fix(ms-ai-architect): RX-P2 reg-kjede-wiring + komprehensiv agent-sti-forankring [skip-docs]

- classify->dpia->ros kjede-wiring: dpia/ros Task-templater far eksplisitte AI Act-klassifiserings- + DPIA-funn-felt (kjede-data agentene allerede branchet pa) - dpia-agent uutforbar "spor bruker" -> "marker vurderingen" (Hvis-ikke-klassifisert + Error Handling); speilet til ros-analysis-agent - adr-writer-agent: Write-verktoy fjernet, returnerer ADR-markdown til hovedkontekst (command er eneste skriver -- subagenter skriver aldri) - KB-sti-forankring (komprehensiv): 36 bare referanse-subdir-stier i 6 agenter + 2 commands fullkvalifisert med CLAUDE_PLUGIN_ROOT/skills/<skill>/references/ --- ogsa innen-skill kortformer, uresolverbare i installert modus - mekanisme: validate-plugin.sh Check 6d (bare-subdir-lint m/ hyphen-guard) + negativ-probe-test (beviser tenner) + RX-P2 wiring-regresjonstester Suite 859/859 exit 0. validate-plugin 250 PASS / 0 FAIL. 145 forankrede stier verifisert eksisterende (0 mangler).
2026-07-15 12:17:01 +02:00 · 2026-07-15 12:17:01 +02:00 · b68514487c
commit b68514487c
parent 5b05009b44
13 changed files with 151 additions and 36 deletions
--- a/agents/adr-writer-agent.md
+++ b/agents/adr-writer-agent.md
@ -2,12 +2,12 @@
 name: adr-writer-agent
 description: |
  Generates Architecture Decision Records (ADR) in MADR v3.0 format from structured input.
-  Reads adr-template.md, fills in from session context, and writes to file.
+  Reads adr-template.md, fills in from session context, and returns the ADR markdown to the main context.
  Use when architect:adr needs to generate a complete ADR document.
  Triggers on: ADR generation, decision documentation, architect:adr delegation.
 model: opus
 color: orange
-tools: ["Read", "Write", "Glob"]
+tools: ["Read", "Glob"]
 ---

 # ADR Writer Agent
@ -87,9 +87,9 @@ Fill in every section of the MADR template:

 **Validering og oppfølging**: Concrete next steps with responsible party.

-### 4. Write to File
+### 4. Return to Main Context

-Write the ADR to the location specified in the input. Default: `docs/adr/ADR-NNN-[slug].md`
+Return the complete ADR markdown as your final message. The main context (the `/architect:adr` command) writes it to file — do not write files yourself (you run as a subagent).

 ## Output Format

@ -101,7 +101,7 @@ The generated ADR should be:

 ## Quality Checklist

-Before writing:
+Before returning:
 - [ ] All template sections filled (no placeholders)
 - [ ] Compliance section included (even if "Not assessed")
 - [ ] Confidence level reflects actual analysis quality
--- a/agents/architecture-review-agent.md
+++ b/agents/architecture-review-agent.md
@ -164,13 +164,13 @@ Read relevant knowledge base files:
 - `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-advisor/references/architecture/licensing-matrix.md` — License requirements

 Load domain-specific references only when dimension requires depth (max 2-3 additional):
- AI Act: `responsible-ai/ai-act-compliance-guide.md`, `responsible-ai/ai-act-annex-iii-checklist.md`
- Governance: `responsible-ai/ai-governance-structure-framework.md`
- Norwegian: `norwegian-public-sector-governance/utredningsinstruksen-ai-methodology.md`
- Security: `ai-security-engineering/ai-threat-modeling-stride.md`
- Cost: `cost-optimization/azure-ai-foundry-cost-governance.md`, `cost-optimization/deterministic-cost-calculation-model.md`
- RAG-arkitektur (når løsningen er RAG-/gjenfinningsbasert): `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `rag-architecture/agentic-rag-patterns.md`, `rag-architecture/rag-evaluation-frameworks.md`
- MLOps/GenAIOps (når løsningen har produksjons-/livssyklusfokus): `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `mlops-genaiops/monitoring-observability-ml-systems.md`, `mlops-genaiops/model-deployment-strategies-azure.md`
+- AI Act: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-compliance-guide.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-annex-iii-checklist.md`
+- Governance: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-governance-structure-framework.md`
+- Norwegian: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/norwegian-public-sector-governance/utredningsinstruksen-ai-methodology.md`
+- Security: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/ai-threat-modeling-stride.md`
+- Cost: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/cost-optimization/azure-ai-foundry-cost-governance.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/cost-optimization/deterministic-cost-calculation-model.md`
+- RAG-arkitektur (når løsningen er RAG-/gjenfinningsbasert): `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/agentic-rag-patterns.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-evaluation-frameworks.md`
+- MLOps/GenAIOps (når løsningen har produksjons-/livssyklusfokus): `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/monitoring-observability-ml-systems.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/model-deployment-strategies-azure.md`

 ## Virksomhetskontekst (automatisk)

--- a/agents/cost-estimation-agent.md
+++ b/agents/cost-estimation-agent.md
@ -60,9 +60,9 @@ Read these core files:
 - `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-advisor/references/architecture/cost-models.md` — Cost model templates

 Load additional files only when estimate requires specific depth:
- PTU: `cost-optimization/ptu-vs-paygo-economics.md`
- Caching: `cost-optimization/semantic-caching-patterns.md`
- Model selection: `cost-optimization/model-selection-price-performance.md`
+- PTU: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/cost-optimization/ptu-vs-paygo-economics.md`
+- Caching: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/cost-optimization/semantic-caching-patterns.md`
+- Model selection: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/cost-optimization/model-selection-price-performance.md`

 ## Virksomhetskontekst (automatisk)

--- a/agents/dpia-agent.md
+++ b/agents/dpia-agent.md
@ -22,10 +22,10 @@ Read these core files:
 - `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-impact-assessment-framework.md` — Konsekvensvurdering

 Load additional files only when assessment requires specific depth:
- Bias: `responsible-ai/bias-detection-mitigation-strategies.md`
- PII: `ai-security-engineering/pii-detection-norwegian-context.md`
- Data leakage: `ai-security-engineering/data-leakage-prevention-ai.md`
- **Cross-border / Schrems II (OBLIGATORISK når data kan nås fra tredjeland — se Fase 3, risiko 7):** `monitoring-observability/data-residency-audit-monitoring.md` — EDPB seks-stegs-TIA, CLOUD Act/FISA 702/EO 12333-restanalyse, EO 14086/DPF-status, tekniske tilleggstiltak
+- Bias: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/bias-detection-mitigation-strategies.md`
+- PII: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/pii-detection-norwegian-context.md`
+- Data leakage: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/data-leakage-prevention-ai.md`
+- **Cross-border / Schrems II (OBLIGATORISK når data kan nås fra tredjeland — se Fase 3, risiko 7):** `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/monitoring-observability/data-residency-audit-monitoring.md` — EDPB seks-stegs-TIA, CLOUD Act/FISA 702/EO 12333-restanalyse, EO 14086/DPF-status, tekniske tilleggstiltak

 ## Virksomhetskontekst (automatisk)

@ -47,7 +47,7 @@ Før DPIA-vurderingen, sjekk om AI Act-klassifisering er utført:
 - Integrer deployer-forpliktelser fra `ai-act-deployer-obligations.md` som tiltak i Fase 4

 ### Hvis ikke klassifisert
- Spør om det bør gjøres: "Er det gjennomført AI Act-klassifisering for dette systemet? Hvis nei, anbefaler vi `/architect:classify` — men DPIA fortsetter uansett."
+- Marker i rapporten at AI Act-klassifisering ikke er dokumentert, og anbefal `/architect:classify` som neste steg (du kjører som subagent uten brukertur — still ingen spørsmål)
 - Fortsett DPIA som normalt — klassifisering er ikke forutsetning

 ### Ekstra KB-referanser for AI Act
@ -93,7 +93,7 @@ Risk categories for AI systems:

 #### Cross-border / Schrems II — obligatorisk TIA (risiko 7)

-Når systemet bruker en amerikansk-eid skyleverandør (Azure/Microsoft 365/Foundry) eller data på annen måte kan nås fra tredjeland, **er det ikke nok å navngi risikoen** — load `monitoring-observability/data-residency-audit-monitoring.md` og gjennomfør EDPB seks-stegs Transfer Impact Assessment:
+Når systemet bruker en amerikansk-eid skyleverandør (Azure/Microsoft 365/Foundry) eller data på annen måte kan nås fra tredjeland, **er det ikke nok å navngi risikoen** — load `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/monitoring-observability/data-residency-audit-monitoring.md` og gjennomfør EDPB seks-stegs Transfer Impact Assessment:

 1. Kartlegg overføringene (inkl. residual: support, troubleshooting, telemetri)
 2. Identifiser overføringsverktøyet (adekvansvedtak / SCCs / unntak)
@ -141,9 +141,9 @@ Read the AI system description or architecture proposal. Extract:

 ### 2. Load Reference Knowledge
 Core files are loaded via Knowledge Base References above. For deeper analysis:
- Fairness: `responsible-ai/fairness-testing-measurement.md`
- Transparency: `responsible-ai/transparency-documentation-standards.md`
- Human oversight: `responsible-ai/human-in-the-loop-oversight.md`
+- Fairness: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/fairness-testing-measurement.md`
+- Transparency: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md`
+- Human oversight: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/human-in-the-loop-oversight.md`

 ### 3. Validate Latest Guidance
 Use `microsoft_docs_search` for:
@ -225,7 +225,7 @@ Follow the output format below with all sections completed.

 If missing information:
 - State assumptions clearly
- Request specific details needed
+- Note which specific inputs are missing rather than requesting them (you run as a non-interactive subagent with no user turn)
 - Provide conditional assessments
 - Note "Kan ikke vurdere [area] uten [info]"

--- a/agents/research-agent.md
+++ b/agents/research-agent.md
@ -34,8 +34,8 @@ Et kompakt sammendrag av virksomhetskonteksten injiseres ambient i hovedøkten v
 ## Lokal KB-baseline (betinget — RAG / MLOps / engineering-temaer)

 Når forskningstemaet er RAG, gjenfinning, MLOps eller GenAIOps, les den relevante engineering-kjernefilen **først** som hypotese-baseline — verifiser den deretter mot live Microsoft Learn. KB-en kan være utdatert; **MCP-resultatet er fasit**.
- RAG/gjenfinning: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `rag-architecture/agentic-rag-patterns.md`
- MLOps/GenAIOps: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `mlops-genaiops/llm-evaluation-production.md`
+- RAG/gjenfinning: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/agentic-rag-patterns.md`
+- MLOps/GenAIOps: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/llm-evaluation-production.md`

 Les maks 2 baseline-filer. **Flagg eksplisitt** hvis live docs avviker fra KB-baselinen (samme avviks-flagging som Fase 4).

--- a/agents/ros-analysis-agent.md
+++ b/agents/ros-analysis-agent.md
@ -36,12 +36,12 @@ Alle stier under `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/norwe
 | Sektor oppdaget (helse/transport/finans/justis/utdanning) | `ros-sector-checklists.md` |
 | Multi-agent / agent-orkestrering | `ros-maestro-multiagent.md` (MAESTRO 7-lag) |
 | DPIA eller sikkerhetsvurdering skal integreres | `ros-dpia-security-integration.md` |
-| AI Act-dybde i dimensjon 6 | `responsible-ai/ai-act-classification-methodology.md` + `responsible-ai/ai-act-provider-obligations.md` (maks 2) |
+| AI Act-dybde i dimensjon 6 | `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-classification-methodology.md` + `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-provider-obligations.md` (maks 2) |

 ### Referanse (last kun ved eksplisitt behov, ikke default)
 - `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/security-scoring-rubrics-6x5.md` — scoringsmønster-referanse
 - `ros-analyse-ai-systems.md` — generell ROS-bakgrunn
- `responsible-ai/ai-risk-taxonomy-classification.md` — risikotaksonomi
+- `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-risk-taxonomy-classification.md` — risikotaksonomi

 **Budsjett:** kjerne (4) + betinget (maks 2-3 på trigger) = typisk 5-7 filer. Aldri last hele katalogen; last ikke en betinget fil hvis triggeren ikke utløses.

@ -273,7 +273,7 @@ Risk Levels: Low (1-6), Medium (7-12), High (13-19), Critical (20-25)

 If missing information:
 - State assumptions clearly
- Request specific details needed
+- Note which specific inputs are missing rather than requesting them (you run as a non-interactive subagent with no user turn)
 - Provide conditional assessments
 - Note "Kan ikke vurdere [area] uten [info]"

--- a/agents/security-assessment-agent.md
+++ b/agents/security-assessment-agent.md
@ -26,9 +26,9 @@ Read these core files:
 - `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/ai-threat-modeling-stride.md` — STRIDE trusselmodellering

 Load additional files only when assessment requires specific depth:
- Prompt injection: `ai-security-engineering/prompt-injection-defense-patterns.md`
- Governance: `responsible-ai/ai-act-compliance-guide.md`
- Norwegian context: `norwegian-public-sector-governance/nsm-grunnprinsipper-ai-mapping.md`
+- Prompt injection: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-security/references/ai-security-engineering/prompt-injection-defense-patterns.md`
+- Governance: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-compliance-guide.md`
+- Norwegian context: `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/norwegian-public-sector-governance/nsm-grunnprinsipper-ai-mapping.md`

 ## Virksomhetskontekst (automatisk)

--- a/commands/dpia.md
+++ b/commands/dpia.md
@ -41,6 +41,7 @@ Gjennomfør en komplett DPIA for følgende AI-system:
 **Registrerte:** [hvem som berøres]
 **Behandlingsgrunnlag:** [GDPR art. 6/9]
 **Kontekst:** [sektor/kontekst — offentlig, privat, finans, helse, etc.]
+**AI Act-klassifisering (fra /architect:classify, hvis utført):** [risikonivå + rolle + Annex III-kategori — ellers "ikke klassifisert"]

 Les kunnskapsbasene (kjerne):
 - ${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/norwegian-public-sector-governance/dpia-norwegian-methodology-ai.md
--- a/commands/poc.md
+++ b/commands/poc.md
@ -34,8 +34,8 @@ Les `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-advisor/references/architecture/poc-temp
 ### 3b. Les domene-spesifikke mønstre (betinget)

 Hvis use-caset treffer et engineering-domene, les 1-2 kjernefiler for å forankre scope og suksesskriterier (ikke hele katalogen):
- **RAG / gjenfinning:** `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `rag-architecture/rag-evaluation-frameworks.md` — sett målbare gjenfinnings-/grounding-kriterier
- **MLOps / produksjonssetting:** `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `mlops-genaiops/llm-evaluation-production.md` — POC-evaluering + driftskriterier
+- **RAG / gjenfinning:** `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-core-patterns.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/rag-architecture/rag-evaluation-frameworks.md` — sett målbare gjenfinnings-/grounding-kriterier
+- **MLOps / produksjonssetting:** `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/genaiops-llm-specific-practices.md`, `${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-engineering/references/mlops-genaiops/llm-evaluation-production.md` — POC-evaluering + driftskriterier

 ### 4. Generer POC-plan

--- a/commands/ros.md
+++ b/commands/ros.md
@ -43,6 +43,8 @@ Gjennomfør en [komplett / quick] ROS-analyse for følgende AI-system:
 **Sektor:** [sektor]
 **Borgermøtende:** [ja/nei]
 **Kontekst:** [ytterligere kontekst]
+**AI Act-klassifisering (dimensjon 6, fra /architect:classify):** [risikonivå + rolle — ellers "ikke klassifisert"]
+**DPIA-funn (fra /architect:dpia, hvis utført):** [sentrale personvernrisikoer — ellers "ikke utført"]
 [**Modus:** Quick (top-10 risikoer, trafikklys) — if --quick]

 Les kunnskapsbasene per last-kontrakten i agentfilen — kjerne (alltid) + betinget (kun på trigger):
@ -57,7 +59,7 @@ Betinget (kun når triggeren utløses):
 - ros-sector-checklists.md (hvis relevant sektor oppdaget)
 - ros-maestro-multiagent.md (hvis multi-agent / agent-orkestrering)
 - ros-dpia-security-integration.md (hvis DPIA/sikkerhet skal integreres)
- responsible-ai/ai-act-classification-methodology.md + ai-act-provider-obligations.md (hvis AI Act-dybde i dimensjon 6)
+- ${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-classification-methodology.md + ${CLAUDE_PLUGIN_ROOT}/skills/ms-ai-governance/references/responsible-ai/ai-act-provider-obligations.md (hvis AI Act-dybde i dimensjon 6)

 Lever en [komplett ROS-rapport med alle 8 faser / Quick ROS med top-10 og trafikklys]."
 ```
--- a/tests/kb-update/test-rx-p2-wiring.test.mjs
+++ b/tests/kb-update/test-rx-p2-wiring.test.mjs
@ -0,0 +1,45 @@
+// RX-P2 regression guards: classify→dpia→ros chain wiring (del 1), removal of the
+// unexecutable "ask user" subagent instructions (del 2), and single ADR write-ownership
+// (del 4). Path anchoring (del 3) + its lint (del 5) are enforced by validate-plugin.sh
+// Check 6d via test-validate-plugin.test.mjs.
+import { test } from 'node:test';
+import assert from 'node:assert/strict';
+import { readFileSync } from 'node:fs';
+import { fileURLToPath } from 'node:url';
+import { dirname, join } from 'node:path';
+
+const here = dirname(fileURLToPath(import.meta.url));
+const root = join(here, '..', '..'); // tests/kb-update -> plugin root
+const read = (rel) => readFileSync(join(root, rel), 'utf8');
+
+// --- Del 1: chain wiring — the Task templates must carry explicit chain-data fields ---
+test('dpia.md Task template passes AI Act classification as chain-data', () => {
+  const dpia = read('commands/dpia.md');
+  assert.match(dpia, /\*\*AI Act-klassifisering \(fra \/architect:classify/);
+});
+
+test('ros.md Task template passes AI Act classification (dim. 6) + DPIA findings as chain-data', () => {
+  const ros = read('commands/ros.md');
+  assert.match(ros, /\*\*AI Act-klassifisering \(dimensjon 6/);
+  assert.match(ros, /\*\*DPIA-funn \(fra \/architect:dpia/);
+});
+
+// --- Del 2: no unexecutable "ask the user" instructions in Task-spawned subagents ---
+test('dpia-agent.md no longer instructs the subagent to ask the user', () => {
+  const agent = read('agents/dpia-agent.md');
+  assert.doesNotMatch(agent, /Spør om det bør gjøres/);
+  assert.doesNotMatch(agent, /Request specific details needed/);
+});
+
+test('ros-analysis-agent.md no longer instructs the subagent to ask the user', () => {
+  const agent = read('agents/ros-analysis-agent.md');
+  assert.doesNotMatch(agent, /Request specific details needed/);
+});
+
+// --- Del 4: adr-writer-agent returns content; the command is the single writer ---
+test('adr-writer-agent.md does not carry the Write tool and does not write files', () => {
+  const agent = read('agents/adr-writer-agent.md');
+  const toolsLine = agent.split('\n').find((l) => l.startsWith('tools:')) || '';
+  assert.doesNotMatch(toolsLine, /Write/, 'adr-writer-agent must not have the Write tool (subagents never write)');
+  assert.doesNotMatch(agent, /### 4\. Write to File/);
+});
--- a/tests/kb-update/test-validate-plugin.test.mjs
+++ b/tests/kb-update/test-validate-plugin.test.mjs
@ -6,6 +6,7 @@ import assert from 'node:assert/strict';
 import { execFileSync } from 'node:child_process';
 import { fileURLToPath } from 'node:url';
 import { dirname, join } from 'node:path';
+import { writeFileSync, rmSync } from 'node:fs';

 const here = dirname(fileURLToPath(import.meta.url));
 const pluginRoot = join(here, '..', '..'); // tests/kb-update -> plugin root
@ -23,3 +24,42 @@ test('validate-plugin.sh passes (frontmatter, encoding, KB refs, plugin.json, in
  assert.match(out, /VALIDATION PASSED/);
  assert.doesNotMatch(out, /VALIDATION FAILED/);
 });
+
+// RX-P2: prove Check 6d has teeth — an unanchored reference-subdir path must be caught.
+// Without a negative probe, the positive test above only proves the plugin is currently
+// clean, not that 6d actually fires.
+test('validate-plugin.sh Check 6d catches an unanchored reference-subdir path (negative probe)', () => {
+  const probe = join(pluginRoot, 'agents', '__rx_p2_6d_probe__.md');
+  writeFileSync(
+    probe,
+    [
+      '---',
+      'name: rx-p2-6d-probe',
+      'description: temporary probe for the 6d negative test',
+      'model: opus',
+      'color: blue',
+      'tools: ["Read"]',
+      '---',
+      '',
+      '# Probe',
+      '',
+      'Load: `responsible-ai/ai-act-compliance-guide.md`',
+      '',
+    ].join('\n'),
+  );
+  try {
+    let failed = false;
+    let out = '';
+    try {
+      out = execFileSync('bash', [script], { encoding: 'utf8', cwd: pluginRoot });
+    } catch (err) {
+      failed = true;
+      out = `${err.stdout || ''}${err.stderr || ''}`;
+    }
+    assert.ok(failed, 'validate-plugin.sh must exit non-zero when a bare reference-subdir path is present');
+    assert.match(out, /bare reference-subdir path/);
+    assert.match(out, /__rx_p2_6d_probe__\.md/);
+  } finally {
+    rmSync(probe, { force: true });
+  }
+});
--- a/tests/validate-plugin.sh
+++ b/tests/validate-plugin.sh
@ -349,6 +349,33 @@ else
  pass "No sonnet/haiku pinned in Task delegations"
 fi

+# 6d: every bundled reference-subdir path must be ${CLAUDE_PLUGIN_ROOT}-anchored.
+#     Short-forms like `responsible-ai/foo.md` (a reference subdir + file, without the
+#     full skills/<skill>/references/ prefix) resolve only in dev cwd -- installed mode
+#     breaks. A path is anchored iff the subdir is immediately preceded by '/'
+#     (.../references/<subdir>/). The (^|[^A-Za-z0-9._/-]) guard fires on bare short-forms
+#     (preceded by backtick/space/comma/colon) while skipping anchored paths (preceded
+#     by '/') and compound subdir names (e.g. 'architecture' inside 'rag-architecture',
+#     preceded by '-').
+REF_SUBDIRS='architecture|copilot-extensibility|development|platforms|prompt-engineering|agent-orchestration|api-management|azure-ai-services|data-engineering|mlops-genaiops|multi-modal|rag-architecture|monitoring-observability|norwegian-public-sector-governance|responsible-ai|bcdr|hybrid-edge|ai-security-engineering|cost-optimization|performance-scalability'
+subdir_offenders=0
+for f in "$PLUGIN_ROOT"/commands/*.md "$PLUGIN_ROOT"/agents/*.md; do
+  [ -f "$f" ] || continue
+  rel="${f#$PLUGIN_ROOT/}"
+  bare="$(grep -nE "(^|[^A-Za-z0-9._/-])($REF_SUBDIRS)/[A-Za-z0-9._-]+\.md" "$f" || true)"
+  if [ -n "$bare" ]; then
+    while IFS= read -r bline; do
+      [ -n "$bline" ] || continue
+      lineno="${bline%%:*}"
+      fail "$rel:$lineno: bare reference-subdir path (needs \${CLAUDE_PLUGIN_ROOT}/skills/<skill>/references/ prefix for installed mode)"
+      subdir_offenders=$((subdir_offenders + 1))
+    done <<< "$bare"
+  fi
+done
+if [ "$subdir_offenders" -eq 0 ]; then
+  pass "All reference-subdir paths are \${CLAUDE_PLUGIN_ROOT}-anchored"
+fi
+
 echo ""

 # -------------------------------------------------------