diff --git a/README.md b/README.md
index 3534d5a..6af1a2c 100644
--- a/README.md
+++ b/README.md
@@ -651,7 +651,7 @@ node scripts/kb-update/build-registry.mjs [--merge]
 /architect:generate-skills
 ```
 
-Category-to-skill routing is defined in `scripts/skill-gen/category-skill-map.json` (20 categories mapped to 5 skills), used by the generate-skills workflow to place new reference documents in the correct skill directory.
+Category-to-skill routing is defined in `scripts/kb-update/data/domain-taxonomy.json` (the lag-0 single source of truth, read via `lib/taxonomy.mjs` → `getCategorySkill`). It uses physical-disk truth and supersedes the old `category-skill-map.json` (removed — it had drifted from disk and was consumed by no code).
 
 ---
 
diff --git a/scripts/skill-gen/category-skill-map.json b/scripts/skill-gen/category-skill-map.json
deleted file mode 100644
index 6dffe04..0000000
--- a/scripts/skill-gen/category-skill-map.json
+++ /dev/null
@@ -1,32 +0,0 @@
-{
-  "version": "1.0",
-  "description": "Maps KB categories to their target skill directories",
-  "mapping": {
-    "rag-architecture": "ms-ai-engineering",
-    "azure-ai-services": "ms-ai-engineering",
-    "copilot-extensibility": "ms-ai-engineering",
-    "prompt-engineering": "ms-ai-engineering",
-    "data-engineering": "ms-ai-engineering",
-    "api-management": "ms-ai-engineering",
-    "agent-orchestration": "ms-ai-engineering",
-    "multi-modal": "ms-ai-engineering",
-    "mlops-genaiops": "ms-ai-engineering",
-    "performance-scalability": "ms-ai-engineering",
-    "monitoring-observability": "ms-ai-engineering",
-    "responsible-ai": "ms-ai-governance",
-    "norwegian-public-sector-governance": "ms-ai-governance",
-    "cost-optimization": "ms-ai-security",
-    "ai-security-engineering": "ms-ai-security",
-    "security-scoring": "ms-ai-security",
-    "hybrid-edge": "ms-ai-infrastructure",
-    "bcdr": "ms-ai-infrastructure",
-    "platforms": "ms-ai-advisor",
-    "architecture": "ms-ai-advisor"
-  },
-  "priority_thresholds": {
-    "critical": 30,
-    "high": 60,
-    "medium": 90,
-    "low": 180
-  }
-}