feat(environment): integrate llm_pipeline call in step() with mock fallback

server/environment.py

@@ -68,4 +68,18 @@ class RedTeamEnvironment:
             {"role": "user", "content": h["framing"]}
             for h in self.attack_history[-5:]
         ]
+
+        # ------ Step 3: Call Person 3's LLM pipeline ------
+        if self.llm_pipeline:
+            llm_result = await self.llm_pipeline(action, conversation_history)
+        else:
+            # Mock — lets you run the server before P3 is ready
+            logger.warning("LLM pipeline not wired — using mock response")
+            llm_result = {
+                "defender_response": "[Mock] I cannot help with that request.",
+                "attack_success":    0.1,
+                "safety_flagged":    False,
+                "defense_score":     0.9,
+                "llm_grade":         {"overall": 3.0},
+            }
         return None # Placeholder