diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 1b3002d..12ab684 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -2,6 +2,9 @@ name: Tests
 
 on:
   push:
+  pull_request:
+    branches:
+      - main
 
 jobs:
   test:
diff --git a/nwb_linkml/src/nwb_linkml/adapters/classes.py b/nwb_linkml/src/nwb_linkml/adapters/classes.py
index c700d53..fcdf43c 100644
--- a/nwb_linkml/src/nwb_linkml/adapters/classes.py
+++ b/nwb_linkml/src/nwb_linkml/adapters/classes.py
@@ -133,7 +133,24 @@ class ClassAdapter(Adapter):
         """The full name of the object in the generated linkml
 
         Distinct from 'name' which is the thing that's used to define position in
-        a hierarchical data setting
+        a hierarchical data setting.
+
+        Combines names from ``parent``, if present, using ``"__"`` .
+        Rather than concatenating the full series of names with ``__`` like
+
+        * ``Parent``
+        * ``Parent__child1``
+        * ``Parent__child1__child2``
+
+        we only keep the last parent, so
+
+        * ``Parent``
+        * ``Parent__child1``
+        * ``child1__child2``
+
+        The assumption is that a child name may not be unique, but the combination of
+        a parent/child pair should be unique enough to avoid name shadowing without
+        making humongous and cumbersome names.
         """
         if self.cls.neurodata_type_def:
             name = self.cls.neurodata_type_def
@@ -141,7 +158,8 @@ class ClassAdapter(Adapter):
             # not necessarily a unique name, so we combine parent names
             name_parts = []
             if self.parent is not None:
-                name_parts.append(self.parent._get_full_name())
+                parent_name = self.parent._get_full_name().split("__")[-1]
+                name_parts.append(parent_name)
 
             name_parts.append(self.cls.name)
             name = "__".join(name_parts)