ms-ai-architect/scripts/kb-eval/data
Kjell Tore Guttormsen 2935039f44 feat(ms-ai-architect): R7.2 stempling — 9/11 pass born-verified etter re-judge-gate; 2 diskvalifisert → flagged (R11) [skip-docs]
Re-judge-gate (frossen v3.1, 11 blinde Opus-4.8-xhigh subagenter, live MS Learn-fetch,
G6 Layer A dekket alle fetches). Kvalitetsgate på irreversibelt steg: born-verified stempel
krever TO uavhengige alt-grounded pass.

9 ENIGHET (alle claims grounded) → stemplet Verified 2026-07-24 / judge-v3.1:
  apim-azure-front-door-ai, apim-vs-direct-access-comparison, cost-tracking-apim-policies,
  developer-portal-ai-apis, multi-region-ai-gateway-design, streaming-support-apim,
  versioning-ai-api-endpoints, data-versioning-lineage, real-time-streaming-ai

2 UENIGHET (≥1 non-grounded) → pass→flagged, IKKE stemplet (R11 work-list):
  - security-hardening-ai-gateway: #2 source_silent («20+ security policies» usourcet),
    #7 not_grounded R8 (feature = Prompt Injection Protection, ikke «Prompt Shields»;
    indirekte injection-deteksjon ikke dekket)
  - etl-vs-elt-ai: #2 not_grounded R8 (Gold-lag = curated for reports/dashboards, ikke ML-features)

Gaten fanget gull/judge-run-uenighet (jf. gull-friskhet kan invertere adopsjon). Ledger:
94 records, R7.2 9 pass / 40 flagged (validateManifest valid). Stempel = kun Verified/
Verified by-header (body byte-identisk, per-fil invariant-sjekket, atomicWrite). Suite 947/947.
2026-07-24 20:14:45 +02:00
..
fase0-returns feat(ms-ai-architect): Fase 0 steg 2 — verifiserings-fan-out (373 påstander) + gull-sett m/ deterministisk lastmod-join + TDD [skip-docs] 2026-06-26 15:16:10 +02:00
base-rate-report.json feat(ms-ai-architect): Fase 0 steg 3 — base-rate-rapport + GATE: BYGG Fase 3 (scoped/hybrid) [skip-docs] 2026-06-26 17:15:54 +02:00
base-rate-report.md feat(ms-ai-architect): Fase 0 steg 3 — base-rate-rapport + GATE: BYGG Fase 3 (scoped/hybrid) [skip-docs] 2026-06-26 17:15:54 +02:00
eval-baseline.json feat(ms-ai-architect): Sesjon 15 — B2 K10 søsken-scope-ikke-overlapp 2026-06-20 11:59:59 +02:00
fase0-sample-frame.json feat(ms-ai-architect): Fase 0 steg 1 — deterministisk sample-frame (45 volatil + 10 kontroll) + TDD [skip-docs] 2026-06-26 14:18:53 +02:00
gold-correctness-set.json feat(ms-ai-architect): G5b gull-friskhets-spot-sjekk LUKKET — 4 v3-FP re-adjudert mot live, ALLE stale gull (v3 flagget korrekt), baseline løftet v3 P89.7/R92.1 → P100/R92.9/0FP; v3.1 forfattet (ren recall-hardning, FP-vakt droppet) [skip-docs] 2026-06-30 13:43:52 +02:00
judge-bakeoff-claims.json feat(ms-ai-architect): S1 judge bake-off harness + forhåndsregistrert gate (TDD) [skip-docs] 2026-06-26 20:10:58 +02:00
judge-bakeoff-report-v2-g5bgold.json feat(ms-ai-architect): G5b gull-friskhets-spot-sjekk LUKKET — 4 v3-FP re-adjudert mot live, ALLE stale gull (v3 flagget korrekt), baseline løftet v3 P89.7/R92.1 → P100/R92.9/0FP; v3.1 forfattet (ren recall-hardning, FP-vakt droppet) [skip-docs] 2026-06-30 13:43:52 +02:00
judge-bakeoff-report-v2-g5bgold.md feat(ms-ai-architect): G5b gull-friskhets-spot-sjekk LUKKET — 4 v3-FP re-adjudert mot live, ALLE stale gull (v3 flagget korrekt), baseline løftet v3 P89.7/R92.1 → P100/R92.9/0FP; v3.1 forfattet (ren recall-hardning, FP-vakt droppet) [skip-docs] 2026-06-30 13:43:52 +02:00
judge-bakeoff-report-v2-g5gold.json feat(ms-ai-architect): G5 gull-friskhets-mikropass LUKKET — 2 gull-feil rettet (genaiops#2, model-selection#8), v3 REHABILITERT — slår v2 på fersk gull (P 89.7/R 92.1 vs 86.8/86.8); adopsjonsbeslutning reversert [skip-docs] 2026-06-30 11:21:45 +02:00
judge-bakeoff-report-v2-g5gold.md feat(ms-ai-architect): G5 gull-friskhets-mikropass LUKKET — 2 gull-feil rettet (genaiops#2, model-selection#8), v3 REHABILITERT — slår v2 på fersk gull (P 89.7/R 92.1 vs 86.8/86.8); adopsjonsbeslutning reversert [skip-docs] 2026-06-30 11:21:45 +02:00
judge-bakeoff-report-v2-reconciled.json feat(ms-ai-architect): Spor 2b gull-rekonsiliering — 12 uenigheter løst, fasit herdet (P 84→92%, R 84→88%) [skip-docs] 2026-06-30 07:20:21 +02:00
judge-bakeoff-report-v2-reconciled.md feat(ms-ai-architect): Spor 2b gull-rekonsiliering — 12 uenigheter løst, fasit herdet (P 84→92%, R 84→88%) [skip-docs] 2026-06-30 07:20:21 +02:00
judge-bakeoff-report-v2.json feat(ms-ai-architect): S1 v2 targeted iteration — GATE PASS (recall 84.2%, precision 84.2%) [skip-docs] 2026-06-26 21:23:50 +02:00
judge-bakeoff-report-v2.md feat(ms-ai-architect): S1 v2 targeted iteration — GATE PASS (recall 84.2%, precision 84.2%) [skip-docs] 2026-06-26 21:23:50 +02:00
judge-bakeoff-report-v3-g5bgold.json feat(ms-ai-architect): G5b gull-friskhets-spot-sjekk LUKKET — 4 v3-FP re-adjudert mot live, ALLE stale gull (v3 flagget korrekt), baseline løftet v3 P89.7/R92.1 → P100/R92.9/0FP; v3.1 forfattet (ren recall-hardning, FP-vakt droppet) [skip-docs] 2026-06-30 13:43:52 +02:00
judge-bakeoff-report-v3-g5bgold.md feat(ms-ai-architect): G5b gull-friskhets-spot-sjekk LUKKET — 4 v3-FP re-adjudert mot live, ALLE stale gull (v3 flagget korrekt), baseline løftet v3 P89.7/R92.1 → P100/R92.9/0FP; v3.1 forfattet (ren recall-hardning, FP-vakt droppet) [skip-docs] 2026-06-30 13:43:52 +02:00
judge-bakeoff-report-v3-g5gold.json feat(ms-ai-architect): G5 gull-friskhets-mikropass LUKKET — 2 gull-feil rettet (genaiops#2, model-selection#8), v3 REHABILITERT — slår v2 på fersk gull (P 89.7/R 92.1 vs 86.8/86.8); adopsjonsbeslutning reversert [skip-docs] 2026-06-30 11:21:45 +02:00
judge-bakeoff-report-v3-g5gold.md feat(ms-ai-architect): G5 gull-friskhets-mikropass LUKKET — 2 gull-feil rettet (genaiops#2, model-selection#8), v3 REHABILITERT — slår v2 på fersk gull (P 89.7/R 92.1 vs 86.8/86.8); adopsjonsbeslutning reversert [skip-docs] 2026-06-30 11:21:45 +02:00
judge-bakeoff-report-v3.1.json feat(ms-ai-architect): G1 LUKKET — v3.1 judge MÅLT + ADOPTERT (P100/R100/0FP/0FN, max-utfall) 2026-06-30 21:33:45 +02:00
judge-bakeoff-report-v3.1.md feat(ms-ai-architect): G1 LUKKET — v3.1 judge MÅLT + ADOPTERT (P100/R100/0FP/0FN, max-utfall) 2026-06-30 21:33:45 +02:00
judge-bakeoff-report-v3.json feat(ms-ai-architect): G1 v3 bake-off MÅLT — v3 regredierte (P 92.1→89.7, R flat), v2 beholdt; G5 gull-friskhet åpnet [skip-docs] 2026-06-30 10:54:39 +02:00
judge-bakeoff-report-v3.md feat(ms-ai-architect): G1 v3 bake-off MÅLT — v3 regredierte (P 92.1→89.7, R flat), v2 beholdt; G5 gull-friskhet åpnet [skip-docs] 2026-06-30 10:54:39 +02:00
judge-bakeoff-report.json feat(ms-ai-architect): S1 judge bake-off RESULT — GATE FAIL (recall 68.4% < 80%) [skip-docs] 2026-06-26 20:45:09 +02:00
judge-bakeoff-report.md feat(ms-ai-architect): S1 judge bake-off RESULT — GATE FAIL (recall 68.4% < 80%) [skip-docs] 2026-06-26 20:45:09 +02:00
judge-bakeoff-results-v2.json feat(ms-ai-architect): S1 v2 targeted iteration — GATE PASS (recall 84.2%, precision 84.2%) [skip-docs] 2026-06-26 21:23:50 +02:00
judge-bakeoff-results-v3.1.json feat(ms-ai-architect): G1 LUKKET — v3.1 judge MÅLT + ADOPTERT (P100/R100/0FP/0FN, max-utfall) 2026-06-30 21:33:45 +02:00
judge-bakeoff-results-v3.json feat(ms-ai-architect): G1 v3 bake-off MÅLT — v3 regredierte (P 92.1→89.7, R flat), v2 beholdt; G5 gull-friskhet åpnet [skip-docs] 2026-06-30 10:54:39 +02:00
judge-bakeoff-results.json feat(ms-ai-architect): S1 judge bake-off RESULT — GATE FAIL (recall 68.4% < 80%) [skip-docs] 2026-06-26 20:45:09 +02:00
judge-pass-manifest.json feat(ms-ai-architect): R7.2 stempling — 9/11 pass born-verified etter re-judge-gate; 2 diskvalifisert → flagged (R11) [skip-docs] 2026-07-24 20:14:45 +02:00
judge-results.json feat(ms-ai-architect): Spor D steg A — eng+infra K10-fix → korpus-invariant ≥90 oppfylt 2026-06-23 17:56:43 +02:00
k1-trigger-prompts.json fix(ms-ai-architect): Sesjon 11 — K1 trigger-presisjon, blindet judge PASS alle 5 2026-06-20 10:00:47 +02:00
r71-payloads.json feat(ms-ai-architect): R7.1 — judge-pass 30/45 filer (362/544 claims; 1 pass, 29 flagged, 106 flagg; v3.1-judge, ledger durabel per fil) [skip-docs] 2026-07-18 17:34:49 +02:00
spor0-fix-manifest.json fix(ms-ai-architect): R5 G5b — 4 innholds-fikser (retired gpt-35-turbo→gpt-4o-mini, vector-quant GA 2024-07-01, 2 overclaims nyansert), hver live-verifisert per MS Learn-kilde [skip-docs] 2026-07-04 17:09:23 +02:00