feat(example): add per-entity LLM evaluations for 985 WoN entities (S3.3)

Batch evaluation of all 988 entities via OpenRouter. 984 succeeded on
first pass; 3 failed (network errors). eval-summary --update-metrics
written with per_entity_mean=3.9556.

Viability dashboard: 6/6 PASS
  redundancy_ratio   0.0061  (max 0.10)
  coverage_ratio     0.6190  (min 0.40)
  coherence_comps    0.0000  (max 3)
  consistency_cycles 0.0000  (max 0)
  granularity_entropy 2.6748 (min 1.0)
  per_entity_mean    3.9556  (min 3.5)

Dimension breakdown (mean across 985 entities):
  definition_precision  3.62
  source_grounding      4.36
  domain_placement      4.56
  vsm_relevance         3.31
  explanatory_value     3.94

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

examples/infospace-with-history/output/evaluations/tax_administration_systems.md

@@ -0,0 +1,62 @@
+---
+entity_slug: tax_administration_systems
+evaluator: null
+evaluated_at: '2026-02-23T06:29:23.055639'
+overall_score: 2.4
+scores:
+- name: definition_precision
+  value: 1.0
+  max_value: 5.0
+  rationale: There is no definition provided at all, making this entity completely
+    imprecise. Without any definitional content, it's impossible to assess whether
+    the concept is distinct or merely a vague umbrella term.
+- name: source_grounding
+  value: 2.0
+  max_value: 5.0
+  rationale: While Smith does discuss tax collection and administration in "The Wealth
+    of Nations," the entity lacks specific grounding since no source chapter or context
+    is provided. The concept likely exists in the source but this entity fails to
+    demonstrate that connection.
+- name: domain_placement
+  value: 3.0
+  max_value: 5.0
+  rationale: Tax administration systems would logically belong in a public finance
+    or fiscal policy domain within economic theory, which seems appropriate for Smith's
+    work. However, without a specified domain, it's difficult to assess whether the
+    placement is optimal.
+- name: vsm_relevance
+  value: 4.0
+  max_value: 5.0
+  rationale: Tax administration systems map well to VSM System 3 (internal regulation/audit)
+    as they involve the operational control and monitoring of tax collection processes.
+    This represents a clear structural function within a viable economic system.
+- name: explanatory_value
+  value: 2.0
+  max_value: 5.0
+  rationale: While tax administration is mechanistically important for understanding
+    how governments collect revenue, this entity as presented offers no explanatory
+    insight due to its lack of definition and context. It names a phenomenon without
+    illuminating any structural relations or mechanisms.
+---
+
+# Evaluation: Tax Administration Systems
+
+## definition_precision — 1.0 / 5.0
+
+There is no definition provided at all, making this entity completely imprecise. Without any definitional content, it's impossible to assess whether the concept is distinct or merely a vague umbrella term.
+
+## source_grounding — 2.0 / 5.0
+
+While Smith does discuss tax collection and administration in "The Wealth of Nations," the entity lacks specific grounding since no source chapter or context is provided. The concept likely exists in the source but this entity fails to demonstrate that connection.
+
+## domain_placement — 3.0 / 5.0
+
+Tax administration systems would logically belong in a public finance or fiscal policy domain within economic theory, which seems appropriate for Smith's work. However, without a specified domain, it's difficult to assess whether the placement is optimal.
+
+## vsm_relevance — 4.0 / 5.0
+
+Tax administration systems map well to VSM System 3 (internal regulation/audit) as they involve the operational control and monitoring of tax collection processes. This represents a clear structural function within a viable economic system.
+
+## explanatory_value — 2.0 / 5.0
+
+While tax administration is mechanistically important for understanding how governments collect revenue, this entity as presented offers no explanatory insight due to its lack of definition and context. It names a phenomenon without illuminating any structural relations or mechanisms.