feat(cli): bootstrap default configs on CLI startup (#401)

* feat(cli): bootstrap default configs on command run

* fix(cli): use active interpreter in bootstrap warning

* refactor(cli): simplify bootstrap warning flow

* refactor(cli): bootstrap defaults in main entrypoint

* refactor(cli): keep bootstrap ownership in main

* test(cli): cover lazy dispatch and runtime failure flag

* refactor(cli): remove redundant bootstrap state

* test(cli): assert bootstrap warning includes error

* test: address cli bootstrap review feedback

Author: johnnygreco

packages/data-designer/src/data_designer/cli/lazy_group.py

@@ -60,9 +60,6 @@ def make_context(
     ) -> click.Context:
         return self._resolve().make_context(info_name, args, parent, **extra)
 
-    def invoke(self, ctx: click.Context) -> Any:
-        return self._resolve().invoke(ctx)
-
 
 def create_lazy_typer_group(
     lazy_subcommands: dict[str, dict[str, str]],

packages/data-designer/src/data_designer/cli/main.py

@@ -6,6 +6,7 @@
 import typer
 
 from data_designer.cli.lazy_group import create_lazy_typer_group
+from data_designer.cli.runtime import ensure_cli_default_model_settings
 
 _CMD = "data_designer.cli.commands"
 
@@ -105,6 +106,7 @@
 
 def main() -> None:
     """Main entry point for the CLI."""
+    ensure_cli_default_model_settings()
     app()
 
 

packages/data-designer/src/data_designer/cli/runtime.py

@@ -0,0 +1,24 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from data_designer.cli.ui import print_warning
+from data_designer.config.default_model_settings import resolve_seed_default_model_settings
+
+
+def ensure_cli_default_model_settings() -> None:
+    """Best-effort bootstrap for CLI default model settings.
+
+    Repeated calls are safe because ``resolve_seed_default_model_settings()``
+    only writes missing files/directories.
+    """
+    try:
+        resolve_seed_default_model_settings()
+    except Exception as e:
+        print_warning(
+            "Could not initialize default model providers and model configs automatically. "
+            f"The command will continue. Error: {e}. "
+            "You will need to configure providers and models manually with "
+            "`data-designer config providers` and `data-designer config models`."
+        )

packages/data-designer/src/data_designer/cli/utils/config_loader.py

@@ -9,7 +9,6 @@
 from urllib.parse import urlparse
 
 from data_designer.config.config_builder import DataDesignerConfigBuilder
-from data_designer.config.default_model_settings import resolve_seed_default_model_settings
 from data_designer.config.utils.io_helpers import VALID_CONFIG_FILE_EXTENSIONS, is_http_url
 
 
@@ -23,18 +22,6 @@ class ConfigLoadError(Exception):
 USER_MODULE_FUNC_NAME = "load_config_builder"
 
 
-_default_settings_initialized = False
-
-
-def _ensure_default_model_settings() -> None:
-    """Initialize default model/provider files once before loading CLI configs."""
-    global _default_settings_initialized
-    if _default_settings_initialized:
-        return
-    resolve_seed_default_model_settings()
-    _default_settings_initialized = True
-
-
 def load_config_builder(config_source: str) -> DataDesignerConfigBuilder:
     """Load a DataDesignerConfigBuilder from a file path or URL.
 
@@ -52,8 +39,6 @@ def load_config_builder(config_source: str) -> DataDesignerConfigBuilder:
     Raises:
         ConfigLoadError: If the file cannot be loaded or is invalid.
     """
-    _ensure_default_model_settings()
-
     if is_http_url(config_source):
         return _load_from_config_url(config_source)
 

packages/data-designer/tests/cli/test_main.py

@@ -0,0 +1,43 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from unittest.mock import Mock, call, patch
+
+from typer.testing import CliRunner
+
+from data_designer.cli.main import app, main
+from data_designer.config.utils.constants import DEFAULT_NUM_RECORDS
+
+runner = CliRunner()
+
+
+@patch("data_designer.cli.main.app")
+@patch("data_designer.cli.main.ensure_cli_default_model_settings")
+def test_main_bootstraps_before_running_app(mock_bootstrap: Mock, mock_app: Mock) -> None:
+    """The CLI entrypoint bootstraps defaults before invoking Typer."""
+    call_order = Mock()
+    call_order.attach_mock(mock_bootstrap, "bootstrap")
+    call_order.attach_mock(mock_app, "app")
+
+    main()
+
+    assert call_order.mock_calls == [call.bootstrap(), call.app()]
+
+
+@patch("data_designer.cli.commands.create.GenerationController")
+def test_app_dispatches_lazy_create_command(mock_controller_cls: Mock) -> None:
+    """The Typer app dispatches lazy-loaded commands through the resolved callback."""
+    mock_controller = Mock()
+    mock_controller_cls.return_value = mock_controller
+
+    result = runner.invoke(app, ["create", "config.yaml"])
+
+    assert result.exit_code == 0
+    mock_controller.run_create.assert_called_once_with(
+        config_source="config.yaml",
+        num_records=DEFAULT_NUM_RECORDS,
+        dataset_name="dataset",
+        artifact_path=None,
+    )

packages/data-designer/tests/cli/test_runtime.py

@@ -0,0 +1,41 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from unittest.mock import patch
+
+import data_designer.cli.runtime as runtime_mod
+
+
+def test_ensure_cli_default_model_settings_attempts_default_setup() -> None:
+    """CLI bootstrap delegates to default setup when the CLI starts."""
+    with (
+        patch("data_designer.cli.runtime.print_warning") as mock_print_warning,
+        patch("data_designer.cli.runtime.resolve_seed_default_model_settings") as mock_resolve,
+    ):
+        runtime_mod.ensure_cli_default_model_settings()
+
+    mock_resolve.assert_called_once_with()
+    mock_print_warning.assert_not_called()
+
+
+def test_ensure_cli_default_model_settings_warns_and_continues() -> None:
+    """CLI bootstrap prints an actionable warning when setup fails."""
+    with (
+        patch("data_designer.cli.runtime.print_warning") as mock_print_warning,
+        patch(
+            "data_designer.cli.runtime.resolve_seed_default_model_settings",
+            side_effect=RuntimeError("boom"),
+        ) as mock_resolve,
+    ):
+        runtime_mod.ensure_cli_default_model_settings()
+
+    mock_resolve.assert_called_once_with()
+    mock_print_warning.assert_called_once()
+    warning = mock_print_warning.call_args[0][0]
+    assert "Could not initialize default model providers and model configs automatically." in warning
+    assert "The command will continue." in warning
+    assert "boom" in warning
+    assert "data-designer config providers" in warning
+    assert "data-designer config models" in warning

packages/data-designer/tests/cli/utils/test_config_loader.py

@@ -1,12 +1,13 @@
 # SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
+from __future__ import annotations
+
 from pathlib import Path
 from unittest.mock import MagicMock, patch
 
 import pytest
 
-import data_designer.cli.utils.config_loader as config_loader_mod
 from data_designer.cli.utils.config_loader import (
     ConfigLoadError,
     load_config_builder,
@@ -288,13 +289,3 @@ def test_load_config_builder_empty_yaml(tmp_path: Path) -> None:
 
     with pytest.raises(ConfigLoadError, match="Failed to load config from"):
         load_config_builder(str(yaml_file))
-
-
-def test_ensure_default_model_settings_runs_once(monkeypatch: pytest.MonkeyPatch) -> None:
-    """_ensure_default_model_settings only calls resolve_seed_default_model_settings once."""
-    monkeypatch.setattr(config_loader_mod, "_default_settings_initialized", False)
-
-    with patch("data_designer.cli.utils.config_loader.resolve_seed_default_model_settings") as mock_resolve:
-        config_loader_mod._ensure_default_model_settings()
-        config_loader_mod._ensure_default_model_settings()
-        mock_resolve.assert_called_once()