fix: dedup memory consolidation with existing memory search (#105)

_maybe_consolidate() now checks get_memories(subject=agent_id)
before storing. Skips if a memory of the same type (pattern/anomaly)
was created within the last hour. Prevents duplicate consolidation
entries on repeated task completion/failure events.

Also restructured branching: neutral success rates (0.3-0.8) now
return early instead of falling through.

9 new tests. 1465 total passing.

src/spark/engine.py

@@ -273,6 +273,8 @@ class SparkEngine:
 
     def _maybe_consolidate(self, agent_id: str) -> None:
         """Consolidate events into memories when enough data exists."""
+        from datetime import UTC, datetime, timedelta
+
         agent_events = spark_memory.get_events(agent_id=agent_id, limit=50)
         if len(agent_events) < 5:
             return
@@ -286,7 +288,34 @@ class SparkEngine:
 
         success_rate = len(completions) / total if total else 0
 
+        # Determine target memory type based on success rate
         if success_rate >= 0.8:
+            target_memory_type = "pattern"
+        elif success_rate <= 0.3:
+            target_memory_type = "anomaly"
+        else:
+            return  # No consolidation needed for neutral success rates
+
+        # Check for recent memories of the same type for this agent
+        existing_memories = spark_memory.get_memories(subject=agent_id, limit=5)
+        now = datetime.now(UTC)
+        one_hour_ago = now - timedelta(hours=1)
+
+        for memory in existing_memories:
+            if memory.memory_type == target_memory_type:
+                try:
+                    created_at = datetime.fromisoformat(memory.created_at)
+                    if created_at >= one_hour_ago:
+                        logger.info(
+                            "Consolidation: skipping — recent memory exists for %s",
+                            agent_id[:8],
+                        )
+                        return
+                except (ValueError, TypeError):
+                    continue
+
+        # Store the new memory
+        if target_memory_type == "pattern":
             spark_memory.store_memory(
                 memory_type="pattern",
                 subject=agent_id,
@@ -295,7 +324,7 @@ class SparkEngine:
                 confidence=min(0.95, 0.6 + total * 0.05),
                 source_events=total,
             )
-        elif success_rate <= 0.3:
+        else:  # anomaly
             spark_memory.store_memory(
                 memory_type="anomaly",
                 subject=agent_id,