feat(ms-ai-architect): Sesjon 7 steg 3 - K5 progressive disclosure (3 skills PASS)
Lukker K5-FAIL (navngitte fil-lenker / totale ref-filer ≥ 0,20) ved å følge
infrastructure-forbildet (0,97): full-sti `references/<mappe>/<fil>.md`-pekere
til kjernefiler, ikke bare mappe-refs.
Funn: security + advisor hadde filene navngitt allerede, men som BARE filnavn
uten `references/`-prefiks → ufanget av både eval-regex og kb-integrity (samme
klasse som de døde ref-paths i steg 2). Fiks = konverter til full sti:
- security: 10→16 navngitte (0,16→0,26). Konverterte 5 perf-filer (§3) + owasp (§1).
- advisor: 1→25 navngitte (0,016→0,40). Konverterte ~18 bare-filnavn i Kunnskaps-
basen + la til model-catalog-2026 og entry-points for copilot-extensi-
bility/prompt-engineering (40 filer som manglet ALLE navngitte pekere).
- engineering: 0→35 navngitte (0,0→0,23). Genuint 0 før; la til `> Kjernefiler:`-linje
med 4-6 kuraterte filer per §1-7.
Bivirkning: kb-integrity-checks 115→181 (de nye full-sti-refsene valideres nå),
orphan-warnings 260→223. Verifisert: K5 PASS alle 5 · K3/refTall ikke regredert ·
validate 239 · kb-eval 15 · kb-update 122 · kb-integrity 181/181.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01REiKFhP4w6xGXXqWKpPCJJ
This commit is contained in:
parent
3c097f5fb4
commit
abac74fe67
3 changed files with 47 additions and 28 deletions
|
|
@ -200,35 +200,40 @@ Du har tilgang til følgende MCP-servere:
|
|||
Du har tilgang til forhåndsresearchede kunnskapsbaser i `references/`-mappen:
|
||||
|
||||
**Plattformer (`references/platforms/`):**
|
||||
- `azure-ai-foundry.md` - Azure AI Foundry vs Copilot Studio vs Azure OpenAI
|
||||
- `m365-copilot.md` - Microsoft 365 Copilot: kapasiteter, lisensiering, extensibility
|
||||
- `copilot-studio.md` - Copilot Studio: agenttyper, MCP-støtte, autonome agenter
|
||||
- `power-platform.md` - Power Automate, Power Apps, AI Builder
|
||||
- `references/platforms/azure-ai-foundry.md` - Azure AI Foundry vs Copilot Studio vs Azure OpenAI
|
||||
- `references/platforms/m365-copilot.md` - Microsoft 365 Copilot: kapasiteter, lisensiering, extensibility
|
||||
- `references/platforms/copilot-studio.md` - Copilot Studio: agenttyper, MCP-støtte, autonome agenter
|
||||
- `references/platforms/power-platform.md` - Power Automate, Power Apps, AI Builder
|
||||
- `references/platforms/model-catalog-2026.md` - Modellkatalog: kapasiteter, kontekstvindu, prising
|
||||
|
||||
**Arkitektur (`references/architecture/`):**
|
||||
- `decision-trees.md` - Beslutningstrær for plattformvalg, agenttyper, RAG, sikkerhet
|
||||
- `security.md` - Content Safety, Purview, Defender, identity, compliance
|
||||
- `ai-utredning-template.md` - Utredningsmal for offentlig sektor
|
||||
- `cost-models.md` - Kostnadsmodeller per plattform
|
||||
- `licensing-matrix.md` - Lisensmatrise for Microsoft AI
|
||||
- `poc-template.md` - POC-rammeverk
|
||||
- `migration-patterns.md` - Migrasjonsmønstre mellom plattformer
|
||||
- `public-sector-checklist.md` - Sjekkliste for offentlig sektor
|
||||
- `adr-template.md` - ADR-mal (MADR v3.0)
|
||||
- `diagram-prompt-templates.md` - Diagramprompts for Imagen 3
|
||||
- `recommended-mcp-servers.md` - Anbefalte MCP-servere
|
||||
- `alternativanalyse-methodology.md` - Vektet multi-kriterie-analyse (brukt av `/architect:compare --weighted`)
|
||||
- `capacity-feasibility-benchmarks.md` - Kompetanse-gap + tidsplan-benchmarks (brukt av `/architect:cost --capacity`)
|
||||
- `references/architecture/decision-trees.md` - Beslutningstrær for plattformvalg, agenttyper, RAG, sikkerhet
|
||||
- `references/architecture/security.md` - Content Safety, Purview, Defender, identity, compliance
|
||||
- `references/architecture/ai-utredning-template.md` - Utredningsmal for offentlig sektor
|
||||
- `references/architecture/cost-models.md` - Kostnadsmodeller per plattform
|
||||
- `references/architecture/licensing-matrix.md` - Lisensmatrise for Microsoft AI
|
||||
- `references/architecture/poc-template.md` - POC-rammeverk
|
||||
- `references/architecture/migration-patterns.md` - Migrasjonsmønstre mellom plattformer
|
||||
- `references/architecture/public-sector-checklist.md` - Sjekkliste for offentlig sektor
|
||||
- `references/architecture/adr-template.md` - ADR-mal (MADR v3.0)
|
||||
- `references/architecture/diagram-prompt-templates.md` - Diagramprompts for Imagen 3
|
||||
- `references/architecture/recommended-mcp-servers.md` - Anbefalte MCP-servere
|
||||
- `references/architecture/alternativanalyse-methodology.md` - Vektet multi-kriterie-analyse (brukt av `/architect:compare --weighted`)
|
||||
- `references/architecture/capacity-feasibility-benchmarks.md` - Kompetanse-gap + tidsplan-benchmarks (brukt av `/architect:cost --capacity`)
|
||||
- (+ øvrige filer i architecture/)
|
||||
|
||||
**Utvikling (`references/development/`):**
|
||||
- `agent-framework.md` - Microsoft Agent Framework
|
||||
- `references/development/agent-framework.md` - Microsoft Agent Framework
|
||||
|
||||
**Copilot-utvidbarhet (`references/copilot-extensibility/`):**
|
||||
- Declarative agents, custom engine, plugins, connectors (22 filer)
|
||||
**Copilot-utvidbarhet (`references/copilot-extensibility/`, 22 filer — kjernefiler:):**
|
||||
- `references/copilot-extensibility/declarative-agents-fundamentals.md` - Declarative agents (grunnlag)
|
||||
- `references/copilot-extensibility/copilot-connectors-design-patterns.md` - Copilot connectors (designmønstre)
|
||||
- `references/copilot-extensibility/custom-engine-agents-development.md` - Custom engine agents
|
||||
|
||||
**Prompt Engineering (`references/prompt-engineering/`):**
|
||||
- System messages, few-shot, CoT, reasoning, grounding (18 filer)
|
||||
**Prompt Engineering (`references/prompt-engineering/`, 18 filer — kjernefiler:):**
|
||||
- `references/prompt-engineering/system-message-design-patterns.md` - System message-design
|
||||
- `references/prompt-engineering/chain-of-thought-prompting.md` - Chain-of-thought-mønstre
|
||||
- `references/prompt-engineering/prompt-testing-and-evaluation.md` - Testing og evaluering
|
||||
|
||||
### Kryss-referanser til andre skills
|
||||
|
||||
|
|
|
|||
|
|
@ -41,6 +41,8 @@ RAG er det mest brukte mønsteret for å gi LLM-er organisasjonsspesifikk kunnsk
|
|||
|
||||
For detailed guidance, see `references/rag-architecture/` (28 files).
|
||||
|
||||
> **Kjernefiler:** `references/rag-architecture/rag-core-patterns.md`, `references/rag-architecture/agentic-rag-patterns.md`, `references/rag-architecture/azure-ai-search-setup.md`, `references/rag-architecture/chunking-strategies.md`, `references/rag-architecture/hybrid-search-configuration.md`, `references/rag-architecture/rag-evaluation-frameworks.md`
|
||||
|
||||
---
|
||||
|
||||
## 2. Agent-orkestrering
|
||||
|
|
@ -51,6 +53,8 @@ For offentlig sektor: Agenter krever AI Act-klassifisering, DPIA for persondata,
|
|||
|
||||
For detailed guidance, see `references/agent-orchestration/` (24 files).
|
||||
|
||||
> **Kjernefiler:** `references/agent-orchestration/multi-agent-orchestration-patterns.md`, `references/agent-orchestration/semantic-kernel-agents-implementation.md`, `references/agent-orchestration/foundry-agent-service-ga.md`, `references/agent-orchestration/tool-use-and-function-calling-patterns.md`, `references/agent-orchestration/agent-to-agent-a2a-protocol.md`
|
||||
|
||||
---
|
||||
|
||||
## 3. Azure AI Services
|
||||
|
|
@ -61,6 +65,8 @@ Start med GPT-4o for prototyping, bruk spesialiserte tjenester for ytelse/kostna
|
|||
|
||||
For detailed guidance, see `references/azure-ai-services/` (20 files).
|
||||
|
||||
> **Kjernefiler:** `references/azure-ai-services/ai-services-enterprise-architecture.md`, `references/azure-ai-services/document-intelligence-prebuilt-models.md`, `references/azure-ai-services/speech-services-speech-to-text.md`, `references/azure-ai-services/azure-ai-vision-image-analysis.md`, `references/azure-ai-services/language-services-text-analytics.md`
|
||||
|
||||
---
|
||||
|
||||
## 4. Dataingeniør for AI
|
||||
|
|
@ -71,6 +77,8 @@ For offentlig sektor: Anonymiser/pseudonymiser personopplysninger, bruk syntetis
|
|||
|
||||
For detailed guidance, see `references/data-engineering/` (22 files).
|
||||
|
||||
> **Kjernefiler:** `references/data-engineering/fabric-lakehouse-architecture.md`, `references/data-engineering/onelake-data-strategy.md`, `references/data-engineering/data-factory-ai-pipelines.md`, `references/data-engineering/data-quality-ai-frameworks.md`, `references/data-engineering/microsoft-purview-governance.md`
|
||||
|
||||
---
|
||||
|
||||
## 5. MLOps / GenAIOps
|
||||
|
|
@ -87,6 +95,8 @@ MLOps og GenAIOps sikrer pålitelig bygging, deployment og drift av AI-løsninge
|
|||
|
||||
For detailed guidance, see `references/mlops-genaiops/` (22 files).
|
||||
|
||||
> **Kjernefiler:** `references/mlops-genaiops/mlops-fundamentals-overview.md`, `references/mlops-genaiops/genaiops-llm-specific-practices.md`, `references/mlops-genaiops/ci-cd-for-ml-models.md`, `references/mlops-genaiops/model-versioning-registry-management.md`, `references/mlops-genaiops/llm-evaluation-production.md`
|
||||
|
||||
---
|
||||
|
||||
## 6. Multimodal AI
|
||||
|
|
@ -95,6 +105,8 @@ Multimodal AI kombinerer tekst, bilde, lyd og video. GPT-4o vision analyserer bi
|
|||
|
||||
For detailed guidance, see `references/multi-modal/` (18 files).
|
||||
|
||||
> **Kjernefiler:** `references/multi-modal/gpt4o-vision-architecture.md`, `references/multi-modal/multimodal-rag-architecture.md`, `references/multi-modal/document-vision-processing.md`, `references/multi-modal/whisper-speech-recognition.md`
|
||||
|
||||
---
|
||||
|
||||
## 7. API Management for AI
|
||||
|
|
@ -105,6 +117,8 @@ Bruk semantisk caching for FAQ-lignende domener. Unngå for personaliserte svar
|
|||
|
||||
For detailed guidance, see `references/api-management/` (19 files).
|
||||
|
||||
> **Kjernefiler:** `references/api-management/apim-ai-gateway-overview.md`, `references/api-management/genai-gateway-policies.md`, `references/api-management/token-rate-limiting-policies.md`, `references/api-management/load-balancing-openai-instances.md`, `references/api-management/semantic-caching-apim.md`
|
||||
|
||||
---
|
||||
|
||||
## 8. Referansekatalog
|
||||
|
|
|
|||
|
|
@ -87,7 +87,7 @@ Map each threat to the solution under assessment. Use the reference files for de
|
|||
| LLM09 | Misinformation | RAG grounding, Groundedness Detection, citation patterns, confidence scoring | `owasp-llm-top10-azure-mitigations.md` |
|
||||
| LLM10 | Unbounded Consumption | Rate limits, token budgets, PTU for capacity, Cost Management alerts | — |
|
||||
|
||||
All reference files are in `references/ai-security-engineering/`. LLM04/06/08/09 deler den konsoliderte filen `owasp-llm-top10-azure-mitigations.md`; LLM10 dekkes av rate-limit-/kostnadsfiler.
|
||||
All reference files are in `references/ai-security-engineering/`. LLM04/06/08/09 deler den konsoliderte filen `references/ai-security-engineering/owasp-llm-top10-azure-mitigations.md`; LLM10 dekkes av rate-limit-/kostnadsfiler.
|
||||
|
||||
Kjøretids-trusseldeteksjon for AI-endepunkter dekkes av `defender-threat-protection-ai-services.md` (Defender for Cloud AI threat protection — GA for AI applications, Preview for AI agents; merk: **ikke** tilgjengelig i Azure Government).
|
||||
|
||||
|
|
@ -155,11 +155,11 @@ Optimize latency, throughput, and scalability for AI workloads. Key strategies:
|
|||
- **Load testing:** Establish baseline, simulate peak traffic, identify breaking points, long-running soak tests
|
||||
|
||||
For detailed implementation guidance, see specific files in `references/performance-scalability/`:
|
||||
- `latency-optimization-azure-openai.md` — Latency tuning
|
||||
- `auto-scaling-ai-infrastructure.md` — Scaling patterns
|
||||
- `rate-limit-management.md` — TPM/RPM quota management
|
||||
- `load-testing-ai-services.md` — Load testing methodology
|
||||
- `gpu-compute-sizing.md` — GPU VM-sizing for selvhostet inferens (brukt av `/architect:cost --capacity`)
|
||||
- `references/performance-scalability/latency-optimization-azure-openai.md` — Latency tuning
|
||||
- `references/performance-scalability/auto-scaling-ai-infrastructure.md` — Scaling patterns
|
||||
- `references/performance-scalability/rate-limit-management.md` — TPM/RPM quota management
|
||||
- `references/performance-scalability/load-testing-ai-services.md` — Load testing methodology
|
||||
- `references/performance-scalability/gpu-compute-sizing.md` — GPU VM-sizing for selvhostet inferens (brukt av `/architect:cost --capacity`)
|
||||
|
||||
---
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue