diff --git a/core/wren/src/wren/memory/schema_indexer.py b/core/wren/src/wren/memory/schema_indexer.py
index 7e28cd44fa..92dfbcc9cc 100644
--- a/core/wren/src/wren/memory/schema_indexer.py
+++ b/core/wren/src/wren/memory/schema_indexer.py
@@ -228,16 +228,30 @@ def extract_schema_items(manifest: dict) -> list[dict]:
     mdl_h = manifest_hash(manifest)
     items: list[dict] = []
 
-    for model in manifest.get("models", []):
-        items.append(_model_record(model, mdl_h, now))
-        for col in model.get("columns", []):
-            items.append(_column_record(col, model["name"], mdl_h, now))
-
-    for rel in manifest.get("relationships", []):
-        items.append(_relationship_record(rel, mdl_h, now))
-
-    for view in manifest.get("views", []):
-        items.append(_view_record(view, mdl_h, now))
+    models = manifest.get("models", []) or []
+    if isinstance(models, list):
+        for model in models:
+            if not isinstance(model, dict) or model.get("name") is None:
+                continue
+            items.append(_model_record(model, mdl_h, now))
+            cols = model.get("columns", []) or []
+            if not isinstance(cols, list):
+                continue
+            for col in cols:
+                if isinstance(col, dict) and col.get("name") is not None:
+                    items.append(_column_record(col, model["name"], mdl_h, now))
+
+    rels = manifest.get("relationships", []) or []
+    if isinstance(rels, list):
+        for rel in rels:
+            if isinstance(rel, dict) and rel.get("name") is not None:
+                items.append(_relationship_record(rel, mdl_h, now))
+
+    views = manifest.get("views", []) or []
+    if isinstance(views, list):
+        for view in views:
+            if isinstance(view, dict) and view.get("name") is not None:
+                items.append(_view_record(view, mdl_h, now))
 
     cubes = manifest.get("cubes", []) or []
     if isinstance(cubes, list):
@@ -264,8 +278,13 @@ def extract_schema_items(manifest: dict) -> list[dict]:
 
 def _model_record(model: dict, mdl_h: str, now: datetime) -> dict:
     name = model["name"]
-    cols = model.get("columns", [])
-    col_summaries = ", ".join(f"{c['name']} ({c.get('type', '?')})" for c in cols[:20])
+    cols = model.get("columns", []) or []
+    if not isinstance(cols, list):
+        cols = []
+    safe_cols = [c for c in cols if isinstance(c, dict) and c.get("name") is not None]
+    col_summaries = ", ".join(
+        f"{c['name']} ({c.get('type', '?')})" for c in safe_cols[:20]
+    )
     pk = model.get("primaryKey") or ""
 
     description = _prop_description(model)
diff --git a/core/wren/tests/unit/test_schema_extract_nonduct_models.py b/core/wren/tests/unit/test_schema_extract_nonduct_models.py
new file mode 100644
index 0000000000..456087ed4a
--- /dev/null
+++ b/core/wren/tests/unit/test_schema_extract_nonduct_models.py
@@ -0,0 +1,27 @@
+from wren.memory.schema_indexer import extract_schema_items
+
+
+def test_extract_skips_nonduct_models_and_columns():
+    items = extract_schema_items(
+        {
+            "models": [
+                {
+                    "name": "orders",
+                    "columns": [
+                        {"name": "id", "type": "int"},
+                        "bad",
+                        {"type": "str"},  # missing name
+                    ],
+                },
+                "nope",
+            ],
+            "relationships": ["x", {"name": "r", "models": ["a", "b"]}],
+            "views": [1, {"name": "v", "statement": "SELECT 1"}],
+        }
+    )
+    kinds = {(i.get("item_type"), i.get("item_name")) for i in items}
+    assert ("model", "orders") in kinds
+    assert ("column", "id") in kinds
+    assert ("relationship", "r") in kinds
+    assert ("view", "v") in kinds
+    assert all(i.get("item_name") != "nope" for i in items)