these are the ",
881
+ "source_id": "thesis_session",
882
+ "maturity": "defined"
883
  },
884
  {
885
  "id": "F0085",
 
887
  "source_line": 599,
888
  "latex": "\\lambda_2",
889
  "context": "e conjunctive AND structure prevents this: every axis is a blocking veto. The two elevated floors for \\(\\lambda_1\\) and \\(\\lambda_2\\) add a second layer of asymmetry \u2014 these are the axes most directly",
890
+ "source_id": "thesis_session",
891
+ "maturity": "defined"
892
  },
893
  {
894
  "id": "F0086",
 
896
  "source_line": 605,
897
  "latex": "m \\in \\{0,1\\}^9",
898
  "context": "to the receipt structure. Rather than publishing the raw nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evalua",
899
+ "source_id": "thesis_session",
900
+ "maturity": "defined"
901
  },
902
  {
903
  "id": "F0087",
 
905
  "source_line": 605,
906
  "latex": "m_i = 1",
907
  "context": "her than publishing the raw nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evaluated and passed its floor. The",
908
+ "source_id": "thesis_session",
909
+ "maturity": "defined"
910
  },
911
  {
912
  "id": "F0088",
 
914
  "source_line": 605,
915
  "latex": "\\lambda_i",
916
  "context": "nine (or ten) axis scores, the receipt carries a bitfield mask \\(m \\in \\{0,1\\}^9\\) in which \\(m_i = 1\\) if and only if \\(\\lambda_i\\) was evaluated and passed its floor. The raw scores are withheld fro",
917
+ "source_id": "thesis_session",
918
+ "maturity": "defined"
919
  },
920
  {
921
  "id": "F0089",
 
923
  "source_line": 611,
924
  "latex": "\\theta_i = 0.95",
925
  "context": "ity profile of the agent. Formally, the mask is computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The",
926
+ "source_id": "thesis_session",
927
+ "maturity": "defined"
928
  },
929
  {
930
  "id": "F0090",
 
932
  "source_line": 611,
933
  "latex": "i \\in \\{1,2\\}",
934
  "context": ". Formally, the mask is computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_",
935
+ "source_id": "thesis_session",
936
+ "maturity": "defined"
937
  },
938
  {
939
  "id": "F0091",
 
941
  "source_line": 611,
942
  "latex": "\\theta_i = 0.90",
943
  "context": "s computed as: \\[ m_i = \\mathbf{1}[\\lambda_i(c) \\geq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_i m_i = 9\\) (or 10 und",
944
+ "source_id": "thesis_session",
945
+ "maturity": "defined"
946
  },
947
  {
948
  "id": "F0092",
 
950
  "source_line": 611,
951
  "latex": "\\sum_i m_i = 9",
952
  "context": "eq \\theta_i] \\] where \\(\\theta_i = 0.95\\) for \\(i \\in \\{1,2\\}\\) and \\(\\theta_i = 0.90\\) otherwise. The gate passes iff \\(\\sum_i m_i = 9\\) (or 10 under \u039b\u2081\u2080). The mask is committed via SHA-256 and incl",
953
+ "source_id": "thesis_session",
954
+ "maturity": "defined"
955
  },
956
  {
957
  "id": "F0093",
 
959
  "source_line": 627,
960
  "latex": "\\textit{parent\\_hash}",
961
  "context": "\\textit{timestamp},\\; \\vec{\\lambda},\\; \\rho\\_\\textit{witness\\_set},\\; \\textit{signature}\\bigr) \\] The fields are: - **\\(\\textit{parent\\_hash}\\)**: The SHA-256 digest of receipt \\(r_{i-1}\\). For the ",
962
+ "source_id": "thesis_session",
963
+ "maturity": "defined"
964
  },
965
  {
966
  "id": "F0094",
 
968
  "source_line": 627,
969
  "latex": "r_{i-1}",
970
  "context": "s\\_set},\\; \\textit{signature}\\bigr) \\] The fields are: - **\\(\\textit{parent\\_hash}\\)**: The SHA-256 digest of receipt \\(r_{i-1}\\). For the genesis receipt, this is the SHA-256 of a protocol-specifie",
971
+ "source_id": "thesis_session",
972
+ "maturity": "defined"
973
  },
974
  {
975
  "id": "F0095",
 
977
  "source_line": 628,
978
  "latex": "\\textit{content\\_digest}",
979
  "context": "this is the SHA-256 of a protocol-specified null seed. This field creates the backward-pointing link of the chain. - **\\(\\textit{content\\_digest}\\)**: The SHA-256 of the canonical JSON serialization o",
980
+ "source_id": "thesis_session",
981
+ "maturity": "defined"
982
  },
983
  {
984
  "id": "F0096",
 
986
  "source_line": 629,
987
  "latex": "\\textit{actor}",
988
  "context": "fore any side-effectful execution. This binds the gate verdict irrevocably to the specific input that triggered it. - **\\(\\textit{actor}\\)**: The identifier of the acting agent as registered in the pr",
989
+ "source_id": "thesis_session",
990
+ "maturity": "defined"
991
  },
992
  {
993
  "id": "F0097",
 
995
  "source_line": 629,
996
  "latex": "\\lambda_7",
997
  "context": "t. - **\\(\\textit{actor}\\)**: The identifier of the acting agent as registered in the principal registry. Corresponds to \\(\\lambda_7\\) (actorIdentity). - **\\(\\textit{timestamp}\\)**: A monotonic timesta",
998
+ "source_id": "thesis_session",
999
+ "maturity": "defined"
1000
  },
1001
  {
1002
  "id": "F0098",
 
1004
  "source_line": 630,
1005
  "latex": "\\textit{timestamp}",
1006
  "context": "entifier of the acting agent as registered in the principal registry. Corresponds to \\(\\lambda_7\\) (actorIdentity). - **\\(\\textit{timestamp}\\)**: A monotonic timestamp in milliseconds since the Unix e",
1007
+ "source_id": "thesis_session",
1008
+ "maturity": "defined"
1009
  },
1010
  {
1011
  "id": "F0099",
 
1013
  "source_line": 631,
1014
  "latex": "\\vec{\\lambda}",
1015
  "context": ": A monotonic timestamp in milliseconds since the Unix epoch, drawn from a pinned, non-forgeable source (see \u00a74.6). - **\\(\\vec{\\lambda}\\)**: The full nine-dimensional \u039b vector, or the `lambda9_mask` b",
1016
+ "source_id": "thesis_session",
1017
+ "maturity": "defined"
1018
  },
1019
  {
1020
  "id": "F0100",
 
1022
  "source_line": 632,
1023
  "latex": "\\rho\\_\\textit{witness\\_set}",
1024
  "context": "vec{\\lambda}\\)**: The full nine-dimensional \u039b vector, or the `lambda9_mask` bitfield under the \u039b\u2081\u2080 privacy variant. - **\\(\\rho\\_\\textit{witness\\_set}\\)**: The set of co-witnesses whose signatures are ",
1025
+ "source_id": "thesis_session",
1026
+ "maturity": "defined"
1027
  }
1028
  ],
1029
  "definitions": [],
 
1685
  "10.5281/zenodo.20053163",
1686
  "10.5281/zenodo.20119582",
1687
  "10.5281/zenodo.20162352"
1688
+ ],
1689
+ "proof_summary": {
1690
+ "locked_proven": 5,
1691
+ "locked_ids": [
1692
+ "F1",
1693
+ "F11",
1694
+ "F12",
1695
+ "F18",
1696
+ "F19"
1697
+ ],
1698
+ "experimental_sorry_free": 21,
1699
+ "axiom_gated": 3,
1700
+ "axiom_gated_detail": {
1701
+ "f13_tamper_evident": "hash_collision_resistant",
1702
+ "f14_dsse_verifiable": "ecdsa_unforgeable",
1703
+ "f15_inclusion_binding": "h2_collision_resistant"
1704
+ },
1705
+ "conjecture": [
1706
+ "F23"
1707
+ ],
1708
+ "note": "Locked kernel proven=5; experimental scope Lutar/Puriq/Formulas has 21 sorry-free (excluded from locked count); F23 = Conjecture 1, NOT a theorem.",
1709
+ "lean_repo": "szl-holdings/lutar-lean",
1710
+ "lean_files": [
1711
+ "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
1712
+ "Lutar/Puriq/Formulas/F23_Uniqueness.lean"
1713
+ ],
1714
+ "verification": "bare `lean` 4.13.0, 0 errors, 1 sorry (F23 only); #print axioms shows no sorryAx in any proved theorem.",
1715
+ "source_report": "team/PROOFS_WAVE2_REPORT.md",
1716
+ "wave3": {
1717
+ "campaign": "prove-wave-3 (C1-C20 research candidates)",
1718
+ "source_report": "team/PROVE_WAVE3_REPORT.md",
1719
+ "lean_repo": "szl-holdings/lutar-lean",
1720
+ "commit_proofs": "775093f0f8ef7f530272c38d513c28fdaec3366b",
1721
+ "commit_root_wiring": "02e44c30657c9986475ff7373113728f4ba38f67",
1722
+ "lean_files": [
1723
+ "Lutar/Wave3/Consensus.lean",
1724
+ "Lutar/Wave3/MerkleKraft.lean",
1725
+ "Lutar/Wave3/InfoEstim.lean",
1726
+ "Lutar/Wave3/Tier1Mathlib.lean (CI-pending, not wired into lake build)"
1727
+ ],
1728
+ "verification": "Mathlib-free modules bare-`lean` 4.13.0 verified sorry-free (0 errors); #print axioms ledger shows no sorryAx. Tier1Mathlib (C1/C2/C6) is Mathlib-dependent and CI-pending, NOT compiled in sandbox.",
1729
+ "new_proven_sorry_free": 19,
1730
+ "new_proven_ids": [
1731
+ "C8",
1732
+ "C9",
1733
+ "C10",
1734
+ "C11",
1735
+ "C12",
1736
+ "C17",
1737
+ "C20"
1738
+ ],
1739
+ "new_axiom_gated": 4,
1740
+ "new_axiom_gated_detail": {
1741
+ "c13_md_step_cr": "compression_collision_resistant",
1742
+ "c13a_md_append_cr": "compression_collision_resistant",
1743
+ "c14_merkle_binding": "node_collision_resistant, leaf_collision_resistant, domain_separation",
1744
+ "c14b_no_second_preimage": "domain_separation (structural tag only, no hardness)"
1745
+ },
1746
+ "ci_pending": [
1747
+ "C1",
1748
+ "C2",
1749
+ "C6"
1750
+ ],
1751
+ "ci_pending_detail": "C1 tsirelson_inequality, C2 CHSH_inequality_of_comm, C6 ConvexOn.map_sum_le re-exports; Mathlib-dependent, awaiting green lake build.",
1752
+ "maturity": {
1753
+ "C1": "ci-pending",
1754
+ "C2": "ci-pending",
1755
+ "C3": "mathlib-available-not-instantiated",
1756
+ "C4": "mathlib-available-not-instantiated",
1757
+ "C5": "mathlib-available-not-instantiated",
1758
+ "C6": "ci-pending",
1759
+ "C7": "axiom-gated (A6_bisymmetric); Lambda still Conjecture 1",
1760
+ "C8": "proven",
1761
+ "C9": "proven (Mathlib-free fragment; full L>=H is Mathlib target)",
1762
+ "C10": "proven",
1763
+ "C11": "proven",
1764
+ "C12": "proven (bivalence core; full FLP not claimed)",
1765
+ "C13": "axiom-gated",
1766
+ "C14": "axiom-gated",
1767
+ "C15": "lean-exists-not-ported",
1768
+ "C16": "not-attempted",
1769
+ "C17": "proven (Mathlib-free scalar core; full matrix-PSD is Mathlib target)",
1770
+ "C18": "lean-exists-not-ported",
1771
+ "C19": "not-attempted",
1772
+ "C20": "proven (Mathlib-free order-preservation core; tight 1/2-Lipschitz is Mathlib target)"
1773
+ },
1774
+ "lambda_status": "F23 = Conjecture 1 (UNCHANGED). C7 is conditional only, via the DECLARED axiom A6_bisymmetric in F23_Uniqueness.lean; unconditional uniqueness is FALSE under A1-A5 (maxAgg_ne_Lambda).",
1775
+ "locked_kernel": "749/14/163 @ c7c0ba17 (Doctrine v11) UNCHANGED; wave3 is experimental and counter-excluded from the locked count.",
1776
+ "headline": "+19 sorry-free (Lean-core axioms only, bare-lean verified), +4 axiom-gated (declared idealizations), 3 Mathlib re-exports CI-pending, Lambda still Conjecture 1."
1777
+ },
1778
+ "wave4": {
1779
+ "campaign": "prove-wave-4 (conditional Lambda uniqueness on the WEAKER block-consistency axiom)",
1780
+ "source_report": "team/PROVE_WAVE4_REPORT.md",
1781
+ "candidate_research": "team/RESEARCH_WAVE4/CANDIDATE_FORMULAS_V4.md",
1782
+ "lean_repo": "szl-holdings/lutar-lean",
1783
+ "commit_final": "043c3df4bcbe55c60f1ce2d5c59b91284a7cc1d4",
1784
+ "commit_ci_green_lambda": "52d9bf542bcb1adb8a0a5a5de694f2ca96bf9b68",
1785
+ "lean_files": [
1786
+ "Lutar/Wave4/LambdaBlockConsistency.lean (Mathlib-dependent, CI-green: lake build + kernel check success @ 043c3df)",
1787
+ "Lutar/Wave4/LambdaBisymmetryWitness.lean (bare-`lean` 4.13.0 verified sorry-free, ZERO axioms; also CI-green)",
1788
+ "Lutar/Wave3/Tier1Mathlib.lean (CI-PENDING, NOT wired into the compiled root)"
1789
+ ],
1790
+ "ci_status": "build + lake build + numbers + check/doctrine all GREEN @ 043c3df; only doi-title-gate fails (PRE-EXISTING live-network README DOI check, unrelated to wave4).",
1791
+ "verification": "LambdaBlockConsistency kernel-checked by lutar-lean CI lake build (green). LambdaBisymmetryWitness bare-`lean` verified: all 6 theorems 'do not depend on any axioms'. Every theorem carries #print axioms.",
1792
+ "new_proven_ci_green": {
1793
+ "lambda_unique_under_block": "CLOSED, conditional on declared axiom A6'_block_consistent; #print axioms = [A6'_block_consistent, propext, Quot.sound, Classical.choice]",
1794
+ "lambda_factors": "CLOSED, AXIOM-FREE (Mathlib core only): Lambda factors with exponents 1/k, so A6' is non-vacuous",
1795
+ "unconditional_lambda_is_false": "CLOSED (= maxAgg_ne_Lambda): unconditional Lambda uniqueness is FALSE under A1-A5"
1796
+ },
1797
+ "witness_theorems_zero_axiom": [
1798
+ "Fmax_not_strict",
1799
+ "Fmin_not_strict",
1800
+ "geo_separates_where_max_collapses",
1801
+ "geo_bisym_product_eq",
1802
+ "geo_fourth_root_consistent",
1803
+ "geo_inner_products_consistent"
1804
+ ],
1805
+ "lambda_axiom_set": "{A1,A2,A3,A4,A5} + A6'_block_consistent (single DECLARED, disclosed, NON-core axiom).",
1806
+ "lambda_weakest_axiom": "Cleanest published: Aczel-Saaty 1983 (doi:10.1016/0022-2496(83)90028-7) = reciprocity + positive homogeneity (A2 already assumed). Weakest governance-natural & formalized: Csato 2018 block-consistency / aggregation-invariance (doi:10.1007/s10726-018-9589-3, arXiv:1706.07256), WEAKER than the prior A6_bisymmetric.",
1807
+ "lambda_status": "F23 = Conjecture 1 (UNCHANGED, unconditional). Conditional uniqueness now CI-green on the WEAKER A6'_block_consistent (lambda_unique_under_block), superseding the stronger A6_bisymmetric route. Unconditional uniqueness FALSE (maxAgg_ne_Lambda). NEVER conflated.",
1808
+ "ci_pending": [
1809
+ "C1",
1810
+ "C2",
1811
+ "C6"
1812
+ ],
1813
+ "ci_pending_detail": "C1 tsirelson_inequality / C2 CHSH_inequality_of_comm / C6 ConvexOn.map_sum_le re-exports. Signatures verified VERBATIM vs pinned Mathlib d731765, but wiring Tier1Mathlib into the compiled root reproducibly red-lights lake build (bisected: a4299fb/52d9bf5 un-wired = green). Exact error not retrievable (CI log download proxy-blocked). File stays in-tree, NOT imported; NOT claimed proven.",
1814
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED",
1815
+ "locked_proven": 5,
1816
+ "canonical_numbers": {
1817
+ "declarations": 1182,
1818
+ "axioms_raw": 20,
1819
+ "axioms_unique": 19,
1820
+ "new_axiom": "A6'_block_consistent (declared, disclosed, NON-core, NOT in locked kernel)",
1821
+ "sorries_raw": 308,
1822
+ "sorries_noncomment": 256,
1823
+ "drift_gate": "PASS"
1824
+ },
1825
+ "citations": [
1826
+ "Aczel 1948",
1827
+ "Aczel-Saaty 1983 doi:10.1016/0022-2496(83)90028-7",
1828
+ "Csato 2018 doi:10.1007/s10726-018-9589-3 arXiv:1706.07256",
1829
+ "Kolmogorov 1930",
1830
+ "Maksa-Munnich-Mokken",
1831
+ "Burai-Kiss-Szokol 2021"
1832
+ ]
1833
+ },
1834
+ "wave5": {
1835
+ "campaign": "prove-wave-5: un-block C1/C2/C6 Mathlib re-exports (CI-GREEN) + new substrate re-exports (AM-GM/Cauchy-Schwarz) + Mathlib-free discrete substrate guarantees (bare-lean verified)",
1836
+ "source_report": "team/PROVE_WAVE5_REPORT.md",
1837
+ "lean_repo": "szl-holdings/lutar-lean",
1838
+ "branch": "prove-wave5/c1c2c6-rewire-plus-amgm-cs",
1839
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/186",
1840
+ "commit_ci_green": "0a552a90dd7f3b8b668ae761bf6e39eca17c62f1",
1841
+ "ci_run_ids": {
1842
+ "lean_kernel_check": "27053443102 (success)",
1843
+ "lake_build_gate_numbers": "27053443099 (success)",
1844
+ "doctrine": "27053443200 (success)",
1845
+ "dco": "27053443096 (success)"
1846
+ },
1847
+ "ci_status": "build (Lean kernel check) + lake build + numbers + check/doctrine + DCO all GREEN @ 0a552a90 (and @ 099d6caa). Only doi-title-gate + PR-title-lint fail (PRE-EXISTING / cosmetic, unrelated to proofs).",
1848
+ "headline": "C1 Tsirelson 2sqrt2 / C2 CHSH<=2 / C6 Jensen are now CI-GREEN (wave-4 had them CI-PENDING). Root cause fixed: dropped the non-load-bearing c1a_tsirelson_constant numeric remark and its two extra SpecialFunctions imports, minimizing Tier1Mathlib's build closure to exactly the two modules that define the instantiated theorems.",
1849
+ "ci_green_mathlib_dependent": {
1850
+ "Wave3.Tier1.c1_lutar_omega_tsirelson_ceiling": "C1 Tsirelson 2sqrt2 ceiling (tsirelson_inequality) \u2014 PROVEN, CI-green. EPR-Bell governance diagnostic (entangled-agent ceiling).",
1851
+ "Wave3.Tier1.c2_lutar_omega_classical_ceiling": "C2 CHSH classical ceiling <=2 (CHSH_inequality_of_comm) \u2014 PROVEN, CI-green. Local/independent-prior agent ceiling.",
1852
+ "Wave3.Tier1.c6_jensen_forecaster": "C6 finite Jensen (ConvexOn.map_sum_le) \u2014 PROVEN, CI-green. Active-inference ELBO-direction conservative forecaster.",
1853
+ "Wave5.MathlibCore.w5_1_lambda_le_arith_mean": "W5-1 weighted AM-GM (Real.geom_mean_le_arith_mean_weighted) \u2014 PROVEN, CI-green. Lambda (geometric-mean aggregator) <= arithmetic mean: no-inflation guarantee.",
1854
+ "Wave5.MathlibCore.w5_1b_lambda2_le_arith_mean": "W5-1b two-point weighted AM-GM \u2014 PROVEN, CI-green. Pairwise consensus diagnostic.",
1855
+ "Wave5.MathlibCore.w5_2_trust_inner_le_norm": "W5-2 Cauchy-Schwarz (real_inner_le_norm) \u2014 PROVEN, CI-green. Trust-vector similarity bound (cosine in [-1,1])."
1856
+ },
1857
+ "proven_mathlib_free_bare_lean": {
1858
+ "Wave5.DiscreteSubstrate.w5_3a_miscover_le_total": "miscoverage<=sample size. axioms=[propext]. killinchu conformal coverage.",
1859
+ "Wave5.DiscreteSubstrate.w5_3b_cover_miscover_partition": "covered+miscovered=total. axioms=[propext, Quot.sound]. coverage=1-miscoverage conservation.",
1860
+ "Wave5.DiscreteSubstrate.w5_3c_threshold_count_mono": "stricter threshold selects fewer. axioms=[propext, Quot.sound]. a11oy threshold monotonicity.",
1861
+ "Wave5.DiscreteSubstrate.w5_4_collision_of_image_dup": "image-duplicate => hash collision (pigeonhole). axioms=[propext, Classical.choice, Quot.sound]. UDS forgery-detection.",
1862
+ "Wave5.DiscreteSubstrate.w5_5_no_early_stop_deflation": "monotone optional-stopping anti-deflation. ZERO axioms. UDS receipt-stream anti-gaming."
1863
+ },
1864
+ "axiom_disclosure": "Mathlib-dependent re-exports use the standard Mathlib trio [propext, Classical.choice, Quot.sound] (NO sorryAx, NO declared Lutar axioms); their #print axioms are emitted in the CI build log (blob log download proxy-blocked here, but the build is green and they are pure term-mode instantiations of axiom-clean Mathlib theorems). Mathlib-free theorems' #print axioms pasted verbatim in PROVE_WAVE5_REPORT.md section 3 (bare lean 4.13.0, exit 0).",
1865
+ "not_available_at_pinned_mathlib": "C3 Hoeffding / C4 Azuma (Mathlib.Probability.Moments.SubGaussian) and C5 KL>=0 (Mathlib.InformationTheory.KullbackLeibler.Basic) modules DO NOT EXIST at the pinned rev d7317655 (v4.13.0) \u2014 verified HTTP 404. They cannot be re-exported on this toolchain; deferred to a future Mathlib bump. Honestly NOT claimed.",
1866
+ "lambda_status": "Lambda (F23) STAYS Conjecture 1 unconditionally. W5-1 AM-GM is a building block Lambda relies on; it does NOT prove uniqueness. Unconditional uniqueness remains FALSE (wave-4 counterexample in-tree).",
1867
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED. locked_proven=5 UNCHANGED. All wave-5 work is experimental scope (counter-excluded).",
1868
+ "canonical_numbers": {
1869
+ "declarations": 1189,
1870
+ "axioms_raw": 20,
1871
+ "axioms_unique": 19,
1872
+ "sorries_raw": 308,
1873
+ "sorries_noncomment": 256,
1874
+ "delta_decls_from_wave4": "+5 net (1184->1189; -1 c1a, +3 MathlibCore, +5 DiscreteSubstrate vs wave4 baseline 1182 -> 1189)"
1875
+ },
1876
+ "citations": [
1877
+ "Tsirelson (1980) doi:10.1007/BF00417500",
1878
+ "CHSH (1969) doi:10.1103/PhysRevLett.23.880",
1879
+ "Jensen (1906)",
1880
+ "Hardy-Littlewood-Polya, Inequalities (1934) [AM-GM]",
1881
+ "Cauchy (1821); Schwarz (1888)",
1882
+ "Vovk-Gammerman-Shafer (2005); Lei et al. (2018) JASA 113:1094 [conformal]",
1883
+ "Dirichlet (1834) [pigeonhole]",
1884
+ "Doob (1953) Stochastic Processes [optional stopping]"
1885
+ ]
1886
+ },
1887
+ "experimental_sorry_free_note": "wave5 adds 11 kernel-verified experimental theorems (6 Mathlib-dependent CI-green: C1/C2/C6 + W5-1/W5-1b/W5-2; 5 Mathlib-free bare-lean: W5-3a/b/c, W5-4, W5-5). Prior experimental_sorry_free baseline was 21 (wave-2 F-pack ceiling).",
1888
+ "wave5_proven_count": {
1889
+ "mathlib_dependent_ci_green": 6,
1890
+ "mathlib_free_bare_lean": 5,
1891
+ "total_new": 11
1892
+ },
1893
+ "wave6": {
1894
+ "campaign": "prove-wave-6 (graph substrate)",
1895
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/189",
1896
+ "commit_ci_green": "dc7ae26d",
1897
+ "ci_status": "GREEN (Lean kernel check + lake build+numbers + doctrine + DCO)",
1898
+ "count": 11,
1899
+ "axioms": "0 new",
1900
+ "new_proven_ids": [
1901
+ "F-G1",
1902
+ "F-G2",
1903
+ "F-G3",
1904
+ "F-G4",
1905
+ "F-G5",
1906
+ "F-G6"
1907
+ ],
1908
+ "headline": "Frechet/Kuratowski embedding core, GNN<=1-WL ceiling, spectral contraction, Lambda-graph iso-invariance, bounded-frontier DAG termination, relabel-invariance",
1909
+ "maturity": "experimental-CI-green",
1910
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; experimental scope"
1911
+ },
1912
+ "wave7": {
1913
+ "campaign": "prove-wave-7",
1914
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/190",
1915
+ "commit_ci_green": "d6a232ba",
1916
+ "ci_status": "GREEN",
1917
+ "count": 10,
1918
+ "axioms": "0 new",
1919
+ "new_proven_ci_green": [
1920
+ "W7-1a",
1921
+ "W7-1",
1922
+ "W7-5a",
1923
+ "W7-5b",
1924
+ "W7-5"
1925
+ ],
1926
+ "new_proven_bare_lean": [
1927
+ "W7-4a",
1928
+ "W7-4b",
1929
+ "W7-4c",
1930
+ "W7-6a",
1931
+ "W7-6"
1932
+ ],
1933
+ "headline": "conformal rank-count/p-value + Doob two-sided audit envelope + degree-sum iso-invariance + PAC-Bayes routing envelope",
1934
+ "maturity": "experimental-CI-green",
1935
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; experimental scope"
1936
+ },
1937
+ "agentic_loop": {
1938
+ "campaign": "prove-agentic-loop (the governed RAG->MCP->kernel->receipt loop, proven as a SYSTEM)",
1939
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/188",
1940
+ "commit_ci_green": "2ede47a2",
1941
+ "ci_status": "GREEN",
1942
+ "namespace": "Lutar.Agentic.Pipeline (EXPERIMENTAL_SCOPES; NOT imported into Lutar.lean)",
1943
+ "theorems": 28,
1944
+ "axiom_free": 14,
1945
+ "lean_core_only": 10,
1946
+ "axiom_gated": 4,
1947
+ "declared_axiom": "hashFn_collision_resistant (P5 only; NIST FIPS 180-4)",
1948
+ "properties": {
1949
+ "P1": "receipt-completeness",
1950
+ "P2": "gate-soundness",
1951
+ "P3": "non-interference (Goguen-Meseguer 1982, axiom-free core)",
1952
+ "P4": "replay-determinism (axiom-free)",
1953
+ "P5": "tamper-evidence (axiom-gated)",
1954
+ "P6": "monotone auditability"
1955
+ },
1956
+ "headline": "the RAG->MCP->kernel loop is proven end-to-end; P3 (poisoned input can't flip the verdict) is the Cannonico bullseye",
1957
+ "maturity": "experimental-CI-green (P5 axiom-gated)",
1958
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; experimental scope"
1959
+ },
1960
+ "coder_formulas": {
1961
+ "campaign": "prove-coder (a11oy Code governed coder formulas)",
1962
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/193",
1963
+ "commit_ci_green": "29e33534",
1964
+ "ci_status": "GREEN (bare-lean sorry-free + CI build/numbers/doctrine/DCO)",
1965
+ "theorems": 27,
1966
+ "axiom_free": 5,
1967
+ "lean_core_only": 23,
1968
+ "axiom_gated": 1,
1969
+ "declared_axiom": "codeHash_collision_resistant (1 only; standard, disclosed)",
1970
+ "areas": {
1971
+ "CS1": "sandbox containment (extends P2)",
1972
+ "CS2": "bounded exec/termination (extends F-G5)",
1973
+ "CR3": "router envelope + argmin stability (W7-5/C20)",
1974
+ "CV4": "consensus/Byzantine majority (C10)",
1975
+ "CC5": "conformal code-confidence <1 (W5-3/W7-4)",
1976
+ "CK6": "receipt-log compression (Kraft/Shannon C8/C9)",
1977
+ "NI7": "code-context non-interference (extends P3) \u2014 poisoned dependency can't flip DENY->ALLOW"
1978
+ },
1979
+ "headline": "27 theorems innovated for the governed coder; kernel-verified two ways",
1980
+ "maturity": "experimental-CI-green (1 axiom-gated)",
1981
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; experimental scope"
1982
+ },
1983
+ "lambda_setalpha_setdelta": {
1984
+ "campaign": "lambda-uniqueness Set alpha + Set delta (conditional uniqueness within strengthened axiom classes)",
1985
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/192",
1986
+ "commit_ci_green": "5f0bb5ee",
1987
+ "ci_status": "GREEN (build + lake+numbers + doctrine + DCO)",
1988
+ "results": 22,
1989
+ "headline_theorems_approx": 12,
1990
+ "declared_bridge_axioms": [
1991
+ "setAlpha_cauchy",
1992
+ "KS_theorem_1_1",
1993
+ "setDelta_stage2"
1994
+ ],
1995
+ "impostor_deaths_axiom_free": 10,
1996
+ "what_proven": "Lambda (geometric mean) is UNIQUE within Set alpha {A1,A2,A3,A4,A5' multiplicativity} (cond. on setAlpha_cauchy) and within Set delta {d1..d4,d5' multiplicativity} (cond. on KS_theorem_1_1+setDelta_stage2). All 10 impostor-deaths AXIOM-FREE.",
1997
+ "what_NOT_claimed": "NOT unconditional uniqueness under original A1-A5 (machine-checked FALSE: Round13.maxAgg_ne_Lambda). Lambda STAYS Conjecture 1.",
1998
+ "maturity": "conditional (axiom-gated bridge); Lambda = Conjecture 1 unconditionally",
1999
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; locked_proven STAYS 5"
2000
+ },
2001
+ "mathlib_bump_c3c4c5": {
2002
+ "campaign": "Mathlib v4.18 bump \u2014 concentration/KL re-exports C3/C4/C5",
2003
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/187",
2004
+ "ci_status": "PROVEN on the Mathlib-v4.18 bump branch (CI-green), PENDING MERGE to main",
2005
+ "ids": [
2006
+ "C3",
2007
+ "C4",
2008
+ "C5"
2009
+ ],
2010
+ "status_honest": "proven on bump branch (PR#187), pending merge to main \u2014 NOT on-main, NOT blocked",
2011
+ "maturity": "branch-pending"
2012
+ },
2013
+ "unify_governed_run_sound": {
2014
+ "campaign": "unify governance substrate meta-theorem (monoid-action spine unifying P1/P4/P6 + coder corpus)",
2015
+ "pull_request": "https://github.com/szl-holdings/lutar-lean/pull/194",
2016
+ "commit_ci_green": "9f9c1bbd",
2017
+ "artifact": "Lutar/Unify/GovernanceSubstrate.lean (EXPERIMENTAL scope; NOT wired into Lutar.lean)",
2018
+ "spine": "run is a left monoid action of the free monoid (List Hop,++,[]) on St; run_append homomorphism is the unifier",
2019
+ "theorems": [
2020
+ "run_nil (axiom-free)",
2021
+ "run_append",
2022
+ "run_singleton (axiom-free)",
2023
+ "completeness_additive",
2024
+ "determinism_composes",
2025
+ "chainEnd_append (axiom-free)",
2026
+ "auditability_multiplicative",
2027
+ "governed_run_sound"
2028
+ ],
2029
+ "headline": "every compositional corpus guarantee is a corollary of one homomorphism; a synthesis (unification) theorem, not new deep pure math",
2030
+ "maturity": "experimental-CI-green",
2031
+ "locked_kernel": "749/14/163 @ c7c0ba17 UNCHANGED; experimental scope"
2032
+ },
2033
+ "maturity_legend": [
2034
+ "locked",
2035
+ "experimental-CI-green",
2036
+ "branch-pending",
2037
+ "axiom-gated",
2038
+ "conditional",
2039
+ "conjecture"
2040
+ ],
2041
+ "experimental_total_note": "LOCKED proven = exactly 5 {F1,F11,F12,F18,F19} @ c7c0ba17 (749/14/163), UNCHANGED. PLUS 80+ experimental kernel-verified theorems (all CI-green, never folded into the locked 5): wave5 (11, PR#186), wave6 (11, PR#189), wave7 (10, PR#190), agentic-loop P1-P6 (28, PR#188), coder formulas (27, PR#193), Lambda Set alpha/delta (22 results / ~12 theorems, PR#192), unify governed_run_sound (PR#194). C3/C4/C5 proven on the Mathlib-v4.18 bump branch (PR#187), pending merge to main. Lambda = Conjecture 1 unconditionally (uniqueness machine-checked FALSE) + PROVEN conditionally under declared strengthened Set alpha/delta axioms (PR#192).",
2042
+ "experimental_count_min": 80
2043
+ },
2044
+ "puriq_formulas": [
2045
+ {
2046
+ "id": "F1",
2047
+ "name": "Replay-Hash Determinism",
2048
+ "statement": "Deterministic step-fold replay: equal logs replay to equal final state; folded state equals last of the explicit replay trace.",
2049
+ "maturity": "proven",
2050
+ "lean_ref": "f1_replay_fold_deterministic",
2051
+ "lean_repo": "szl-holdings/lutar-lean",
2052
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2053
+ "locked_kernel": true
2054
+ },
2055
+ {
2056
+ "id": "F2",
2057
+ "name": "Scheduler Liveness",
2058
+ "statement": "Fair round-robin scheduler: every ready organ eventually ticks (strictly-decreasing Nat ranking measure reaches 0).",
2059
+ "maturity": "proven",
2060
+ "lean_ref": "f2_scheduler_liveness",
2061
+ "lean_repo": "szl-holdings/lutar-lean",
2062
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2063
+ },
2064
+ {
2065
+ "id": "F3",
2066
+ "name": "Organ Boot-Gate Soundness",
2067
+ "statement": "If the boot gate permits an organ, its genome is valid (decidable implication).",
2068
+ "maturity": "proven",
2069
+ "lean_ref": "f3_genome_gate_sound",
2070
+ "lean_repo": "szl-holdings/lutar-lean",
2071
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2072
+ },
2073
+ {
2074
+ "id": "F4",
2075
+ "name": "Receipt-Chain Acyclicity",
2076
+ "statement": "Appending the new largest-index node preserves DAG acyclicity (backward-edge invariant).",
2077
+ "maturity": "proven",
2078
+ "lean_ref": "f4_khipu_dag_acyclic",
2079
+ "lean_repo": "szl-holdings/lutar-lean",
2080
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2081
+ },
2082
+ {
2083
+ "id": "F5",
2084
+ "name": "Unay Receipt Recall",
2085
+ "statement": "Insert-then-lookup on the same receipt key returns the inserted value (exact-key recall correctness).",
2086
+ "maturity": "proven",
2087
+ "lean_ref": "f5_unay_recall_correct",
2088
+ "lean_repo": "szl-holdings/lutar-lean",
2089
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2090
+ },
2091
+ {
2092
+ "id": "F6",
2093
+ "name": "LMDB Durability",
2094
+ "statement": "Commit-then-restart-then-read returns the committed value; uncommitted writes are lost on crash (WAL model).",
2095
+ "maturity": "proven",
2096
+ "lean_ref": "f6_lmdb_durability",
2097
+ "lean_repo": "szl-holdings/lutar-lean",
2098
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2099
+ },
2100
+ {
2101
+ "id": "F7",
2102
+ "name": "Chaski FIFO Ordering",
2103
+ "statement": "Enqueue-batch-then-drain yields send order; head is the oldest message (true FIFO, no tautology).",
2104
+ "maturity": "proven",
2105
+ "lean_ref": "f7_chaski_fifo",
2106
+ "lean_repo": "szl-holdings/lutar-lean",
2107
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2108
+ },
2109
+ {
2110
+ "id": "F8",
2111
+ "name": "Wallpa OSS-Only Safety",
2112
+ "statement": "Governed-voice admission gate admits only OSS sources; no humanClone or synthetic config is ever admitted.",
2113
+ "maturity": "proven",
2114
+ "lean_ref": "f8_wallpa_oss_only",
2115
+ "lean_repo": "szl-holdings/lutar-lean",
2116
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2117
+ },
2118
+ {
2119
+ "id": "F9",
2120
+ "name": "Wasi-Rikuq Non-Interference",
2121
+ "statement": "Advisory non-interference (Goguen-Meseguer 1982): the low view is unchanged by high inputs.",
2122
+ "maturity": "proven",
2123
+ "lean_ref": "f9_wasi_rikuq_noninterference",
2124
+ "lean_repo": "szl-holdings/lutar-lean",
2125
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2126
+ },
2127
+ {
2128
+ "id": "F10",
2129
+ "name": "Hatun MCP Idempotency",
2130
+ "statement": "MCP request normalizer is idempotent: normalizing twice equals normalizing once.",
2131
+ "maturity": "proven",
2132
+ "lean_ref": "f10_hatun_mcp_idempotent",
2133
+ "lean_repo": "szl-holdings/lutar-lean",
2134
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2135
+ },
2136
+ {
2137
+ "id": "F11",
2138
+ "name": "Ayni Reciprocity Conservation",
2139
+ "statement": "Event-sourcing replay invariant: balance reciprocity is conserved; tit-for-tat parity (Axelrod-Hamilton).",
2140
+ "maturity": "proven",
2141
+ "lean_ref": "f11_ayni_reciprocity_conservation",
2142
+ "lean_repo": "szl-holdings/lutar-lean",
2143
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2144
+ "locked_kernel": true
2145
+ },
2146
+ {
2147
+ "id": "F12",
2148
+ "name": "Kuramoto Phase-Coupling Boundedness",
2149
+ "statement": "Discrete additive coupling is bounded and superposes over an organ set. CAVEAT: additive fragment only, NOT nonlinear Kuramoto synchronisation.",
2150
+ "maturity": "proven",
2151
+ "lean_ref": "f12_kuramoto_superposition",
2152
+ "lean_repo": "szl-holdings/lutar-lean",
2153
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2154
+ "locked_kernel": true
2155
+ },
2156
+ {
2157
+ "id": "F13",
2158
+ "name": "Wayra Hash-Chain Verification",
2159
+ "statement": "Hash-chain verification is sound by induction: a verified chain has every link consistent.",
2160
+ "maturity": "proven",
2161
+ "lean_ref": "f13_wayra_chain_verified",
2162
+ "lean_repo": "szl-holdings/lutar-lean",
2163
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2164
+ "declared_axiom": "hash_collision_resistant (tamper-evidence f13_tamper_evident)"
2165
+ },
2166
+ {
2167
+ "id": "F14",
2168
+ "name": "Signature Verifiable Attribution",
2169
+ "statement": "A signature-envelope signature that verifies attributes the message to the key. AXIOM-GATED on declared `ecdsa_unforgeable`.",
2170
+ "maturity": "axiom-gated",
2171
+ "lean_ref": "f14_dsse_verifiable",
2172
+ "lean_repo": "szl-holdings/lutar-lean",
2173
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2174
+ "declared_axiom": "ecdsa_unforgeable"
2175
+ },
2176
+ {
2177
+ "id": "F15",
2178
+ "name": "Rekor Merkle Inclusion",
2179
+ "statement": "Merkle inclusion checker is sound (structural). Binding form (equal roots => equal leaves) is AXIOM-GATED on declared `h2_collision_resistant`.",
2180
+ "maturity": "proven",
2181
+ "lean_ref": "f15_rekor_inclusion",
2182
+ "lean_repo": "szl-holdings/lutar-lean",
2183
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2184
+ "declared_axiom": "h2_collision_resistant (binding form f15_inclusion_binding)"
2185
+ },
2186
+ {
2187
+ "id": "F16",
2188
+ "name": "Safety-Gate Coverage Completeness",
2189
+ "statement": "Immune cross-cut completeness: 8 gates cover all 8 enumerated threats; gate set is exhaustive.",
2190
+ "maturity": "proven",
2191
+ "lean_ref": "f16_safegate_immune_complete",
2192
+ "lean_repo": "szl-holdings/lutar-lean",
2193
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2194
+ },
2195
+ {
2196
+ "id": "F17",
2197
+ "name": "Three-Vertical Isolation",
2198
+ "statement": "The three verticals are pairwise disjoint (isolation by construction).",
2199
+ "maturity": "proven",
2200
+ "lean_ref": "f17_three_vertical_isolation",
2201
+ "lean_repo": "szl-holdings/lutar-lean",
2202
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2203
+ },
2204
+ {
2205
+ "id": "F18",
2206
+ "name": "Reed-Solomon RS(10,6) Recovery",
2207
+ "statement": "RS(10,6) parity arithmetic: data is recoverable iff at least 6 of 10 shards survive (tolerates 4 erasures).",
2208
+ "maturity": "proven",
2209
+ "lean_ref": "f18_reed_solomon_parity_count",
2210
+ "lean_repo": "szl-holdings/lutar-lean",
2211
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2212
+ "locked_kernel": true
2213
+ },
2214
+ {
2215
+ "id": "F19",
2216
+ "name": "Bekenstein Entropy Budget",
2217
+ "statement": "Entropy budget is additive and monotone over a region partition; each region <= total. CAVEAT: additive scaffolding only, NOT the full Bekenstein bound S <= 2*pi*k*R*E/(hbar*c).",
2218
+ "maturity": "proven",
2219
+ "lean_ref": "f19_budget_total_cons",
2220
+ "lean_repo": "szl-holdings/lutar-lean",
2221
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean",
2222
+ "locked_kernel": true
2223
+ },
2224
+ {
2225
+ "id": "F20",
2226
+ "name": "Mobile Input Equivalence",
2227
+ "statement": "Touch and pointer inputs are equivalent under the normalization map (decidable and sound).",
2228
+ "maturity": "proven",
2229
+ "lean_ref": "f20_mobile_input_equiv",
2230
+ "lean_repo": "szl-holdings/lutar-lean",
2231
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2232
+ },
2233
+ {
2234
+ "id": "F21",
2235
+ "name": "Genome Validator Totality",
2236
+ "statement": "The genome validator is total over Fin 16: every organ validates.",
2237
+ "maturity": "proven",
2238
+ "lean_ref": "f21_all_organs_valid",
2239
+ "lean_repo": "szl-holdings/lutar-lean",
2240
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2241
+ },
2242
+ {
2243
+ "id": "F22",
2244
+ "name": "Receipt Emit Monotonicity",
2245
+ "statement": "Emit appends to the sequence log with strictly increasing sequence numbers (monotone emit).",
2246
+ "maturity": "proven",
2247
+ "lean_ref": "f22_khipu_emit_monotone",
2248
+ "lean_repo": "szl-holdings/lutar-lean",
2249
+ "lean_file": "Lutar/Puriq/Formulas/PuriqFormulaLean.lean"
2250
+ },
2251
+ {
2252
+ "id": "F23",
2253
+ "name": "Lambda-Aggregator Uniqueness",
2254
+ "statement": "CONJECTURE 1 (NOT a theorem). Unconditional uniqueness is FALSE under A1-A5 (maxAgg counterexample). Conditional `lambda_unique_of_factors` IS proved; unconditional uniqueness closes only under declared axiom A6_bisymmetric.",
2255
+ "maturity": "conjectured",
2256
+ "lean_ref": "f23_lambda_aggregator_sound",
2257
+ "lean_repo": "szl-holdings/lutar-lean",
2258
+ "lean_file": "Lutar/Puriq/Formulas/F23_Uniqueness.lean",
2259
+ "declared_axiom": "A6_bisymmetric (optional, only for conditional lambda_unique_under_A6)"
2260
+ }
2261
+ ],
2262
+ "vertical_policies": [
2263
+ {
2264
+ "policy_id": "academic",
2265
+ "policy_name": "Academic / Research Integrity",
2266
+ "version": "0.3.0",
2267
+ "regulations": [
2268
+ "NIH NOT-OD-23-149",
2269
+ "NSF PAPPG Chapter II.E",
2270
+ "ORI Standards",
2271
+ "COPE Guidelines"
2272
+ ],
2273
+ "required_attestors": [
2274
+ "principal_investigator",
2275
+ "research_integrity_officer"
2276
+ ],
2277
+ "lambda_floors": {
2278
+ "measurabilityHonesty": 1.0,
2279
+ "constructiveTransparency": 1.0,
2280
+ "informationIntegrity": 0.99,
2281
+ "temporalConsistency": 0.99
2282
+ },
2283
+ "forbidden_inputs": [
2284
+ "undisclosed_ai_authored_content",
2285
+ "missing_doi_citation"
2286
+ ],
2287
+ "required_output_formats": [
2288
+ "zenodo_deposit",
2289
+ "json_receipt",
2290
+ "orcid_linked_artifact"
2291
+ ],
2292
+ "retention_days": "3650",
2293
+ "primitives_applicable": [
2294
+ "A5",
2295
+ "A8",
2296
+ "A12",
2297
+ "T5",
2298
+ "T10",
2299
+ "TH2"
2300
+ ],
2301
+ "acv_range_usd": {
2302
+ "low": 10000.0,
2303
+ "mid": 50000.0,
2304
+ "high": 200000.0
2305
+ }
2306
+ },
2307
+ {
2308
+ "policy_id": "capital_markets",
2309
+ "policy_name": "Capital Markets / Quant / Hedge Funds",
2310
+ "version": "0.3.0",
2311
+ "regulations": [
2312
+ "SEC Rule 17a-4",
2313
+ "MiFID II RTS 6",
2314
+ "FINRA Rule 4370",
2315
+ "Reg SCI 17 CFR 242"
2316
+ ],
2317
+ "required_attestors": [
2318
+ "chief_compliance_officer",
2319
+ "quant_review_committee",
2320
+ "external_auditor"
2321
+ ],
2322
+ "lambda_floors": {
2323
+ "moralGrounding": 0.97,
2324
+ "measurabilityHonesty": 1.0,
2325
+ "temporalConsistency": 1.0,
2326
+ "economicGrounding": 1.0,
2327
+ "constructiveTransparency": 0.99,
2328
+ "informationIntegrity": 0.99
2329
+ },
2330
+ "forbidden_inputs": [
2331
+ "non_sec_registered_model",
2332
+ "missing_algo_documentation"
2333
+ ],
2334
+ "required_output_formats": [
2335
+ "sec_17a4_compliant_log",
2336
+ "json_receipt",
2337
+ "worm_storage_manifest"
2338
+ ],
2339
+ "retention_days": "2190",
2340
+ "primitives_applicable": [
2341
+ "A5",
2342
+ "A6",
2343
+ "A10",
2344
+ "A12",
2345
+ "A14",
2346
+ "T5",
2347
+ "T9",
2348
+ "TH2"
2349
+ ],
2350
+ "acv_range_usd": {
2351
+ "low": 500000.0,
2352
+ "mid": 2000000.0,
2353
+ "high": 10000000.0
2354
+ }
2355
+ },
2356
+ {
2357
+ "policy_id": "critical_infrastructure",
2358
+ "policy_name": "Critical Infrastructure / Utilities",
2359
+ "version": "0.3.0",
2360
+ "regulations": [
2361
+ "NERC CIP-013-2",
2362
+ "IEC 62443-3-3",
2363
+ "TSA Pipeline Security Directive SD-02C",
2364
+ "NIST CSF 2.0"
2365
+ ],
2366
+ "required_attestors": [
2367
+ "system_security_officer",
2368
+ "control_systems_engineer",
2369
+ "incident_commander"
2370
+ ],
2371
+ "lambda_floors": {
2372
+ "moralGrounding": 0.99,
2373
+ "actionReversibility": 0.99,
2374
+ "scopeContainment": 1.0,
2375
+ "informationIntegrity": 0.99,
2376
+ "adversarialRobustness": 0.99,
2377
+ "temporalConsistency": 0.99
2378
+ },
2379
+ "forbidden_inputs": [
2380
+ "unauthenticated_control_commands",
2381
+ "non_air_gapped_ot_data"
2382
+ ],
2383
+ "required_output_formats": [
2384
+ "ics_audit_log",
2385
+ "json_receipt",
2386
+ "nerc_compliance_report"
2387
+ ],
2388
+ "retention_days": "1825",
2389
+ "primitives_applicable": [
2390
+ "A4",
2391
+ "A5",
2392
+ "A6",
2393
+ "A10",
2394
+ "A13",
2395
+ "T5",
2396
+ "T9",
2397
+ "T10",
2398
+ "TH1"
2399
+ ],
2400
+ "acv_range_usd": {
2401
+ "low": 1000000.0,
2402
+ "mid": 5000000.0,
2403
+ "high": 20000000.0
2404
+ }
2405
+ },
2406
+ {
2407
+ "policy_id": "defense",
2408
+ "policy_name": "Defense / DoD",
2409
+ "version": "0.3.0",
2410
+ "regulations": [
2411
+ "NIST SP 800-53 Rev 5",
2412
+ "CMMC 2.0",
2413
+ "FedRAMP High",
2414
+ "DISA STIGs"
2415
+ ],
2416
+ "required_attestors": [
2417
+ "authorizing_official",
2418
+ "system_owner",
2419
+ "security_control_assessor"
2420
+ ],
2421
+ "lambda_floors": {
2422
+ "moralGrounding": 0.99,
2423
+ "measurabilityHonesty": 0.99,
2424
+ "actionReversibility": 0.95,
2425
+ "scopeContainment": 0.99,
2426
+ "informationIntegrity": 0.99,
2427
+ "consentBoundary": 0.95
2428
+ },
2429
+ "forbidden_inputs": [
2430
+ "unclassified_cui_without_marking",
2431
+ "foreign_national_data"
2432
+ ],
2433
+ "required_output_formats": [
2434
+ "json_audit_log",
2435
+ "nist_oscal"
2436
+ ],
2437
+ "retention_days": "7300",
2438
+ "primitives_applicable": [
2439
+ "A1",
2440
+ "A4",
2441
+ "A5",
2442
+ "A6",
2443
+ "A8",
2444
+ "A9",
2445
+ "A12",
2446
+ "A13",
2447
+ "T5",
2448
+ "T9",
2449
+ "T10",
2450
+ "TH1"
2451
+ ],
2452
+ "acv_range_usd": {
2453
+ "low": 500000.0,
2454
+ "mid": 2000000.0,
2455
+ "high": 5000000.0
2456
+ }
2457
+ },
2458
+ {
2459
+ "policy_id": "financial_services",
2460
+ "policy_name": "Financial Services / Banking",
2461
+ "version": "0.3.0",
2462
+ "regulations": [
2463
+ "SR 11-7",
2464
+ "OCC 2011-12",
2465
+ "MiFID II RTS 6",
2466
+ "Basel III"
2467
+ ],
2468
+ "required_attestors": [
2469
+ "model_risk_officer",
2470
+ "chief_risk_officer",
2471
+ "internal_audit"
2472
+ ],
2473
+ "lambda_floors": {
2474
+ "moralGrounding": 0.97,
2475
+ "measurabilityHonesty": 0.99,
2476
+ "temporalConsistency": 0.99,
2477
+ "informationIntegrity": 0.99,
2478
+ "economicGrounding": 1.0
2479
+ },
2480
+ "forbidden_inputs": [
2481
+ "insider_information",
2482
+ "unregistered_model_version"
2483
+ ],
2484
+ "required_output_formats": [
2485
+ "csv_model_log",
2486
+ "json_receipt",
2487
+ "pdf_board_report"
2488
+ ],
2489
+ "retention_days": "2190",
2490
+ "primitives_applicable": [
2491
+ "A1",
2492
+ "A5",
2493
+ "A6",
2494
+ "A8",
2495
+ "A9",
2496
+ "A14",
2497
+ "T5",
2498
+ "T9",
2499
+ "T10",
2500
+ "TH1",
2501
+ "TH2"
2502
+ ],
2503
+ "acv_range_usd": {
2504
+ "low": 200000.0,
2505
+ "mid": 800000.0,
2506
+ "high": 2000000.0
2507
+ }
2508
+ },
2509
+ {
2510
+ "policy_id": "healthcare",
2511
+ "policy_name": "Healthcare / Clinical AI",
2512
+ "version": "0.3.0",
2513
+ "regulations": [
2514
+ "HIPAA 45 CFR Part 164",
2515
+ "FDA 21 CFR Part 11",
2516
+ "FDA SaMD Guidance Q3 2023"
2517
+ ],
2518
+ "required_attestors": [
2519
+ "licensed_clinician",
2520
+ "clinical_informatics_officer",
2521
+ "hipaa_privacy_officer"
2522
+ ],
2523
+ "lambda_floors": {
2524
+ "moralGrounding": 0.99,
2525
+ "measurabilityHonesty": 0.99,
2526
+ "consentBoundary": 0.99,
2527
+ "informationIntegrity": 0.99,
2528
+ "causalSeparability": 0.99
2529
+ },
2530
+ "forbidden_inputs": [
2531
+ "plaintext_phi",
2532
+ "deidentification_not_verified"
2533
+ ],
2534
+ "required_output_formats": [
2535
+ "hl7_fhir_audit",
2536
+ "json_receipt",
2537
+ "pdf_clinical_audit"
2538
+ ],
2539
+ "retention_days": "2190",
2540
+ "primitives_applicable": [
2541
+ "A1",
2542
+ "A4",
2543
+ "A5",
2544
+ "A8",
2545
+ "A9",
2546
+ "A11",
2547
+ "A12",
2548
+ "T5",
2549
+ "T7",
2550
+ "T10",
2551
+ "TH1"
2552
+ ],
2553
+ "acv_range_usd": {
2554
+ "low": 150000.0,
2555
+ "mid": 600000.0,
2556
+ "high": 2000000.0
2557
+ }
2558
+ },
2559
+ {
2560
+ "policy_id": "insurance",
2561
+ "policy_name": "Insurance",
2562
+ "version": "0.3.0",
2563
+ "regulations": [
2564
+ "NAIC Model Law 881",
2565
+ "NY DFS Circular Letter 7 (2022)",
2566
+ "NAIC AI Principles (2020)"
2567
+ ],
2568
+ "required_attestors": [
2569
+ "chief_actuary",
2570
+ "ai_ethics_board",
2571
+ "compliance_officer"
2572
+ ],
2573
+ "lambda_floors": {
2574
+ "moralGrounding": 0.97,
2575
+ "measurabilityHonesty": 0.99,
2576
+ "informationIntegrity": 0.99,
2577
+ "constructiveTransparency": 0.99,
2578
+ "economicGrounding": 0.97
2579
+ },
2580
+ "forbidden_inputs": [
2581
+ "prohibited_rating_factors",
2582
+ "non_actuarially_justified_proxies"
2583
+ ],
2584
+ "required_output_formats": [
2585
+ "csv_underwriting_log",
2586
+ "json_receipt",
2587
+ "pdf_state_filing"
2588
+ ],
2589
+ "retention_days": "1825",
2590
+ "primitives_applicable": [
2591
+ "A1",
2592
+ "A5",
2593
+ "A8",
2594
+ "A9",
2595
+ "A12",
2596
+ "A14",
2597
+ "T6",
2598
+ "T9",
2599
+ "T10"
2600
+ ],
2601
+ "acv_range_usd": {
2602
+ "low": 200000.0,
2603
+ "mid": 750000.0,
2604
+ "high": 2000000.0
2605
+ }
2606
+ },
2607
+ {
2608
+ "policy_id": "legal",
2609
+ "policy_name": "Legal / e-Discovery",
2610
+ "version": "0.3.0",
2611
+ "regulations": [
2612
+ "FRCP 26",
2613
+ "FRCP 34",
2614
+ "FRE 902(13)",
2615
+ "FRE 902(14)",
2616
+ "ABA Model Rule 1.1"
2617
+ ],
2618
+ "required_attestors": [
2619
+ "supervising_attorney",
2620
+ "records_custodian"
2621
+ ],
2622
+ "lambda_floors": {
2623
+ "moralGrounding": 0.97,
2624
+ "measurabilityHonesty": 0.99,
2625
+ "informationIntegrity": 0.99,
2626
+ "constructiveTransparency": 0.99,
2627
+ "actionReversibility": 0.95
2628
+ },
2629
+ "forbidden_inputs": [
2630
+ "privileged_attorney_client_without_waiver"
2631
+ ],
2632
+ "required_output_formats": [
2633
+ "json_chain_of_custody",
2634
+ "pdf_court_exhibit"
2635
+ ],
2636
+ "retention_days": "2555",
2637
+ "primitives_applicable": [
2638
+ "A5",
2639
+ "A6",
2640
+ "A8",
2641
+ "A12",
2642
+ "T5",
2643
+ "T8",
2644
+ "T10",
2645
+ "TH2"
2646
+ ],
2647
+ "acv_range_usd": {
2648
+ "low": 75000.0,
2649
+ "mid": 300000.0,
2650
+ "high": 1000000.0
2651
+ }
2652
+ },
2653
+ {
2654
+ "policy_id": "pharma",
2655
+ "policy_name": "Pharma / Life Sciences R&D",
2656
+ "version": "0.3.0",
2657
+ "regulations": [
2658
+ "FDA 21 CFR Part 11",
2659
+ "EMA Annex 11",
2660
+ "ICH E6(R3) GCP",
2661
+ "GxP"
2662
+ ],
2663
+ "required_attestors": [
2664
+ "qualified_person",
2665
+ "gxp_compliance_officer",
2666
+ "computational_scientist"
2667
+ ],
2668
+ "lambda_floors": {
2669
+ "moralGrounding": 0.97,
2670
+ "measurabilityHonesty": 1.0,
2671
+ "informationIntegrity": 1.0,
2672
+ "temporalConsistency": 0.99,
2673
+ "constructiveTransparency": 1.0
2674
+ },
2675
+ "forbidden_inputs": [
2676
+ "non_gxp_validated_software_output",
2677
+ "unversioned_model"
2678
+ ],
2679
+ "required_output_formats": [
2680
+ "ectd_submission_package",
2681
+ "json_audit_trail",
2682
+ "csv_gxp_log"
2683
+ ],
2684
+ "retention_days": "3650",
2685
+ "primitives_applicable": [
2686
+ "A5",
2687
+ "A6",
2688
+ "A8",
2689
+ "A10",
2690
+ "A12",
2691
+ "T5",
2692
+ "TH2"
2693
+ ],
2694
+ "acv_range_usd": {
2695
+ "low": 500000.0,
2696
+ "mid": 2000000.0,
2697
+ "high": 10000000.0
2698
+ }
2699
+ },
2700
+ {
2701
+ "policy_id": "public_sector",
2702
+ "policy_name": "Public Sector / Civic AI",
2703
+ "version": "0.3.0",
2704
+ "regulations": [
2705
+ "EU AI Act Annex III",
2706
+ "NYC Local Law 144 (2023)",
2707
+ "NIST AI RMF 1.0",
2708
+ "OMB M-24-10"
2709
+ ],
2710
+ "required_attestors": [
2711
+ "agency_ai_officer",
2712
+ "civil_rights_officer",
2713
+ "inspector_general"
2714
+ ],
2715
+ "lambda_floors": {
2716
+ "moralGrounding": 0.99,
2717
+ "measurabilityHonesty": 0.99,
2718
+ "stakeholderAlignment": 0.99,
2719
+ "constructiveTransparency": 1.0,
2720
+ "adversarialRobustness": 0.95
2721
+ },
2722
+ "forbidden_inputs": [
2723
+ "biometric_data_without_explicit_consent",
2724
+ "prohibited_social_scoring"
2725
+ ],
2726
+ "required_output_formats": [
2727
+ "json_public_audit_log",
2728
+ "csv_bias_audit",
2729
+ "pdf_annual_report"
2730
+ ],
2731
+ "retention_days": "3650",
2732
+ "primitives_applicable": [
2733
+ "A1",
2734
+ "A5",
2735
+ "A8",
2736
+ "A12",
2737
+ "A13",
2738
+ "T6",
2739
+ "T10",
2740
+ "TH1",
2741
+ "TH3"
2742
+ ],
2743
+ "acv_range_usd": {
2744
+ "low": 300000.0,
2745
+ "mid": 1200000.0,
2746
+ "high": 5000000.0
2747
+ }
2748
+ }
2749
+ ],
2750
+ "instill_wave": "build-wave (locked 5 + wave3 19 sorry-free/4 axiom-gated + wave5 6 Mathlib-CI-green/5 bare-lean + conditional Lambda)"
2751
  }
pages/console.html CHANGED
The diff for this file is too large to render. See raw diff