diff --git a/plugins/memory/hindsight/__init__.py b/plugins/memory/hindsight/__init__.py
index bc82bc40fb..39dfe94f6c 100644
--- a/plugins/memory/hindsight/__init__.py
+++ b/plugins/memory/hindsight/__init__.py
@@ -1231,9 +1231,19 @@ class HindsightMemoryProvider(MemoryProvider):
         if self._client is not None:
             try:
                 if self._mode == "local_embedded":
-                    # Use the public close() API. The RuntimeError from
-                    # aiohttp's "attached to a different loop" is expected
-                    # and harmless — the daemon keeps running independently.
+                    # HindsightEmbedded.close() delegates to its sync client.close().
+                    # When Hermes created/used that client on the shared async loop,
+                    # closing it from this thread can raise "attached to a different
+                    # loop" before aiohttp releases the session. Close the embedded
+                    # inner async client on the shared loop first, then let the
+                    # wrapper clean up daemon/UI bookkeeping.
+                    inner_client = getattr(self._client, "_client", None)
+                    if inner_client is not None and hasattr(inner_client, "aclose"):
+                        _run_sync(inner_client.aclose())
+                        try:
+                            self._client._client = None
+                        except Exception:
+                            pass
                     try:
                         self._client.close()
                     except RuntimeError:
diff --git a/tests/plugins/memory/test_hindsight_provider.py b/tests/plugins/memory/test_hindsight_provider.py
index 5f1290b2f1..2f123b6f05 100644
--- a/tests/plugins/memory/test_hindsight_provider.py
+++ b/tests/plugins/memory/test_hindsight_provider.py
@@ -1102,3 +1102,22 @@ class TestSharedEventLoopLifecycle:
 
         mock_client.aclose.assert_called_once()
         assert provider._client is None
+
+
+class TestShutdown:
+    def test_local_embedded_shutdown_closes_inner_async_client_on_shared_loop(self, provider):
+        inner_client = _make_mock_client()
+        embedded = MagicMock()
+        embedded._client = inner_client
+        embedded.close = MagicMock()
+
+        provider._mode = "local_embedded"
+        provider._client = embedded
+
+        provider.shutdown()
+
+        inner_client.aclose.assert_awaited_once()
+        embedded.close.assert_called_once()
+        assert embedded._client is None
+        assert provider._client is None
+