Repair hallucinated Semantic Consolidator source IDs

src/meeting_lab/consolidation/consolidate_facts.py

@@ -600,11 +600,12 @@ def repair_model_group_coverage(
     """
     Apply deterministic source-coverage repairs to model grouping JSON.
 
-    The repair is intentionally conservative. Repeated source IDs are removed
-    after their first occurrence, empty groups created by that removal are
-    dropped, and missing facts are restored as singleton groups using the
-    original canonicalized fact text. No existing group text, merge reason or
-    semantic merge is rewritten.
+    The repair is intentionally conservative. Unknown source IDs are removed
+    without guessing a replacement. Repeated valid source IDs are removed after
+    their first occurrence, empty groups created by either removal are dropped,
+    and missing facts are restored as singleton groups using the original
+    canonicalized fact text. No existing group text, merge reason or semantic
+    merge is rewritten.
     """
     groups = model_output.get("groups")
     if not isinstance(groups, list):
@@ -625,9 +626,19 @@ def repair_model_group_coverage(
             continue
         kept_ids: list[str] = []
         for id_index, item_id in enumerate(source_ids):
-            if not isinstance(item_id, str) or item_id not in expected_ids:
+            if not isinstance(item_id, str):
                 kept_ids.append(item_id)
                 continue
+            if item_id not in expected_ids:
+                changes.append(
+                    {
+                        "operation": "remove_unknown_source_id",
+                        "id": item_id,
+                        "group_index": group_index,
+                        "id_index": id_index,
+                    }
+                )
+                continue
             if item_id in seen:
                 changes.append(
                     {