test: preserve metadata and passthrough assertions after cap removal

tests/agent/test_models_dev.py

@@ -1144,8 +1144,8 @@ class TestModelOverrides:
     def test_model_info_override_for_unknown_model(self):
         """Canonical-schema override provides metadata for an unknown model.
 
-        Same key space as every other consumer — context_window,
-        max_output_tokens, supports_* — NOT the internal catalog shape.
+        Context and capabilities remain configurable; a legacy output override
+        cannot displace the unknown-model metadata fallback.
         """
         overrides = {
             "custom:my-vllm": {
@@ -1161,10 +1161,13 @@ class TestModelOverrides:
         with self._setup_overrides(overrides), \
              patch("agent.models_dev.fetch_models_dev", return_value={}):
             info = get_model_info("custom:my-vllm", "my-llava-model")
+            del overrides["custom:my-vllm"]["my-llava-model"]["max_output_tokens"]
+            uncapped_info = get_model_info("custom:my-vllm", "my-llava-model")
         assert info is not None
         assert info.family == "llava"
         assert info.context_window == 8192
-        assert info.max_output == 0
+        assert uncapped_info is not None
+        assert info.max_output == uncapped_info.max_output > 0
         assert info.tool_call is True
         assert info.reasoning is False
 

tests/tools/test_delegate.py

@@ -1046,7 +1046,7 @@ class TestDelegationCredentialResolution(unittest.TestCase):
     @patch("hermes_cli.runtime_provider.resolve_runtime_provider")
     def test_base_url_with_provider_carries_runtime_request_overrides(self, mock_resolve):
         """#65035: the base_url short-circuit must not drop the configured
-        provider's request_overrides / max_output_tokens."""
+        provider's generic request_overrides; dedicated output caps are ignored."""
         mock_resolve.return_value = {
             "provider": "custom",
             "base_url": "https://provider-default.example/v1",
@@ -1071,7 +1071,7 @@ class TestDelegationCredentialResolution(unittest.TestCase):
             creds["request_overrides"],
             {"extra_body": {"thinking": {"type": "disabled"}}},
         )
-        self.assertEqual(creds["max_output_tokens"], 8192)
+        self.assertNotIn("max_output_tokens", creds)
 
     def test_bare_base_url_returns_none_overrides(self):
         """No provider alongside base_url → no overrides source; keys are
@@ -1080,7 +1080,7 @@ class TestDelegationCredentialResolution(unittest.TestCase):
         cfg = {"model": "m", "provider": "", "base_url": "http://localhost:1234/v1", "api_key": "k"}
         creds = _resolve_delegation_credentials(cfg, parent)
         self.assertIsNone(creds["request_overrides"])
-        self.assertIsNone(creds["max_output_tokens"])
+        self.assertNotIn("max_output_tokens", creds)
 
     @patch("hermes_cli.runtime_provider.resolve_runtime_provider")
     def test_base_url_survives_runtime_resolution_failure(self, mock_resolve):
@@ -1093,7 +1093,7 @@ class TestDelegationCredentialResolution(unittest.TestCase):
         creds = _resolve_delegation_credentials(cfg, parent)
         self.assertEqual(creds["base_url"], "https://api.xiaomimimo.com/v1")
         self.assertIsNone(creds["request_overrides"])
-        self.assertIsNone(creds["max_output_tokens"])
+        self.assertNotIn("max_output_tokens", creds)
 
     @patch("hermes_cli.runtime_provider.resolve_runtime_provider")
     def test_provider_resolution_failure_raises_valueerror(self, mock_resolve):