feat(honcho): make latency-adding paths configurable

queryRewrite (default off) gates the latest-message rewrite so the
extra auxiliary LLM call is opt-in. firstTurnBaseWait and
firstTurnDialecticWait expose the turn-1 bounded waits in seconds
(0 disables). All three resolve host-block-first like every other
field. Also pins per-host timeout resolution with tests.

tests/honcho_plugin/test_query_rewrite.py

@@ -100,6 +100,7 @@ def test_long_input_keeps_both_ends_with_a_hard_bound():
 
 def _provider(query_rewriter, *, depth=1):
     provider = HonchoMemoryProvider(query_rewriter=query_rewriter)
+    provider._query_rewrite_enabled = True
     provider._manager = MagicMock()
     provider._manager.dialectic_query.return_value = "memory synthesis"
     provider._session_key = "test-session"
@@ -189,6 +190,7 @@ def test_first_user_message_is_not_shadowed_by_generic_dialectic_prewarm():
         enabled=True,
         recall_mode="hybrid",
         timeout=1,
+        query_rewrite=True,
     )
 
     with (
@@ -237,3 +239,24 @@ def test_query_rewrite_has_an_independent_auxiliary_model_config():
     assert task_config["provider"] == "auto"
     assert task_config["timeout"] == 8
     assert TASK_KEY in {key for key, _name, _description in _AUX_TASKS}
+
+
+def test_query_rewrite_disabled_by_default():
+    """queryRewrite defaults OFF — the rewriter must not add an LLM call."""
+    rewriter = MagicMock(return_value="What does the user prefer?")
+    provider = _provider(rewriter)
+    provider._query_rewrite_enabled = False
+
+    provider._run_dialectic_depth("what did we decide?")
+
+    rewriter.assert_not_called()
+    provider._manager.dialectic_query.assert_called_once()
+
+
+def test_config_defaults_keep_latency_additions_off():
+    from plugins.memory.honcho.client import HonchoClientConfig
+
+    cfg = HonchoClientConfig(api_key="k", enabled=True)
+    assert cfg.query_rewrite is False
+    assert cfg.first_turn_base_wait == 3.0
+    assert cfg.first_turn_dialectic_wait == 2.0

tests/test_honcho_client_config.py

@@ -125,3 +125,50 @@ def test_save_config_sets_owner_only_permissions(tmp_path, monkeypatch):
     assert config_file.exists()
     mode = stat.S_IMODE(config_file.stat().st_mode)
     assert mode == 0o600, f"Expected 0o600 (owner-only), got {oct(mode)}"
+
+
+class TestLatencyFlagResolution:
+    def test_defaults(self, tmp_path, monkeypatch):
+        monkeypatch.delenv('HONCHO_BASE_URL', raising=False)
+        config_path = tmp_path / 'config.json'
+        config_path.write_text(json.dumps({'apiKey': 'k'}))
+        cfg = HonchoClientConfig.from_global_config(config_path=config_path)
+        assert cfg.query_rewrite is False
+        assert cfg.first_turn_base_wait == 3.0
+        assert cfg.first_turn_dialectic_wait == 2.0
+
+    def test_host_block_wins(self, tmp_path, monkeypatch):
+        monkeypatch.delenv('HONCHO_BASE_URL', raising=False)
+        config_path = tmp_path / 'config.json'
+        config_path.write_text(json.dumps({
+            'apiKey': 'k',
+            'queryRewrite': False,
+            'firstTurnBaseWait': 3,
+            'hosts': {'hermes': {
+                'queryRewrite': True,
+                'firstTurnBaseWait': 0,
+                'firstTurnDialecticWait': 0.5,
+            }},
+        }))
+        cfg = HonchoClientConfig.from_global_config(config_path=config_path)
+        assert cfg.query_rewrite is True
+        assert cfg.first_turn_base_wait == 0.0
+        assert cfg.first_turn_dialectic_wait == 0.5
+
+    def test_per_host_timeout_wins_over_global(self, tmp_path, monkeypatch):
+        monkeypatch.delenv('HONCHO_TIMEOUT', raising=False)
+        config_path = tmp_path / 'config.json'
+        config_path.write_text(json.dumps({
+            'apiKey': 'k',
+            'timeout': 30,
+            'hosts': {'hermes': {'timeout': 5}},
+        }))
+        cfg = HonchoClientConfig.from_global_config(config_path=config_path)
+        assert cfg.timeout == 5.0
+
+    def test_timeout_falls_back_to_global(self, tmp_path, monkeypatch):
+        monkeypatch.delenv('HONCHO_TIMEOUT', raising=False)
+        config_path = tmp_path / 'config.json'
+        config_path.write_text(json.dumps({'apiKey': 'k', 'timeout': 30}))
+        cfg = HonchoClientConfig.from_global_config(config_path=config_path)
+        assert cfg.timeout == 30.0

tests/test_honcho_startup_fail_open.py

@@ -25,6 +25,9 @@ def _configured_hybrid_config() -> _FakeHonchoConfig:
         injection_frequency="every-turn",
         context_cadence=1,
         dialectic_cadence=1,
+        query_rewrite=False,
+        first_turn_base_wait=3.0,
+        first_turn_dialectic_wait=2.0,
         dialectic_depth=1,
         dialectic_depth_levels=None,
         reasoning_heuristic=True,