refactor: enforce top-level imports for llama-stack-api

Enforce that all imports from llama-stack-api use the form:

from llama_stack_api import <symbol>

 This prevents external code from accessing internal package structure
 (e.g., llama_stack_api.agents, llama_stack_api.common.*) and establishes
 a clear public API boundary.

 Changes:
 - Export 400+ symbols from llama_stack_api/__init__.py
 - Include all API types, common utilities, and strong_typing helpers
 - Update files across src/llama_stack, docs/, tests/, scripts/
 - Convert all submodule imports to top-level imports
 - ensure docs use the proper importing structure

 Addresses PR review feedback requiring explicit __all__ definition to
 prevent "peeking inside" the API package.

Signed-off-by: Charlie Doern <cdoern@redhat.com>

tests/external/llama-stack-api-weather/src/llama_stack_api_weather/weather.py

@@ -6,9 +6,7 @@
 
 from typing import Protocol
 
-from llama_stack_api.datatypes import Api, ProviderSpec, RemoteProviderSpec
-from llama_stack_api.schema_utils import webmethod
-from llama_stack_api.version import LLAMA_STACK_API_V1
+from llama_stack_api import LLAMA_STACK_API_V1, Api, ProviderSpec, RemoteProviderSpec, webmethod
 
 
 def available_providers() -> list[ProviderSpec]:

tests/integration/batches/conftest.py

@@ -13,7 +13,7 @@ from contextlib import contextmanager
 from io import BytesIO
 
 import pytest
-from llama_stack_api.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose
 
 
 class BatchHelper:

tests/integration/files/test_files.py

@@ -9,7 +9,7 @@ from unittest.mock import patch
 
 import pytest
 import requests
-from llama_stack_api.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose
 
 from llama_stack.core.datatypes import User
 

tests/integration/inference/test_provider_data_routing.py

@@ -15,8 +15,8 @@ that enables routing based on provider_data alone.
 from unittest.mock import AsyncMock, patch
 
 import pytest
-from llama_stack_api.datatypes import Api
-from llama_stack_api.inference import (
+from llama_stack_api import (
+    Api,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionUsage,

tests/integration/post_training/test_post_training.py

@@ -9,7 +9,7 @@ import time
 import uuid
 
 import pytest
-from llama_stack_api.post_training import (
+from llama_stack_api import (
     DataConfig,
     DatasetFormat,
     DPOAlignmentConfig,

tests/integration/safety/test_llama_guard.py

@@ -12,7 +12,7 @@ import warnings
 from collections.abc import Generator
 
 import pytest
-from llama_stack_api.safety import ViolationLevel
+from llama_stack_api import ViolationLevel
 
 from llama_stack.models.llama.sku_types import CoreModelId
 

tests/integration/safety/test_safety.py

@@ -7,7 +7,7 @@ import base64
 import mimetypes
 
 import pytest
-from llama_stack_api.safety import ViolationLevel
+from llama_stack_api import ViolationLevel
 
 CODE_SCANNER_ENABLED_PROVIDERS = {"ollama", "together", "fireworks"}
 

tests/integration/safety/test_vision_safety.py

@@ -9,7 +9,7 @@ import mimetypes
 import os
 
 import pytest
-from llama_stack_api.safety import ViolationLevel
+from llama_stack_api import ViolationLevel
 
 VISION_SHIELD_ENABLED_PROVIDERS = {"together"}
 

tests/integration/tool_runtime/test_registration.py

@@ -7,7 +7,7 @@
 import re
 
 import pytest
-from llama_stack_api.common.errors import ToolGroupNotFoundError
+from llama_stack_api import ToolGroupNotFoundError
 
 from llama_stack.core.library_client import LlamaStackAsLibraryClient
 from tests.common.mcp import MCP_TOOLGROUP_ID, make_mcp_server

tests/integration/vector_io/test_openai_vector_stores.py

@@ -8,8 +8,7 @@ import time
 from io import BytesIO
 
 import pytest
-from llama_stack_api.files import ExpiresAfter
-from llama_stack_api.vector_io import Chunk
+from llama_stack_api import Chunk, ExpiresAfter
 from llama_stack_client import BadRequestError
 from openai import BadRequestError as OpenAIBadRequestError
 
@@ -646,7 +645,7 @@ def test_openai_vector_store_attach_file(
 ):
     """Test OpenAI vector store attach file."""
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     compat_client = compat_client_with_empty_stores
 
@@ -710,7 +709,7 @@ def test_openai_vector_store_attach_files_on_creation(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create some files and attach them to the vector store
     valid_file_ids = []
@@ -775,7 +774,7 @@ def test_openai_vector_store_list_files(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -867,7 +866,7 @@ def test_openai_vector_store_retrieve_file_contents(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -928,7 +927,7 @@ def test_openai_vector_store_delete_file(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -994,7 +993,7 @@ def test_openai_vector_store_delete_file_removes_from_vector_store(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -1046,7 +1045,7 @@ def test_openai_vector_store_update_file(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -1103,7 +1102,7 @@ def test_create_vector_store_files_duplicate_vector_store_name(
     This test confirms that client.vector_stores.create() creates a unique ID
     """
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
-    from llama_stack_api.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     compat_client = compat_client_with_empty_stores
 

tests/integration/vector_io/test_vector_io.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 import pytest
-from llama_stack_api.vector_io import Chunk
+from llama_stack_api import Chunk
 
 from ..conftest import vector_provider_wrapper
 

tests/unit/conversations/test_api_models.py

@@ -5,11 +5,7 @@
 # the root directory of this source tree.
 
 
-from llama_stack_api.conversations import (
-    Conversation,
-    ConversationItem,
-    ConversationItemList,
-)
+from llama_stack_api import Conversation, ConversationItem, ConversationItemList
 
 
 def test_conversation_model_defaults():

tests/unit/conversations/test_conversations.py

@@ -8,10 +8,7 @@ import tempfile
 from pathlib import Path
 
 import pytest
-from llama_stack_api.openai_responses import (
-    OpenAIResponseInputMessageContentText,
-    OpenAIResponseMessage,
-)
+from llama_stack_api import OpenAIResponseInputMessageContentText, OpenAIResponseMessage
 from openai.types.conversations.conversation import Conversation as OpenAIConversation
 from openai.types.conversations.conversation_item import ConversationItem as OpenAIConversationItem
 from pydantic import TypeAdapter

tests/unit/core/routers/test_safety_router.py

@@ -6,8 +6,7 @@
 
 from unittest.mock import AsyncMock
 
-from llama_stack_api.safety import ModerationObject, ModerationObjectResults
-from llama_stack_api.shields import ListShieldsResponse, Shield
+from llama_stack_api import ListShieldsResponse, ModerationObject, ModerationObjectResults, Shield
 
 from llama_stack.core.datatypes import SafetyConfig
 from llama_stack.core.routers.safety import SafetyRouter

tests/unit/core/routers/test_vector_io.py

@@ -7,7 +7,7 @@
 from unittest.mock import AsyncMock, Mock
 
 import pytest
-from llama_stack_api.vector_io import OpenAICreateVectorStoreRequestWithExtraBody
+from llama_stack_api import OpenAICreateVectorStoreRequestWithExtraBody
 
 from llama_stack.core.routers.vector_io import VectorIORouter
 

tests/unit/core/test_stack_validation.py

@@ -9,9 +9,7 @@
 from unittest.mock import AsyncMock
 
 import pytest
-from llama_stack_api.datatypes import Api
-from llama_stack_api.models import ListModelsResponse, Model, ModelType
-from llama_stack_api.shields import ListShieldsResponse, Shield
+from llama_stack_api import Api, ListModelsResponse, ListShieldsResponse, Model, ModelType, Shield
 
 from llama_stack.core.datatypes import QualifiedModel, SafetyConfig, StackRunConfig, StorageConfig, VectorStoresConfig
 from llama_stack.core.stack import validate_safety_config, validate_vector_stores_config

tests/unit/distribution/routers/test_routing_tables.py

@@ -9,14 +9,21 @@
 from unittest.mock import AsyncMock
 
 import pytest
-from llama_stack_api.common.content_types import URL
-from llama_stack_api.common.errors import ModelNotFoundError
-from llama_stack_api.common.type_system import NumberType
-from llama_stack_api.datasets import Dataset, DatasetPurpose, URIDataSource
-from llama_stack_api.datatypes import Api
-from llama_stack_api.models import Model, ModelType
-from llama_stack_api.shields import Shield
-from llama_stack_api.tools import ListToolDefsResponse, ToolDef, ToolGroup
+from llama_stack_api import (
+    URL,
+    Api,
+    Dataset,
+    DatasetPurpose,
+    ListToolDefsResponse,
+    Model,
+    ModelNotFoundError,
+    ModelType,
+    NumberType,
+    Shield,
+    ToolDef,
+    ToolGroup,
+    URIDataSource,
+)
 
 from llama_stack.core.datatypes import RegistryEntrySource
 from llama_stack.core.routing_tables.benchmarks import BenchmarksRoutingTable

tests/unit/distribution/test_api_recordings.py

@@ -11,7 +11,7 @@ from unittest.mock import patch
 import pytest
 
 # Import the real Pydantic response types instead of using Mocks
-from llama_stack_api.inference import (
+from llama_stack_api import (
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChoice,

tests/unit/distribution/test_distribution.py

@@ -9,7 +9,7 @@ from unittest.mock import patch
 
 import pytest
 import yaml
-from llama_stack_api.datatypes import ProviderSpec
+from llama_stack_api import ProviderSpec
 from pydantic import BaseModel, Field, ValidationError
 
 from llama_stack.core.datatypes import Api, Provider, StackRunConfig
@@ -312,7 +312,7 @@ pip_packages:
         """Test loading an external provider from a module (success path)."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import Api, ProviderSpec
+        from llama_stack_api import Api, ProviderSpec
 
         # Simulate a provider module with get_provider_spec
         fake_spec = ProviderSpec(
@@ -395,7 +395,7 @@ pip_packages:
 
     def test_external_provider_from_module_building(self, mock_providers):
         """Test loading an external provider from a module during build (building=True, partial spec)."""
-        from llama_stack_api.datatypes import Api
+        from llama_stack_api import Api
 
         from llama_stack.core.datatypes import BuildConfig, BuildProvider, DistributionSpec
 
@@ -457,7 +457,7 @@ class TestGetExternalProvidersFromModule:
         """Test provider with module containing version spec (e.g., package==1.0.0)."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         from llama_stack.core.distribution import get_external_providers_from_module
 
@@ -595,7 +595,7 @@ class TestGetExternalProvidersFromModule:
         """Test when get_provider_spec returns a list of specs."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         from llama_stack.core.distribution import get_external_providers_from_module
 
@@ -644,7 +644,7 @@ class TestGetExternalProvidersFromModule:
         """Test that list return filters specs by provider_type."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         from llama_stack.core.distribution import get_external_providers_from_module
 
@@ -693,7 +693,7 @@ class TestGetExternalProvidersFromModule:
         """Test that list return adds multiple different provider_types when config requests them."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         from llama_stack.core.distribution import get_external_providers_from_module
 
@@ -833,7 +833,7 @@ class TestGetExternalProvidersFromModule:
         """Test multiple APIs with providers."""
         from types import SimpleNamespace
 
-        from llama_stack_api.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         from llama_stack.core.distribution import get_external_providers_from_module
 

tests/unit/files/test_files.py

@@ -6,9 +6,7 @@
 
 
 import pytest
-from llama_stack_api.common.errors import ResourceNotFoundError
-from llama_stack_api.common.responses import Order
-from llama_stack_api.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose, Order, ResourceNotFoundError
 
 from llama_stack.core.access_control.access_control import default_policy
 from llama_stack.core.storage.datatypes import SqliteSqlStoreConfig, SqlStoreReference

tests/unit/providers/batches/test_reference.py

@@ -58,8 +58,7 @@ import json
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
-from llama_stack_api.batches import BatchObject
-from llama_stack_api.common.errors import ConflictError, ResourceNotFoundError
+from llama_stack_api import BatchObject, ConflictError, ResourceNotFoundError
 
 
 class TestReferenceBatchesImpl:

tests/unit/providers/batches/test_reference_idempotency.py

@@ -43,7 +43,7 @@ Key Behaviors Tested:
 import asyncio
 
 import pytest
-from llama_stack_api.common.errors import ConflictError
+from llama_stack_api import ConflictError
 
 
 class TestReferenceBatchesIdempotency:

tests/unit/providers/files/test_s3_files.py

@@ -8,8 +8,7 @@ from unittest.mock import patch
 
 import pytest
 from botocore.exceptions import ClientError
-from llama_stack_api.common.errors import ResourceNotFoundError
-from llama_stack_api.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose, ResourceNotFoundError
 
 
 class TestS3FilesImpl:
@@ -227,7 +226,7 @@ class TestS3FilesImpl:
 
             mock_now.return_value = 0
 
-            from llama_stack_api.files import ExpiresAfter
+            from llama_stack_api import ExpiresAfter
 
             sample_text_file.filename = "test_expired_file"
             uploaded = await s3_provider.openai_upload_file(
@@ -259,7 +258,7 @@ class TestS3FilesImpl:
 
     async def test_unsupported_expires_after_anchor(self, s3_provider, sample_text_file):
         """Unsupported anchor value should raise ValueError."""
-        from llama_stack_api.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         sample_text_file.filename = "test_unsupported_expires_after_anchor"
 
@@ -272,7 +271,7 @@ class TestS3FilesImpl:
 
     async def test_nonint_expires_after_seconds(self, s3_provider, sample_text_file):
         """Non-integer seconds in expires_after should raise ValueError."""
-        from llama_stack_api.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         sample_text_file.filename = "test_nonint_expires_after_seconds"
 
@@ -285,7 +284,7 @@ class TestS3FilesImpl:
 
     async def test_expires_after_seconds_out_of_bounds(self, s3_provider, sample_text_file):
         """Seconds outside allowed range should raise ValueError."""
-        from llama_stack_api.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         with pytest.raises(ValueError, match="greater than or equal to 3600"):
             await s3_provider.openai_upload_file(

tests/unit/providers/files/test_s3_files_auth.py

@@ -7,8 +7,7 @@
 from unittest.mock import patch
 
 import pytest
-from llama_stack_api.common.errors import ResourceNotFoundError
-from llama_stack_api.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose, ResourceNotFoundError
 
 from llama_stack.core.datatypes import User
 from llama_stack.providers.remote.files.s3.files import S3FilesImpl

tests/unit/providers/inference/test_bedrock_adapter.py

@@ -8,7 +8,7 @@ from types import SimpleNamespace
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
-from llama_stack_api.inference import OpenAIChatCompletionRequestWithExtraBody
+from llama_stack_api import OpenAIChatCompletionRequestWithExtraBody
 from openai import AuthenticationError
 
 from llama_stack.providers.remote.inference.bedrock.bedrock import BedrockInferenceAdapter

tests/unit/providers/inference/test_remote_vllm.py

@@ -9,8 +9,9 @@ import time
 from unittest.mock import AsyncMock, MagicMock, PropertyMock, patch
 
 import pytest
-from llama_stack_api.datatypes import HealthStatus
-from llama_stack_api.inference import (
+from llama_stack_api import (
+    HealthStatus,
+    Model,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionRequestWithExtraBody,
@@ -20,7 +21,6 @@ from llama_stack_api.inference import (
     OpenAICompletionRequestWithExtraBody,
     ToolChoice,
 )
-from llama_stack_api.models import Model
 
 from llama_stack.core.routers.inference import InferenceRouter
 from llama_stack.core.routing_tables.models import ModelsRoutingTable

tests/unit/providers/inline/agents/meta_reference/responses/test_streaming.py

@@ -7,7 +7,7 @@
 from unittest.mock import AsyncMock
 
 import pytest
-from llama_stack_api.tools import ToolDef
+from llama_stack_api import ToolDef
 
 from llama_stack.providers.inline.agents.meta_reference.responses.streaming import (
     convert_tooldef_to_chat_tool,

tests/unit/providers/nvidia/test_datastore.py

@@ -8,8 +8,7 @@ import os
 from unittest.mock import patch
 
 import pytest
-from llama_stack_api.datasets import Dataset, DatasetPurpose, URIDataSource
-from llama_stack_api.resource import ResourceType
+from llama_stack_api import Dataset, DatasetPurpose, ResourceType, URIDataSource
 
 from llama_stack.providers.remote.datasetio.nvidia.config import NvidiaDatasetIOConfig
 from llama_stack.providers.remote.datasetio.nvidia.datasetio import NvidiaDatasetIOAdapter

tests/unit/providers/nvidia/test_eval.py

@@ -8,11 +8,17 @@ import os
 from unittest.mock import MagicMock, patch
 
 import pytest
-from llama_stack_api.benchmarks import Benchmark
-from llama_stack_api.common.job_types import Job, JobStatus
-from llama_stack_api.eval import BenchmarkConfig, EvaluateResponse, ModelCandidate, SamplingParams
-from llama_stack_api.inference import TopPSamplingStrategy
-from llama_stack_api.resource import ResourceType
+from llama_stack_api import (
+    Benchmark,
+    BenchmarkConfig,
+    EvaluateResponse,
+    Job,
+    JobStatus,
+    ModelCandidate,
+    ResourceType,
+    SamplingParams,
+    TopPSamplingStrategy,
+)
 
 from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.remote.eval.nvidia.config import NVIDIAEvalConfig

tests/unit/providers/nvidia/test_parameters.py

@@ -9,7 +9,7 @@ import warnings
 from unittest.mock import patch
 
 import pytest
-from llama_stack_api.post_training import (
+from llama_stack_api import (
     DataConfig,
     DatasetFormat,
     EfficiencyConfig,

tests/unit/providers/nvidia/test_rerank_inference.py

@@ -8,7 +8,7 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import aiohttp
 import pytest
-from llama_stack_api.models import ModelType
+from llama_stack_api import ModelType
 
 from llama_stack.providers.remote.inference.nvidia.config import NVIDIAConfig
 from llama_stack.providers.remote.inference.nvidia.nvidia import NVIDIAInferenceAdapter

tests/unit/providers/nvidia/test_safety.py

@@ -9,13 +9,14 @@ from typing import Any
 from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
-from llama_stack_api.inference import (
+from llama_stack_api import (
     OpenAIAssistantMessageParam,
     OpenAIUserMessageParam,
+    ResourceType,
+    RunShieldResponse,
+    Shield,
+    ViolationLevel,
 )
-from llama_stack_api.resource import ResourceType
-from llama_stack_api.safety import RunShieldResponse, ViolationLevel
-from llama_stack_api.shields import Shield
 
 from llama_stack.providers.remote.safety.nvidia.config import NVIDIASafetyConfig
 from llama_stack.providers.remote.safety.nvidia.nvidia import NVIDIASafetyAdapter

tests/unit/providers/nvidia/test_supervised_fine_tuning.py

@@ -9,7 +9,7 @@ import warnings
 from unittest.mock import patch
 
 import pytest
-from llama_stack_api.post_training import (
+from llama_stack_api import (
     DataConfig,
     DatasetFormat,
     LoraFinetuningConfig,

tests/unit/providers/test_bedrock.py

@@ -7,7 +7,7 @@
 from types import SimpleNamespace
 from unittest.mock import AsyncMock, PropertyMock, patch
 
-from llama_stack_api.inference import OpenAIChatCompletionRequestWithExtraBody
+from llama_stack_api import OpenAIChatCompletionRequestWithExtraBody
 
 from llama_stack.providers.remote.inference.bedrock.bedrock import BedrockInferenceAdapter
 from llama_stack.providers.remote.inference.bedrock.config import BedrockConfig

tests/unit/providers/utils/inference/test_openai_mixin.py

@@ -10,8 +10,7 @@ from typing import Any
 from unittest.mock import AsyncMock, MagicMock, Mock, PropertyMock, patch
 
 import pytest
-from llama_stack_api.inference import Model, OpenAIChatCompletionRequestWithExtraBody, OpenAIUserMessageParam
-from llama_stack_api.models import ModelType
+from llama_stack_api import Model, ModelType, OpenAIChatCompletionRequestWithExtraBody, OpenAIUserMessageParam
 from pydantic import BaseModel, Field
 
 from llama_stack.core.request_headers import request_provider_data_context

tests/unit/providers/utils/inference/test_prompt_adapter.py

@@ -4,10 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.inference import (
-    OpenAIAssistantMessageParam,
-    OpenAIUserMessageParam,
-)
+from llama_stack_api import OpenAIAssistantMessageParam, OpenAIUserMessageParam
 
 from llama_stack.models.llama.datatypes import RawTextItem
 from llama_stack.providers.utils.inference.prompt_adapter import (

tests/unit/providers/utils/memory/test_vector_store.py

@@ -7,8 +7,7 @@
 from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
-from llama_stack_api.common.content_types import URL, TextContentItem
-from llama_stack_api.rag_tool import RAGDocument
+from llama_stack_api import URL, RAGDocument, TextContentItem
 
 from llama_stack.providers.utils.memory.vector_store import content_from_data_and_mime_type, content_from_doc
 

tests/unit/providers/utils/test_model_registry.py

@@ -34,7 +34,7 @@
 #
 
 import pytest
-from llama_stack_api.models import Model
+from llama_stack_api import Model
 
 from llama_stack.providers.utils.inference.model_registry import ModelRegistryHelper, ProviderModelEntry
 

tests/unit/providers/vector_io/conftest.py

@@ -9,8 +9,7 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import numpy as np
 import pytest
-from llama_stack_api.vector_io import Chunk, ChunkMetadata, QueryChunksResponse
-from llama_stack_api.vector_stores import VectorStore
+from llama_stack_api import Chunk, ChunkMetadata, QueryChunksResponse, VectorStore
 
 from llama_stack.core.storage.datatypes import KVStoreReference, SqliteKVStoreConfig
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig

tests/unit/providers/vector_io/test_faiss.py

@@ -9,10 +9,7 @@ from unittest.mock import MagicMock, patch
 
 import numpy as np
 import pytest
-from llama_stack_api.datatypes import HealthStatus
-from llama_stack_api.files import Files
-from llama_stack_api.vector_io import Chunk, QueryChunksResponse
-from llama_stack_api.vector_stores import VectorStore
+from llama_stack_api import Chunk, Files, HealthStatus, QueryChunksResponse, VectorStore
 
 from llama_stack.providers.inline.vector_io.faiss.config import FaissVectorIOConfig
 from llama_stack.providers.inline.vector_io.faiss.faiss import (

tests/unit/providers/vector_io/test_sqlite_vec.py

@@ -8,7 +8,7 @@ import asyncio
 
 import numpy as np
 import pytest
-from llama_stack_api.vector_io import Chunk, QueryChunksResponse
+from llama_stack_api import Chunk, QueryChunksResponse
 
 from llama_stack.providers.inline.vector_io.sqlite_vec.sqlite_vec import (
     SQLiteVecIndex,

tests/unit/providers/vector_io/test_vector_io_openai_vector_stores.py

@@ -10,16 +10,16 @@ from unittest.mock import AsyncMock, patch
 
 import numpy as np
 import pytest
-from llama_stack_api.common.errors import VectorStoreNotFoundError
-from llama_stack_api.vector_io import (
+from llama_stack_api import (
     Chunk,
     OpenAICreateVectorStoreFileBatchRequestWithExtraBody,
     OpenAICreateVectorStoreRequestWithExtraBody,
     QueryChunksResponse,
+    VectorStore,
     VectorStoreChunkingStrategyAuto,
     VectorStoreFileObject,
+    VectorStoreNotFoundError,
 )
-from llama_stack_api.vector_stores import VectorStore
 
 from llama_stack.providers.inline.vector_io.sqlite_vec.sqlite_vec import VECTOR_DBS_PREFIX
 
@@ -222,7 +222,7 @@ async def test_insert_chunks_missing_db_raises(vector_io_adapter):
 
 async def test_insert_chunks_with_missing_document_id(vector_io_adapter):
     """Ensure no KeyError when document_id is missing or in different places."""
-    from llama_stack_api.vector_io import Chunk, ChunkMetadata
+    from llama_stack_api import Chunk, ChunkMetadata
 
     fake_index = AsyncMock()
     vector_io_adapter.cache["db1"] = fake_index
@@ -255,7 +255,7 @@ async def test_insert_chunks_with_missing_document_id(vector_io_adapter):
 
 async def test_document_id_with_invalid_type_raises_error():
     """Ensure TypeError is raised when document_id is not a string."""
-    from llama_stack_api.vector_io import Chunk
+    from llama_stack_api import Chunk
 
     # Integer document_id should raise TypeError
     from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id

tests/unit/providers/vector_io/test_vector_utils.py

@@ -4,7 +4,7 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack_api.vector_io import Chunk, ChunkMetadata
+from llama_stack_api import Chunk, ChunkMetadata
 
 from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id
 

tests/unit/rag/test_rag_query.py

@@ -7,12 +7,7 @@
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
-from llama_stack_api.rag_tool import RAGQueryConfig
-from llama_stack_api.vector_io import (
-    Chunk,
-    ChunkMetadata,
-    QueryChunksResponse,
-)
+from llama_stack_api import Chunk, ChunkMetadata, QueryChunksResponse, RAGQueryConfig
 
 from llama_stack.providers.inline.tool_runtime.rag.memory import MemoryToolRuntimeImpl
 

tests/unit/rag/test_vector_store.py

@@ -12,12 +12,7 @@ from unittest.mock import AsyncMock, MagicMock
 
 import numpy as np
 import pytest
-from llama_stack_api.inference import (
-    OpenAIEmbeddingData,
-    OpenAIEmbeddingsRequestWithExtraBody,
-)
-from llama_stack_api.rag_tool import RAGDocument
-from llama_stack_api.vector_io import Chunk
+from llama_stack_api import Chunk, OpenAIEmbeddingData, OpenAIEmbeddingsRequestWithExtraBody, RAGDocument
 
 from llama_stack.providers.utils.memory.vector_store import (
     URL,

tests/unit/registry/test_registry.py

@@ -6,8 +6,7 @@
 
 
 import pytest
-from llama_stack_api.inference import Model
-from llama_stack_api.vector_stores import VectorStore
+from llama_stack_api import Model, VectorStore
 
 from llama_stack.core.datatypes import VectorStoreWithOwner
 from llama_stack.core.storage.datatypes import KVStoreReference, SqliteKVStoreConfig
@@ -304,7 +303,7 @@ async def test_double_registration_different_objects(disk_dist_registry):
 
 async def test_double_registration_with_cache(cached_disk_dist_registry):
     """Test double registration behavior with caching enabled."""
-    from llama_stack_api.models import ModelType
+    from llama_stack_api import ModelType
 
     from llama_stack.core.datatypes import ModelWithOwner
 

tests/unit/registry/test_registry_acl.py

@@ -5,7 +5,7 @@
 # the root directory of this source tree.
 
 
-from llama_stack_api.models import ModelType
+from llama_stack_api import ModelType
 
 from llama_stack.core.datatypes import ModelWithOwner, User
 from llama_stack.core.store.registry import CachedDiskDistributionRegistry

tests/unit/server/test_access_control.py

@@ -8,8 +8,7 @@ from unittest.mock import MagicMock, Mock, patch
 
 import pytest
 import yaml
-from llama_stack_api.datatypes import Api
-from llama_stack_api.models import ModelType
+from llama_stack_api import Api, ModelType
 from pydantic import TypeAdapter, ValidationError
 
 from llama_stack.core.access_control.access_control import AccessDeniedError, is_action_allowed

tests/unit/server/test_auth.py

@@ -144,7 +144,7 @@ def middleware_with_mocks(mock_auth_endpoint):
     middleware = AuthenticationMiddleware(mock_app, auth_config, {})
 
     # Mock the route_impls to simulate finding routes with required scopes
-    from llama_stack_api.schema_utils import WebMethod
+    from llama_stack_api import WebMethod
 
     routes = {
         ("POST", "/test/scoped"): WebMethod(route="/test/scoped", method="POST", required_scope="test.read"),

tests/unit/server/test_resolver.py

@@ -9,8 +9,7 @@ import sys
 from typing import Any, Protocol
 from unittest.mock import AsyncMock, MagicMock
 
-from llama_stack_api.datatypes import InlineProviderSpec, ProviderSpec
-from llama_stack_api.inference import Inference
+from llama_stack_api import Inference, InlineProviderSpec, ProviderSpec
 from pydantic import BaseModel, Field
 
 from llama_stack.core.datatypes import Api, Provider, StackRunConfig

tests/unit/server/test_sse.py

@@ -9,7 +9,7 @@ import logging  # allow-direct-logging
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
-from llama_stack_api.common.responses import PaginatedResponse
+from llama_stack_api import PaginatedResponse
 
 from llama_stack.core.server.server import create_dynamic_typed_route, create_sse_event, sse_generator
 

tests/unit/tools/test_tools_json_schema.py

@@ -9,7 +9,7 @@ Unit tests for JSON Schema-based tool definitions.
 Tests the new input_schema and output_schema fields.
 """
 
-from llama_stack_api.tools import ToolDef
+from llama_stack_api import ToolDef
 from pydantic import ValidationError
 
 from llama_stack.models.llama.datatypes import BuiltinTool, ToolDefinition

tests/unit/utils/inference/test_inference_store.py

@@ -7,7 +7,7 @@
 import time
 
 import pytest
-from llama_stack_api.inference import (
+from llama_stack_api import (
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChoice,

tests/unit/utils/responses/test_responses_store.py

@@ -9,12 +9,7 @@ from tempfile import TemporaryDirectory
 from uuid import uuid4
 
 import pytest
-from llama_stack_api.agents import Order
-from llama_stack_api.inference import OpenAIMessageParam, OpenAIUserMessageParam
-from llama_stack_api.openai_responses import (
-    OpenAIResponseInput,
-    OpenAIResponseObject,
-)
+from llama_stack_api import OpenAIMessageParam, OpenAIResponseInput, OpenAIResponseObject, OpenAIUserMessageParam, Order
 
 from llama_stack.core.storage.datatypes import ResponsesStoreReference, SqliteSqlStoreConfig
 from llama_stack.providers.utils.responses.responses_store import ResponsesStore
@@ -46,7 +41,7 @@ def create_test_response_object(
 
 def create_test_response_input(content: str, input_id: str) -> OpenAIResponseInput:
     """Helper to create a test response input."""
-    from llama_stack_api.openai_responses import OpenAIResponseMessage
+    from llama_stack_api import OpenAIResponseMessage
 
     return OpenAIResponseMessage(
         id=input_id,