fix: run CLI new-session memory flush off-thread

cli.py

@@ -4127,6 +4127,7 @@ class HermesCLI(CLIAgentSetupMixin, CLICommandsMixin):
         # Background task tracking: {task_id: threading.Thread}
         self._background_tasks: Dict[str, threading.Thread] = {}
         self._background_task_counter = 0
+        self._last_session_boundary_thread: Optional[threading.Thread] = None
 
     def _claim_active_session(self, surface: str = "cli", *, stderr: bool = False) -> bool:
         """Claim a global active-session slot for this CLI process."""
@@ -6925,17 +6926,50 @@ class HermesCLI(CLIAgentSetupMixin, CLICommandsMixin):
             )
             return False
 
+    def _launch_session_boundary_memory_flush(
+        self,
+        history_snapshot: list,
+        *,
+        session_id: Optional[str] = None,
+    ) -> None:
+        """Kick old-session memory extraction off-thread so /new is responsive."""
+        self._last_session_boundary_thread = None
+        agent = getattr(self, "agent", None)
+        if not agent or not history_snapshot:
+            return
+
+        def _run_boundary_flush() -> None:
+            if hasattr(agent, "commit_memory_session"):
+                try:
+                    agent.commit_memory_session(history_snapshot, session_id=session_id)
+                except Exception:
+                    pass
+
+        thread = threading.Thread(
+            target=_run_boundary_flush,
+            daemon=True,
+            name=f"session-boundary-flush-{(session_id or 'unknown')[:24]}",
+        )
+        self._last_session_boundary_thread = thread
+        thread.start()
+
     def new_session(self, silent=False, title=None):
         """Start a fresh session with a new session ID and cleared agent state."""
+        old_session_id = self.session_id
         if self.agent and self.conversation_history:
-            # Trigger memory extraction on the old session before session_id rotates.
-            self.agent.commit_memory_session(self.conversation_history)
+            # Trigger memory extraction on the old session without blocking
+            # session rotation. Snapshot the history and old session id first:
+            # the background thread may run after ``self.agent.session_id`` is
+            # updated to the new session below.
+            self._launch_session_boundary_memory_flush(
+                list(self.conversation_history),
+                session_id=old_session_id,
+            )
             self._notify_session_boundary("on_session_finalize")
         elif self.agent:
             # First session or empty history — still finalize the old session
             self._notify_session_boundary("on_session_finalize")
 
-        old_session_id = self.session_id
         if self._session_db and old_session_id:
             # Flush any un-persisted messages from the current turn to the
             # old session *before* rotating.  /new can be called mid-turn

run_agent.py

@@ -3307,11 +3307,18 @@ class AIAgent:
             except Exception:
                 pass
 
-    def commit_memory_session(self, messages: list = None) -> None:
+    def commit_memory_session(
+        self,
+        messages: list = None,
+        session_id: Optional[str] = None,
+    ) -> None:
         """Trigger end-of-session extraction without tearing providers down.
         Called when session_id rotates (e.g. /new, context compression);
         providers keep their state and continue running under the old
         session_id — they just flush pending extraction now."""
+        ended_session_id = (
+            session_id if session_id is not None else (self.session_id or "")
+        )
         if self._memory_manager:
             try:
                 self._memory_manager.on_session_end(messages or [])
@@ -3326,7 +3333,7 @@ class AIAgent:
         if hasattr(self, "context_compressor") and self.context_compressor:
             try:
                 self.context_compressor.on_session_end(
-                    self.session_id or "",
+                    ended_session_id,
                     messages or [],
                 )
             except Exception:

tests/cli/test_cli_new_session.py

@@ -156,6 +156,10 @@ def test_new_command_creates_real_fresh_session_and_resets_agent_state(tmp_path)
 
     cli.process_command("/new")
 
+    boundary_thread = getattr(cli, "_last_session_boundary_thread", None)
+    if boundary_thread is not None:
+        boundary_thread.join(timeout=1)
+
     assert cli.session_id != old_session_id
 
     old_session = cli._session_db.get_session(old_session_id)
@@ -175,6 +179,46 @@ def test_new_command_creates_real_fresh_session_and_resets_agent_state(tmp_path)
     cli.agent._invalidate_system_prompt.assert_called_once()
 
 
+def test_new_session_dispatches_memory_flush_on_background_thread(tmp_path):
+    cli = _prepare_cli_with_active_session(tmp_path)
+    old_session_id = cli.session_id
+
+    import cli as _cli_mod
+
+    started = {}
+
+    class _DummyThread:
+        def __init__(self, *, target=None, args=(), kwargs=None, daemon=None, name=None):
+            started["target"] = target
+            started["args"] = args
+            started["kwargs"] = kwargs or {}
+            started["daemon"] = daemon
+            started["name"] = name
+            started["started"] = False
+
+        def start(self):
+            started["started"] = True
+
+        def join(self, timeout=None):
+            return None
+
+    with patch.object(_cli_mod.threading, "Thread", _DummyThread):
+        cli.process_command("/new")
+
+    assert started["started"] is True
+    assert callable(started["target"])
+    assert started["daemon"] is True
+    assert started["name"] == f"session-boundary-flush-{old_session_id[:24]}"
+    cli.agent.commit_memory_session.assert_not_called()
+
+    started["target"]()
+
+    cli.agent.commit_memory_session.assert_called_once_with(
+        [{"role": "user", "content": "hello"}],
+        session_id=old_session_id,
+    )
+
+
 def test_new_command_rotates_hermes_session_id_env_and_context(tmp_path):
     from gateway.session_context import _VAR_MAP, get_session_env
 

tests/run_agent/test_commit_memory_session_context_engine.py

@@ -56,6 +56,19 @@ def test_commit_memory_session_with_no_messages_passes_empty_list():
     ctx.on_session_end.assert_called_once_with("sess-7", [])
 
 
+def test_commit_memory_session_can_finalize_explicit_old_session_id():
+    """Async callers may run after agent.session_id already changed."""
+    mm = MagicMock()
+    ctx = MagicMock()
+    agent = _make_minimal_agent(mm, ctx, session_id="new-session")
+
+    msgs = [{"role": "user", "content": "old turn"}]
+    agent.commit_memory_session(msgs, session_id="old-session")
+
+    mm.on_session_end.assert_called_once_with(msgs)
+    ctx.on_session_end.assert_called_once_with("old-session", msgs)
+
+
 def test_commit_memory_session_no_memory_manager_still_notifies_context_engine():
     """If only the context engine is configured, it still gets the hook."""
     ctx = MagicMock()