docs(architect): weekly KB update — 66 files refreshed (2026-04)

Updated 66 stale knowledge base reference files (10 critical, 56 high)
across all 5 skills using Microsoft Learn MCP research.

Key factual updates:
- Groundedness Detection API: `correction` → `mitigating` param,
  `correctedText` → `correctionText` (breaking change)
- Copilot Studio: GPT-4.1 mini now default (was GPT-4o mini);
  Claude Sonnet 4.5 + Opus 4.5 added (experimental, 200K ctx)
- Agentic Retrieval: still public preview; 50M free tokens/month
- Azure security baselines: "Cognitive Services" → "Foundry Tools"
- Databricks: Delta Live Tables → Lakeflow Spark Declarative Pipelines
- MLflow 3 GenAI: new Feedback/Expectation data model
- Token tracking doc: "Azure OpenAI in Foundry Models through a gateway"
- Agent Registry: Risks column (M365 E7), Graph API (preview)
- Copilot DLP: new Entra AI Admin + Purview Data Security AI Admin roles
- ISO/IEC 42001: scope expanded to M365 Copilot, Foundry, Security Copilot
- Zero Trust: CAE now via Conditional Access, Strict Location Enforcement
- Purview: new Fabric Copilots/agents governance section
- AG-UI HITL: ApprovalRequiredAIFunction (C#), @tool approval_mode (Python)

All files: Last updated → 2026-04, *(Verified MCP 2026-04)* markers added.
Build registry: 1341 URLs from 387 files (+2 new URLs).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

skills/ms-ai-governance/references/monitoring-observability/sla-monitoring-ai-services.md

@@ -1,6 +1,6 @@
 # SLA Monitoring and Availability Tracking for AI Services
 
-**Last updated:** 2026-02
+**Last updated:** 2026-04
 **Status:** GA
 **Category:** Monitoring & Observability
 
@@ -32,7 +32,8 @@ SLA-monitorering skiller seg fra generell ytelsesmonitorering ved at den er styr
 |-----------|----------|--------------|
 | **Platform Metrics** | Automatisk innsamling av `AvailabilityRate`, `ModelAvailabilityRate` | Sanntids tilgjengelighetsprosent |
 | **Diagnostic Settings** | Rute metrics til Log Analytics for langtidslagring | Revisjonsbevis og historisk analyse |
-| **Metric Alerts** | Automatisk varsling ved SLA-brudd (f.eks. availability < 99.9%) | Proaktiv incident management |
+| **Metric Alerts** | Automatisk varsling ved SLA-brudd (f.eks. availability < 99.9%). Støtter dynamic thresholds og multi-resource. | Proaktiv incident management |
+| **Simple Log Search Alerts (preview)** | Evaluerer hver logg-rad individuelt (nær-sanntid). Raskere enn tradisjonelle log alerts. | Per-request SLA-brudd oppdaget nesten umiddelbart *(Verified MCP 2026-04)* |
 | **Workbooks/Dashboards** | Visuell fremstilling av SLA-status over tid | Executive reporting og trend-analyse |
 | **Azure Service Health** | Plattformvarsler om kjente utfall | Ekstern faktor-tracking (force majeure) |
 
@@ -159,7 +160,7 @@ az monitor metrics alert create \
 |------|------------|---------------|
 | **Bruke `AvailabilityRate` for Azure OpenAI** | Feil metric (gjelder ikke OpenAI) | Bruk `ModelAvailabilityRate` i stedet |
 | **Ikke aktivere Diagnostic Settings** | Kun 93 dagers metrics-retention | Sett opp Log Analytics-export fra dag 1 |
-| **Varsle på enkelthendelser i stedet for trender** | False positives (transiente feil) | Bruk `windowSize` >= 5 min og `evaluationFrequency` for å dempe støy |
+| **Varsle på enkelthendelser i stedet for trender** | False positives (transiente feil) | Bruk `windowSize` >= 5 min og `evaluationFrequency` for å dempe støy. Bruk stateful alerts for infrastruktur-events (én alert per incident) *(Verified MCP 2026-04)* |
 | **Glemme å ekskludere planlagt vedlikehold** | Feilaktig SLA-beregning | Korreiger downtime for Azure Service Health-hendelser |
 | **Lagre SLA-data i samme workspace som debugging-logger** | Overfladisk støy i SLA-rapporter | Bruk dedikert Log Analytics workspace for SLA-metrics |
 
@@ -337,7 +338,7 @@ Metric Alert
 
 | Modenhetsnivå | Anbefaling | Verktøy |
 |---------------|-----------|---------|
-| **Pilot/POC** | Basic metric alerts (email on SLA breach) | Azure Monitor alerts (native) |
+| **Pilot/POC** | Basic metric alerts (email on SLA breach). Vurder Simple Log Search Alerts (preview) for per-request visibility *(Verified MCP 2026-04)* | Azure Monitor alerts (native) |
 | **Produksjon (liten skala)** | Diagnostic settings + Log Analytics + Workbooks | 1 Log Analytics workspace, 3-5 alert rules |
 | **Produksjon (stor skala)** | Multi-tier monitoring + ITSM integration | Dedicated SLA workspace, Action Groups → ServiceNow/Linear |
 | **Enterprise** | Hybrid (hot/warm/cold) + automated reporting + capacity planning | Power BI + Azure DevOps integration + predictive analytics |
@@ -371,7 +372,7 @@ Metric Alert
 
 6. **Azure Monitor alerts overview**
    https://learn.microsoft.com/en-us/azure/azure-monitor/alerts/alerts-overview
-   *Confidence: Verified* — Alert-typer (metric, log, activity log).
+   *Confidence: Verified* — Alert-typer: metric, log search, simple log search (preview, per-row evaluering), activity log, smart detection, Prometheus. Alerts lagres i 30 dager. Stateful log search alerts har detaljerte resolusjonskriterier per evalueringsfrekvens. Query-based metric alerts for Prometheus/OpenTelemetry i public preview. *(Verified MCP 2026-04)*
 
 7. **Reliability in Azure AI Search (SLA example)**
    https://learn.microsoft.com/en-us/azure/reliability/reliability-ai-search#service-level-agreement