open/ms-ai-architect
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity Actions
ms-ai-architect/docs
History
Kjell Tore Guttormsen 4cd290c14b feat(ms-ai-architect): S1 v2 targeted iteration — GATE PASS (recall 84.2%, precision 84.2%) [skip-docs] 
Operatørvalg (c): én målrettet, prinsipiell prompt-iterasjon på den diagnostiserte
grounded-men-feil-feilmoden (eksakt-verdi-entailment). Terskel uendret; v1 frosset.

Full blind v2 fan-out (15 batcher Opus 4.8 xhigh, 255 P-påstander dekket):
- judge v2: recall 84.2% (32/38, PASS >=0.80, Wilson [69.6-92.6%]), presisjon 84.2%
  (PASS >=0.70), F1 0.842, slår staleness 0/38.
- Fiksen løste målet: sku recall 37.5%->75.0%, taxonomy 66.7%->100%.
- +6 ekte fangster (26->32) uten netto nye FP (6->6) => recall OG presisjon opp.

Forbehold (ærlig): andre måling på samme frosne sett etter v1 (erkjent); Wilson nedre
grense 69.6% < 0.80 ved n=38; én iterasjon. Gate-logikk => vei mot S3. Stoppet for
operatør-beslutning (S2/S3), eskalerer ikke selv.

run-judge-bakeoff.mjs: --results/--report-prefix flagg (v2 uten å klobbe v1). Suite 552/552.
2026-06-26 21:23:50 +02:00
..
c3-course-detection-plan.md docs(ms-ai-architect): C3 §8 besluttet (delegert) — spec GODKJENT, +fetch-robusthetskontrakt +removed full-enum-pin [skip-docs] 2026-06-23 12:42:22 +02:00
c3-course-detection-spike.md docs(ms-ai-architect): C3 Fase 0 — Learn Platform API auth-oppsett + spike kartlagt; STATE → spec neste [skip-docs] 2026-06-23 12:09:19 +02:00
cosmo-removal-brief-2026-06.md feat(ms-ai-architect): insertToc + backfill-toc applier (Fase 1b-verktøy) — store-fil-TOC foldet inn i Cosmo-utfasing 2026-06-26 11:43:30 +02:00
development.md feat(ms-ai-architect): Spor D steg 2 — score-skill CLI (--json/--gate/--skill) (TDD) 2026-06-23 17:16:44 +02:00
devils-advocate-audit-2026-06-18.md docs(ms-ai-architect): devil's-advocate-audit (10 dim, 89 verifiserte funn) 2026-06-18 12:00:34 +02:00
eu-ai-act-integration-handover.md chore(privacy): scrub real-org references from plugin internals (phase 2) 2026-05-03 04:28:15 +02:00
gap-analysis-2026-02.md feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
kb-mechanism-redesign-brief.md docs(ms-ai-architect): scope-brief for KB-mekanisme-redesign (neste fase) 2026-06-19 14:24:13 +02:00
kb-mechanism-redesign-plan.md fix(ms-ai-architect): Sesjon 11 — K1 trigger-presisjon, blindet judge PASS alle 5 2026-06-20 10:00:47 +02:00
kb-refresh-backlog-2026-06.md docs(ms-ai-architect): noter ref-fil innholdskvalitet-gap i backlog — Spor D scorer struktur/hygiene (K8/N3/N4/refCount), ikke substansiell korrekthet mot MS Learn; 3 retningsvalg å avklare før bygging 2026-06-26 00:12:04 +02:00
kb-refresh-cost-cluster-2026-06.md docs(ms-ai-architect): apply-brief critical cost-cluster KB-refresh (7 drift-rapporter) 2026-06-23 21:59:11 +02:00
kb-update-apply-brief.md docs(ms-ai-architect): STATE + brief — Medium-bøtte FERDIG (070141f) 2026-06-19 14:03:54 +02:00
kb-update-policy.md feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
okf-second-brain-brief-2026-06.md docs(ms-ai-architect): gjennomtenkt plan for ref-fil-kvalitet + workflow (Fase 0-4, gated) 2026-06-26 01:02:15 +02:00
onboarding-redesign-brief.md test(ms-ai-architect): C2.4 lås K4 privat-sektor-paritet (regresjonsvakt) + omdøp feildøpt doc [skip-docs] 2026-06-22 17:39:11 +02:00
onboarding-redesign-plan.md test(ms-ai-architect): C2.4 lås K4 privat-sektor-paritet (regresjonsvakt) + omdøp feildøpt doc [skip-docs] 2026-06-22 17:39:11 +02:00
playground-ai-act-plan.md feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
playground-plan.md feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
playground-v2-spec.md feat(ultraplan-local): v1.6.0 — /ultraresearch-local deep research command 2026-04-08 08:58:35 +02:00
playground.md chore(ms-ai-architect): release v1.16.1 — KB-refresh (tema a+b+RBAC) 2026-06-23 21:33:28 +02:00
ref-kb-audit-2026-06.md docs(ms-ai-architect): ref-KB audit — verifisert ground truth + reproduserbart audit-skript 2026-06-26 00:25:39 +02:00
ref-kb-direction-note-2026-06.md docs(ms-ai-architect): reconcile ref-KB named/folder-premiss — 169/220/0 autoritativt 2026-06-26 10:15:46 +02:00
ref-kb-workflow-plan-2026-06.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
skill-lifecycle-spor-b-plan.md docs(ms-ai-architect): Spor B plan godkjent — skill-livssyklus (krav 5+6, omdefinert) 2026-06-20 11:09:11 +02:00
skill-quality-scoring-plan.md feat(ms-ai-architect): Spor D — skill-quality score-motor (ren lib + rubrikk grunnet i Anthropic) (TDD) 2026-06-23 16:57:31 +02:00
windows-clone-to-pr-ros-guide.md test(ms-ai-architect): C2.4 lås K4 privat-sektor-paritet (regresjonsvakt) + omdøp feildøpt doc [skip-docs] 2026-06-22 17:39:11 +02:00