fix: audit fixes - aegis_env/environment.py

aegis_env/environment.py

@@ -59,7 +59,7 @@ class AEGISEnvironment(Environment):
         # Ground truth tracks the expected correct response for RL signals
         self.ground_truth = {
             "decision": "ALLOW",
-            "violation_type": "None",
+            "violation_type": "none",
             "applicable_rules": [],
             "suspicious_apis_caught": False,
         }
@@ -173,7 +173,7 @@ class AEGISEnvironment(Environment):
 
         # Calculate Reward using validated action dict + current level for dynamic weights
         level = CurriculumScheduler.get_level(self.step_count)
-        reward = self.rewarder.calculate_total(action_dict, self.ground_truth, "", level=level)
+        reward = self.rewarder.calculate_total(action_dict, self.ground_truth, action_dict.get("explanation", ""), level=level)
 
         # BUG-4 FIX: Log confirmed violations to memory for compound-attack detection (CMP-01)
         if self.memory and self.ground_truth.get("violation_type", "none") != "none":