feat(ms-ai-architect): G1 LUKKET — v3.1 judge MÅLT + ADOPTERT (P100/R100/0FP/0FN, max-utfall)

45-veis v3.1 fan-out kjørt per runbook: 255 claims / 45 filer, 45 Opus-4.8-xhigh-
subagenter (live MS Learn, blinde for gull, én per fil), aggregert 255/255 rent →
deterministisk re-score mot G5b-korrigert gull.

Resultat: v3.1 = P 100,0 / R 100,0 / 0 FP / 0 FN / F1 1,000 (TP 42, TN 198) mot
v3-baren P 100,0 / R 92,9 / 3 FN. Alle 3 gjenstående FN fanget (R1 øvre-grense,
R7 last-bærende-streng, R8 fler-delt) UTEN én ny FP. Forhåndsregistrert gate
(hold P=100 ∧ løft R>92,9) klarert → v3.1 ADOPTERT som judge.

R1-«+»-floor-flagg over full populasjon avkreftet som FP-risiko: alle matchet gull
(decision-changing floors = outdated/TP; tette floors = correct/TN). §8 G1 lukket,
G2 avblokkert (v3.1 = prompt å wire i transform.mjs). Suite 641/641. [skip-docs]

scripts/kb-eval/data/judge-bakeoff-report-v3.1.json

@@ -0,0 +1,234 @@
+{
+  "_meta": {
+    "source": "gold-correctness-set.json + judge-bakeoff-results.json",
+    "thresholds": {
+      "minRecall": 0.7,
+      "minPrecision": 0.6
+    },
+    "judged": 255
+  },
+  "population": {
+    "total": 255,
+    "verifiable": 240,
+    "positives": 42,
+    "negatives": 198,
+    "unsourcedInP": 15
+  },
+  "arms": {
+    "staleness": {
+      "tp": 0,
+      "fp": 0,
+      "fn": 42,
+      "tn": 198,
+      "positives": 42,
+      "negatives": 198,
+      "flagged": 0,
+      "precision": null,
+      "recall": 0,
+      "f1": null,
+      "recallWilson": {
+        "p": 0,
+        "low": 0,
+        "high": 0.08380161250916199
+      },
+      "precisionWilson": null
+    },
+    "judge": {
+      "tp": 42,
+      "fp": 0,
+      "fn": 0,
+      "tn": 198,
+      "positives": 42,
+      "negatives": 198,
+      "flagged": 42,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.9161983874908382,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.9161983874908382,
+        "high": 1
+      }
+    },
+    "hybrid": {
+      "tp": 42,
+      "fp": 0,
+      "fn": 0,
+      "tn": 198,
+      "positives": 42,
+      "negatives": 198,
+      "flagged": 42,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.9161983874908382,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.9161983874908382,
+        "high": 1
+      }
+    }
+  },
+  "sourceSilent": {
+    "onVerifiableNegative": 0,
+    "onVerifiableError": 0,
+    "agreesWithUnsourced": 2,
+    "disagreesWithUnsourced": 13
+  },
+  "byClaimType": {
+    "version": {
+      "tp": 7,
+      "fp": 0,
+      "fn": 0,
+      "tn": 21,
+      "positives": 7,
+      "negatives": 21,
+      "flagged": 7,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.6456611570247934,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.6456611570247934,
+        "high": 1
+      }
+    },
+    "tpm": {
+      "tp": 5,
+      "fp": 0,
+      "fn": 0,
+      "tn": 20,
+      "positives": 5,
+      "negatives": 20,
+      "flagged": 5,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.5655085052479191,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.5655085052479191,
+        "high": 1
+      }
+    },
+    "region": {
+      "tp": 2,
+      "fp": 0,
+      "fn": 0,
+      "tn": 13,
+      "positives": 2,
+      "negatives": 13,
+      "flagged": 2,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.34237195288961925,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.34237195288961925,
+        "high": 1
+      }
+    },
+    "status": {
+      "tp": 10,
+      "fp": 0,
+      "fn": 0,
+      "tn": 43,
+      "positives": 10,
+      "negatives": 43,
+      "flagged": 10,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.7224598312333834,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.7224598312333834,
+        "high": 1
+      }
+    },
+    "taxonomy": {
+      "tp": 11,
+      "fp": 0,
+      "fn": 0,
+      "tn": 86,
+      "positives": 11,
+      "negatives": 86,
+      "flagged": 11,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.7411599827511859,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.7411599827511859,
+        "high": 1
+      }
+    },
+    "sku": {
+      "tp": 7,
+      "fp": 0,
+      "fn": 0,
+      "tn": 15,
+      "positives": 7,
+      "negatives": 15,
+      "flagged": 7,
+      "precision": 1,
+      "recall": 1,
+      "f1": 1,
+      "recallWilson": {
+        "p": 1,
+        "low": 0.6456611570247934,
+        "high": 1
+      },
+      "precisionWilson": {
+        "p": 1,
+        "low": 0.6456611570247934,
+        "high": 1
+      }
+    }
+  },
+  "gate": {
+    "pass": true,
+    "recallOk": true,
+    "precisionOk": true,
+    "beatsStaleness": true,
+    "thresholds": {
+      "minRecall": 0.7,
+      "minPrecision": 0.6
+    },
+    "reasons": [
+      "all criteria met"
+    ]
+  }
+}

scripts/kb-eval/data/judge-bakeoff-report-v3.1.md

@@ -0,0 +1,54 @@
+# Judge bake-off-rapport — S1 (Fase 3 de-risk)
+
+_Generert deterministisk av `run-judge-bakeoff.mjs` over `gold-correctness-set.json` + `judge-bakeoff-results.json`. Tall fra testet `lib/judge-bakeoff.mjs`. Ikke rediger for hånd — regenerer._
+
+**Forhåndsregistrert gate (låst FØR fan-out):** recall ≥ 0.7, presisjon ≥ 0.6, OG judge-recall > staleness-recall.
+
+## Evaluerings-populasjon (P)
+
+Volatil stratum + fetchbare claim_types (price ekskludert) — der feilene bor; unngår «invertert leverage».
+
+| metrikk | verdi |
+|---|---|
+| P totalt | 255 |
+| Verifiserbare (correct/outdated/wrong) | 240 |
+| Positive (reelle feil å fange) | 42 |
+| Negative (correct) | 198 |
+| Unsourced i P (kjørt, men utenfor P/R) | 15 |
+
+## Arm-sammenligning (detektering over de 240 verifiserbare)
+
+| arm | TP | FP | FN | TN | presisjon | recall | recall Wilson 95% | F1 |
+|---|---|---|---|---|---|---|---|---|
+| staleness (billig baseline) | 0 | 0 | 42 | 198 | n/a | 0.0% | [0.0%, 8.4%] | n/a |
+| judge (per-påstand groundedness) | 42 | 0 | 0 | 198 | 100.0% | 100.0% | [91.6%, 100.0%] | 1.000 |
+| hybrid (union) | 42 | 0 | 0 | 198 | 100.0% | 100.0% | [91.6%, 100.0%] | 1.000 |
+
+## Judge per claim_type (verifiserbar delmengde)
+
+| claim_type | positive | TP | FP | FN | presisjon | recall |
+|---|---|---|---|---|---|---|
+| taxonomy | 11 | 11 | 0 | 0 | 100.0% | 100.0% |
+| status | 10 | 10 | 0 | 0 | 100.0% | 100.0% |
+| version | 7 | 7 | 0 | 0 | 100.0% | 100.0% |
+| sku | 7 | 7 | 0 | 0 | 100.0% | 100.0% |
+| tpm | 5 | 5 | 0 | 0 | 100.0% | 100.0% |
+| region | 2 | 2 | 0 | 0 | 100.0% | 100.0% |
+
+## source_silent-diagnostikk
+
+Judgen hentet siden men fant ikke verdien. Diagnostisk, ikke et flagg.
+
+| signal | antall | tolkning |
+|---|---|---|
+| På verifiserbar feil | 0 | judge-bom: reell feil oversett via «kan ikke verifisere» |
+| På verifiserbar correct | 0 | judge reproduserte ikke et korrekt faktum mennesket fant |
+| Enig med unsourced | 2 | judge reproduserer den uverifiserbare grensen (godt) |
+| Uenig med unsourced | 13 | judge hevdet grunnet/ugrunnet der mennesket ikke fant kilde |
+
+## GATE: ✅ PASS — bygg S3
+
+- recall 1.000 ≥ 0.7? **ja**
+- presisjon 1.000 ≥ 0.6? **ja**
+- slår staleness (recall 0.000)? **ja**
+- begrunnelse: all criteria met