letta-server/tests/managers/test_provider_manager.py

"""Tests for ProviderManager encryption/decryption logic."""

import os

import pytest

from letta.orm.provider import Provider as ProviderModel
from letta.schemas.enums import ProviderCategory, ProviderType
from letta.schemas.providers import Provider, ProviderCreate, ProviderUpdate
from letta.schemas.secret import Secret
from letta.server.db import db_registry
from letta.services.organization_manager import OrganizationManager
from letta.services.provider_manager import ProviderManager
from letta.services.user_manager import UserManager
from letta.settings import settings


@pytest.fixture
async def default_organization():
    """Fixture to create and return the default organization."""
    manager = OrganizationManager()
    org = await manager.create_default_organization_async()
    yield org


@pytest.fixture
async def default_user(default_organization):
    """Fixture to create and return the default user within the default organization."""
    manager = UserManager()
    user = await manager.create_default_actor_async(org_id=default_organization.id)
    yield user


@pytest.fixture
async def provider_manager():
    """Fixture to create and return a ProviderManager instance."""
    return ProviderManager()


@pytest.fixture
def encryption_key():
    """Fixture to ensure encryption key is set for tests."""
    original_key = settings.encryption_key
    # Set a test encryption key if not already set
    if not settings.encryption_key:
        settings.encryption_key = "test-encryption-key-32-bytes!!"
    yield settings.encryption_key
    # Restore original
    settings.encryption_key = original_key


# ======================================================================================================================
# Provider Encryption Tests
# ======================================================================================================================


@pytest.mark.asyncio
async def test_provider_create_encrypts_api_key(provider_manager, default_user, encryption_key):
    """Test that creating a provider encrypts the api_key and stores it in api_key_enc."""
    # Create a provider with plaintext api_key
    provider_create = ProviderCreate(
        name="test-openai-provider",
        provider_type=ProviderType.openai,
        api_key="sk-test-plaintext-api-key-12345",
        base_url="https://api.openai.com/v1",
    )

    # Create provider through manager
    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)

    # Verify provider was created
    assert created_provider is not None
    assert created_provider.name == "test-openai-provider"
    assert created_provider.provider_type == ProviderType.openai

    # Verify encrypted api_key can be decrypted
    assert created_provider.api_key_enc.get_plaintext() == "sk-test-plaintext-api-key-12345"

    # Read directly from database to verify encryption
    async with db_registry.async_session() as session:
        provider_orm = await ProviderModel.read_async(
            db_session=session,
            identifier=created_provider.id,
            actor=default_user,
        )

        # Verify encrypted column is populated and decrypts correctly
        assert provider_orm.api_key_enc is not None
        decrypted = Secret.from_encrypted(provider_orm.api_key_enc).get_plaintext()
        assert decrypted == "sk-test-plaintext-api-key-12345"


@pytest.mark.asyncio
async def test_provider_read_decrypts_api_key(provider_manager, default_user, encryption_key):
    """Test that reading a provider decrypts the api_key from api_key_enc."""
    # Create a provider
    provider_create = ProviderCreate(
        name="test-anthropic-provider",
        provider_type=ProviderType.anthropic,
        api_key="sk-ant-test-key-67890",
    )

    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)
    provider_id = created_provider.id

    # Read the provider back
    retrieved_provider = await provider_manager.get_provider_async(provider_id, actor=default_user)

    # Verify the api_key is decrypted correctly via api_key_enc
    decrypted_key = retrieved_provider.api_key_enc.get_plaintext()
    assert decrypted_key == "sk-ant-test-key-67890"


@pytest.mark.asyncio
async def test_provider_update_encrypts_new_api_key(provider_manager, default_user, encryption_key):
    """Test that updating a provider's api_key encrypts the new value."""
    # Create initial provider
    provider_create = ProviderCreate(
        name="test-groq-provider",
        provider_type=ProviderType.groq,
        api_key="gsk-initial-key-123",
    )

    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)
    provider_id = created_provider.id

    # Update the api_key
    provider_update = ProviderUpdate(
        api_key="gsk-updated-key-456",
    )

    updated_provider = await provider_manager.update_provider_async(provider_id, provider_update, actor=default_user)

    # Verify the updated key is accessible via the encrypted field
    assert updated_provider.api_key_enc.get_plaintext() == "gsk-updated-key-456"

    # Read from DB to verify new encrypted value
    async with db_registry.async_session() as session:
        provider_orm = await ProviderModel.read_async(
            db_session=session,
            identifier=provider_id,
            actor=default_user,
        )

        assert provider_orm.api_key_enc is not None
        decrypted = Secret.from_encrypted(provider_orm.api_key_enc).get_plaintext()
        assert decrypted == "gsk-updated-key-456"


@pytest.mark.asyncio
async def test_bedrock_credentials_encryption(provider_manager, default_user, encryption_key):
    """Test that Bedrock provider encrypts both access_key and api_key (secret_key)."""
    # Create Bedrock provider with both keys
    provider_create = ProviderCreate(
        name="test-bedrock-provider",
        provider_type=ProviderType.bedrock,
        api_key="secret-access-key-xyz",  # This is the secret key
        access_key="access-key-id-abc",  # This is the access key ID
        region="us-east-1",
    )

    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)

    # Verify both keys are accessible via encrypted fields
    assert created_provider.api_key_enc.get_plaintext() == "secret-access-key-xyz"
    assert created_provider.access_key_enc.get_plaintext() == "access-key-id-abc"

    # Read from DB to verify both are encrypted
    async with db_registry.async_session() as session:
        provider_orm = await ProviderModel.read_async(
            db_session=session,
            identifier=created_provider.id,
            actor=default_user,
        )

        # Verify both encrypted columns are populated
        assert provider_orm.api_key_enc is not None
        assert provider_orm.access_key_enc is not None

        # Verify encrypted values are different from plaintext
        assert Secret.from_encrypted(provider_orm.api_key_enc).get_plaintext() == "secret-access-key-xyz"
        assert Secret.from_encrypted(provider_orm.access_key_enc).get_plaintext() == "access-key-id-abc"

    # Test the manager method for getting Bedrock credentials
    access_key, secret_key, region = await provider_manager.get_bedrock_credentials_async("test-bedrock-provider", actor=default_user)

    assert access_key == "access-key-id-abc"
    assert secret_key == "secret-access-key-xyz"
    assert region == "us-east-1"


@pytest.mark.asyncio
async def test_provider_secret_not_exposed_in_logs(provider_manager, default_user, encryption_key):
    """Test that Secret objects don't expose plaintext in string representations."""
    # Create a provider
    provider_create = ProviderCreate(
        name="test-secret-provider",
        provider_type=ProviderType.openai,
        api_key="sk-very-secret-key-do-not-log",
    )

    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)

    # Get the Secret object
    api_key_secret = created_provider.api_key_enc

    # Verify string representation doesn't expose the key
    secret_str = str(api_key_secret)
    secret_repr = repr(api_key_secret)

    assert "sk-very-secret-key-do-not-log" not in secret_str
    assert "sk-very-secret-key-do-not-log" not in secret_repr
    assert "****" in secret_str or "Secret" in secret_str
    assert "****" in secret_repr or "Secret" in secret_repr


@pytest.mark.asyncio
async def test_provider_pydantic_to_orm_serialization(provider_manager, default_user, encryption_key):
    """Test the full Pydantic → ORM → Pydantic round-trip maintains data integrity."""
    # Create a provider through the normal flow
    provider_create = ProviderCreate(
        name="test-roundtrip-provider",
        provider_type=ProviderType.openai,
        api_key="sk-roundtrip-test-key-999",
        base_url="https://api.openai.com/v1",
    )

    # Step 1: Create provider (Pydantic → ORM)
    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)
    original_api_key = created_provider.api_key_enc.get_plaintext()

    # Step 2: Read provider back (ORM → Pydantic)
    retrieved_provider = await provider_manager.get_provider_async(created_provider.id, actor=default_user)

    # Verify data integrity
    assert retrieved_provider.api_key_enc.get_plaintext() == original_api_key
    assert retrieved_provider.name == "test-roundtrip-provider"
    assert retrieved_provider.provider_type == ProviderType.openai
    assert retrieved_provider.base_url == "https://api.openai.com/v1"

    # Verify Secret object works correctly
    api_key_secret = retrieved_provider.api_key_enc
    assert api_key_secret.get_plaintext() == original_api_key

    # Step 3: Convert to ORM again (should preserve encrypted field)
    orm_data = retrieved_provider.model_dump(to_orm=True)

    # Verify encrypted field is in the ORM data
    assert "api_key_enc" in orm_data
    assert orm_data["api_key_enc"] is not None
    assert Secret.from_encrypted(orm_data["api_key_enc"]).get_plaintext() == original_api_key


@pytest.mark.asyncio
async def test_provider_with_none_api_key(provider_manager, default_user, encryption_key):
    """Test that providers can be created with None api_key (some providers may not need it)."""
    # Create a provider without an api_key
    provider_create = ProviderCreate(
        name="test-no-key-provider",
        provider_type=ProviderType.ollama,
        api_key="",  # Empty string
        base_url="http://localhost:11434",
    )

    created_provider = await provider_manager.create_provider_async(provider_create, actor=default_user)

    # Verify provider was created
    assert created_provider is not None
    assert created_provider.name == "test-no-key-provider"

    # Read from DB
    async with db_registry.async_session() as session:
        provider_orm = await ProviderModel.read_async(
            db_session=session,
            identifier=created_provider.id,
            actor=default_user,
        )

        # api_key_enc should handle empty string appropriately
        assert provider_orm.api_key_enc is not None
        assert Secret.from_encrypted(provider_orm.api_key_enc).get_plaintext() == ""


@pytest.mark.asyncio
async def test_list_providers_decrypts_all(provider_manager, default_user, encryption_key):
    """Test that listing multiple providers decrypts all their api_keys correctly."""
    # Create multiple providers
    providers_to_create = [
        ProviderCreate(name=f"test-provider-{i}", provider_type=ProviderType.openai, api_key=f"sk-key-{i}") for i in range(3)
    ]

    created_ids = []
    for provider_create in providers_to_create:
        provider = await provider_manager.create_provider_async(provider_create, actor=default_user)
        created_ids.append(provider.id)

    # List all providers
    all_providers = await provider_manager.list_providers_async(actor=default_user)

    # Filter to our test providers
    test_providers = [p for p in all_providers if p.id in created_ids]

    # Verify all are decrypted correctly
    assert len(test_providers) == 3
    for i, provider in enumerate(sorted(test_providers, key=lambda p: p.name)):
        secret = provider.api_key_enc
        assert secret.get_plaintext() == f"sk-key-{i}"


# ======================================================================================================================
# Handle to Config Conversion Tests
# ======================================================================================================================


@pytest.mark.asyncio
async def test_handle_to_llm_config_conversion(provider_manager, default_user):
    """Test that handle to LLMConfig conversion works correctly with database lookup."""
    from letta.orm.errors import NoResultFound
    from letta.schemas.embedding_config import EmbeddingConfig
    from letta.schemas.llm_config import LLMConfig

    # Create a test provider
    provider_create = ProviderCreate(
        name="test-handle-provider", provider_type=ProviderType.openai, api_key="sk-test-handle-key", base_url="https://api.openai.com/v1"
    )
    provider = await provider_manager.create_provider_async(provider_create, actor=default_user)

    # Sync some test models
    llm_models = [
        LLMConfig(
            model="gpt-4",
            model_endpoint_type="openai",
            model_endpoint="https://api.openai.com/v1",
            context_window=8192,
            handle="test-handle-provider/gpt-4",
            provider_name=provider.name,
            provider_category=ProviderCategory.base,
        ),
        LLMConfig(
            model="gpt-3.5-turbo",
            model_endpoint_type="openai",
            model_endpoint="https://api.openai.com/v1",
            context_window=4096,
            handle="test-handle-provider/gpt-3.5-turbo",
            provider_name=provider.name,
            provider_category=ProviderCategory.base,
        ),
    ]

    embedding_models = [
        EmbeddingConfig(
            embedding_model="text-embedding-ada-002",
            embedding_endpoint_type="openai",
            embedding_endpoint="https://api.openai.com/v1",
            embedding_dim=1536,
            embedding_chunk_size=300,
            handle="test-handle-provider/text-embedding-ada-002",
        )
    ]

    await provider_manager.sync_provider_models_async(
        provider=provider, llm_models=llm_models, embedding_models=embedding_models, organization_id=default_user.organization_id
    )

    # Test LLM config from handle
    llm_config = await provider_manager.get_llm_config_from_handle(handle="test-handle-provider/gpt-4", actor=default_user)

    # Verify the returned config
    assert llm_config.model == "gpt-4"
    assert llm_config.handle == "test-handle-provider/gpt-4"
    assert llm_config.context_window == 8192
    assert llm_config.model_endpoint == "https://api.openai.com/v1"
    assert llm_config.provider_name == "test-handle-provider"

    # Test embedding config from handle
    embedding_config = await provider_manager.get_embedding_config_from_handle(
        handle="test-handle-provider/text-embedding-ada-002", actor=default_user
    )

    # Verify the returned config
    assert embedding_config.embedding_model == "text-embedding-ada-002"
    assert embedding_config.handle == "test-handle-provider/text-embedding-ada-002"
    assert embedding_config.embedding_dim == 1536
    assert embedding_config.embedding_chunk_size == 300
    assert embedding_config.embedding_endpoint == "https://api.openai.com/v1"

    # Test context window limit override would be done at server level
    # The provider_manager method doesn't support context_window_limit directly

    # Test error handling for non-existent handle
    with pytest.raises(NoResultFound):
        await provider_manager.get_llm_config_from_handle(handle="nonexistent/model", actor=default_user)


@pytest.mark.asyncio
async def test_byok_provider_auto_syncs_models(provider_manager, default_user, monkeypatch):
    """Test that creating a BYOK provider attempts to sync its models."""
    from letta.schemas.embedding_config import EmbeddingConfig
    from letta.schemas.llm_config import LLMConfig

    # Mock the list_llm_models_async method
    async def mock_list_llm():
        return [
            LLMConfig(
                model="gpt-4o",
                model_endpoint_type="openai",
                model_endpoint="https://api.openai.com/v1",
                context_window=128000,
                handle="openai/gpt-4o",
                provider_name="openai",
                provider_category=ProviderCategory.base,
            ),
            LLMConfig(
                model="gpt-4",
                model_endpoint_type="openai",
                model_endpoint="https://api.openai.com/v1",
                context_window=8192,
                handle="openai/gpt-4",
                provider_name="openai",
                provider_category=ProviderCategory.base,
            ),
        ]

    # Mock the list_embedding_models_async method
    async def mock_list_embedding():
        return [
            EmbeddingConfig(
                embedding_model="text-embedding-ada-002",
                embedding_endpoint_type="openai",
                embedding_endpoint="https://api.openai.com/v1",
                embedding_dim=1536,
                embedding_chunk_size=300,
                handle="openai/text-embedding-ada-002",
            )
        ]

    # Mock the _sync_default_models_for_provider method directly
    async def mock_sync(provider, actor):
        # Get mock models and update them for this provider
        llm_models = await mock_list_llm()
        embedding_models = await mock_list_embedding()

        # Update models to match the BYOK provider
        for model in llm_models:
            model.provider_name = provider.name
            model.handle = f"{provider.name}/{model.model}"
            model.provider_category = provider.provider_category

        for model in embedding_models:
            model.handle = f"{provider.name}/{model.embedding_model}"

        # Call sync_provider_models_async with mock data
        await provider_manager.sync_provider_models_async(
            provider=provider, llm_models=llm_models, embedding_models=embedding_models, organization_id=actor.organization_id
        )

    monkeypatch.setattr(provider_manager, "_sync_default_models_for_provider", mock_sync)

    # Create a BYOK OpenAI provider (simulates UI "Add API Key" flow)
    provider_create = ProviderCreate(name="my-openai-key", provider_type=ProviderType.openai, api_key="sk-my-personal-key-123")

    # Create the BYOK provider (is_byok=True is the default)
    provider = await provider_manager.create_provider_async(provider_create, actor=default_user, is_byok=True)

    # Verify provider was created
    assert provider.name == "my-openai-key"
    assert provider.provider_type == ProviderType.openai

    # List models for this provider - they should have been auto-synced
    models = await provider_manager.list_models_async(actor=default_user, provider_id=provider.id)

    # Should have both LLM and embedding models
    llm_models = [m for m in models if m.model_type == "llm"]
    embedding_models = [m for m in models if m.model_type == "embedding"]

    assert len(llm_models) > 0, "No LLM models were synced"
    assert len(embedding_models) > 0, "No embedding models were synced"

    # Verify handles are correctly formatted with BYOK provider name
    for model in models:
        assert model.handle.startswith(f"{provider.name}/")

    # Test that we can get LLM config from handle
    llm_config = await provider_manager.get_llm_config_from_handle(handle="my-openai-key/gpt-4o", actor=default_user)
    assert llm_config.model == "gpt-4o"
    assert llm_config.provider_name == "my-openai-key"


# ======================================================================================================================
# Server Startup Provider Sync Tests
# ======================================================================================================================


@pytest.mark.asyncio
async def test_server_startup_syncs_base_providers(default_user, default_organization, monkeypatch):
    """Test that server startup properly syncs base provider models from environment.

    This test simulates the server startup process and verifies that:
    1. Base providers from environment variables are synced to database
    2. Provider models are fetched from mocked API endpoints
    3. Models are properly persisted to the database with correct metadata
    4. Models can be retrieved using handles
    """
    from unittest.mock import AsyncMock

    from letta.schemas.embedding_config import EmbeddingConfig
    from letta.schemas.llm_config import LLMConfig
    from letta.schemas.providers import AnthropicProvider, OpenAIProvider
    from letta.server.server import SyncServer

    # Mock OpenAI API responses
    mock_openai_models = {
        "data": [
            {
                "id": "gpt-4",
                "object": "model",
                "created": 1687882411,
                "owned_by": "openai",
                "max_model_len": 8192,
            },
            {
                "id": "gpt-4-turbo",
                "object": "model",
                "created": 1712361441,
                "owned_by": "system",
                "max_model_len": 128000,
            },
            {
                "id": "text-embedding-ada-002",
                "object": "model",
                "created": 1671217299,
                "owned_by": "openai-internal",
            },
            {
                "id": "gpt-4-vision",  # Should be filtered out by OpenAI provider logic (has disallowed keyword)
                "object": "model",
                "created": 1698959748,
                "owned_by": "system",
                "max_model_len": 8192,
            },
        ]
    }

    # Mock Anthropic API responses
    mock_anthropic_models = {
        "data": [
            {
                "id": "claude-3-5-sonnet-20241022",
                "type": "model",
                "display_name": "Claude 3.5 Sonnet",
                "created_at": "2024-10-22T00:00:00Z",
            },
            {
                "id": "claude-3-opus-20240229",
                "type": "model",
                "display_name": "Claude 3 Opus",
                "created_at": "2024-02-29T00:00:00Z",
            },
        ]
    }

    # Mock the API calls for OpenAI
    async def mock_openai_get_model_list_async(*args, **kwargs):
        return mock_openai_models

    # Mock Anthropic models.list() response
    from unittest.mock import MagicMock

    mock_anthropic_response = MagicMock()
    mock_anthropic_response.model_dump.return_value = mock_anthropic_models

    # Mock the Anthropic AsyncAnthropic client
    class MockAnthropicModels:
        async def list(self):
            return mock_anthropic_response

    class MockAsyncAnthropic:
        def __init__(self, *args, **kwargs):
            self.models = MockAnthropicModels()

    # Patch the actual API calling functions
    monkeypatch.setattr(
        "letta.llm_api.openai.openai_get_model_list_async",
        mock_openai_get_model_list_async,
    )
    monkeypatch.setattr(
        "anthropic.AsyncAnthropic",
        MockAsyncAnthropic,
    )

    # Clear ALL provider-related env vars first to ensure clean state
    monkeypatch.delenv("OPENAI_API_KEY", raising=False)
    monkeypatch.delenv("ANTHROPIC_API_KEY", raising=False)
    monkeypatch.delenv("GEMINI_API_KEY", raising=False)
    monkeypatch.delenv("GOOGLE_CLOUD_PROJECT", raising=False)
    monkeypatch.delenv("GOOGLE_CLOUD_LOCATION", raising=False)
    monkeypatch.delenv("AZURE_API_KEY", raising=False)
    monkeypatch.delenv("GROQ_API_KEY", raising=False)
    monkeypatch.delenv("TOGETHER_API_KEY", raising=False)
    monkeypatch.delenv("VLLM_API_BASE", raising=False)
    monkeypatch.delenv("AWS_ACCESS_KEY_ID", raising=False)
    monkeypatch.delenv("AWS_SECRET_ACCESS_KEY", raising=False)
    monkeypatch.delenv("LMSTUDIO_BASE_URL", raising=False)
    monkeypatch.delenv("DEEPSEEK_API_KEY", raising=False)
    monkeypatch.delenv("XAI_API_KEY", raising=False)
    monkeypatch.delenv("OPENROUTER_API_KEY", raising=False)
    monkeypatch.delenv("ZAI_API_KEY", raising=False)

    # Set environment variables to enable only OpenAI and Anthropic
    monkeypatch.setenv("OPENAI_API_KEY", "sk-test-key-12345")
    monkeypatch.setenv("ANTHROPIC_API_KEY", "sk-ant-test-key-67890")

    # Reload model_settings to pick up new env vars
    from letta.settings import model_settings

    monkeypatch.setattr(model_settings, "openai_api_key", "sk-test-key-12345")
    monkeypatch.setattr(model_settings, "anthropic_api_key", "sk-ant-test-key-67890")
    monkeypatch.setattr(model_settings, "gemini_api_key", None)
    monkeypatch.setattr(model_settings, "google_cloud_project", None)
    monkeypatch.setattr(model_settings, "google_cloud_location", None)
    monkeypatch.setattr(model_settings, "azure_api_key", None)
    monkeypatch.setattr(model_settings, "groq_api_key", None)
    monkeypatch.setattr(model_settings, "together_api_key", None)
    monkeypatch.setattr(model_settings, "vllm_api_base", None)
    monkeypatch.setattr(model_settings, "aws_access_key_id", None)
    monkeypatch.setattr(model_settings, "aws_secret_access_key", None)
    monkeypatch.setattr(model_settings, "lmstudio_base_url", None)
    monkeypatch.setattr(model_settings, "deepseek_api_key", None)
    monkeypatch.setattr(model_settings, "xai_api_key", None)
    monkeypatch.setattr(model_settings, "openrouter_api_key", None)
    monkeypatch.setattr(model_settings, "zai_api_key", None)

    # Create server instance (this will load enabled providers from environment)
    server = SyncServer(init_with_default_org_and_user=False)

    # Manually set up the default user/org (since we disabled auto-init)
    server.default_user = default_user
    server.default_org = default_organization

    # Verify enabled providers were loaded
    assert len(server._enabled_providers) == 3  # Exactly: letta, openai, anthropic
    enabled_provider_names = [p.name for p in server._enabled_providers]
    assert "letta" in enabled_provider_names
    assert "openai" in enabled_provider_names
    assert "anthropic" in enabled_provider_names

    # First, sync base providers to database (this is what init_async does)
    await server.provider_manager.sync_base_providers(
        base_providers=server._enabled_providers,
        actor=default_user,
    )

    # Now call the actual _sync_provider_models_async method
    # This simulates what happens during server startup
    await server._sync_provider_models_async()

    # Verify OpenAI models were synced
    openai_providers = await server.provider_manager.list_providers_async(
        name="openai",
        actor=default_user,
    )
    assert len(openai_providers) == 1, "OpenAI provider should exist"
    openai_provider = openai_providers[0]

    # Check OpenAI LLM models
    openai_llm_models = await server.provider_manager.list_models_async(
        actor=default_user,
        provider_id=openai_provider.id,
        model_type="llm",
    )

    # Should have gpt-4 and gpt-4-turbo (gpt-4-vision filtered out due to "vision" keyword)
    assert len(openai_llm_models) >= 2, f"Expected at least 2 OpenAI LLM models, got {len(openai_llm_models)}"
    openai_model_names = [m.name for m in openai_llm_models]
    assert "gpt-4" in openai_model_names
    assert "gpt-4-turbo" in openai_model_names

    # Check OpenAI embedding models
    openai_embedding_models = await server.provider_manager.list_models_async(
        actor=default_user,
        provider_id=openai_provider.id,
        model_type="embedding",
    )
    assert len(openai_embedding_models) >= 1, "Expected at least 1 OpenAI embedding model"
    embedding_model_names = [m.name for m in openai_embedding_models]
    assert "text-embedding-ada-002" in embedding_model_names

    # Verify model metadata is correct
    gpt4_models = [m for m in openai_llm_models if m.name == "gpt-4"]
    assert len(gpt4_models) > 0, "gpt-4 model should exist"
    gpt4_model = gpt4_models[0]
    assert gpt4_model.handle == "openai/gpt-4"
    assert gpt4_model.model_endpoint_type == "openai"
    assert gpt4_model.max_context_window == 8192
    assert gpt4_model.enabled is True

    # Verify Anthropic models were synced
    anthropic_providers = await server.provider_manager.list_providers_async(
        name="anthropic",
        actor=default_user,
    )
    assert len(anthropic_providers) == 1, "Anthropic provider should exist"
    anthropic_provider = anthropic_providers[0]

    anthropic_llm_models = await server.provider_manager.list_models_async(
        actor=default_user,
        provider_id=anthropic_provider.id,
        model_type="llm",
    )

    # Should have Claude models
    assert len(anthropic_llm_models) >= 2, f"Expected at least 2 Anthropic models, got {len(anthropic_llm_models)}"
    anthropic_model_names = [m.name for m in anthropic_llm_models]
    assert "claude-3-5-sonnet-20241022" in anthropic_model_names
    assert "claude-3-opus-20240229" in anthropic_model_names

    # Test that we can retrieve LLMConfig from handle
    llm_config = await server.provider_manager.get_llm_config_from_handle(
        handle="openai/gpt-4",
        actor=default_user,
    )
    assert llm_config.model == "gpt-4"
    assert llm_config.handle == "openai/gpt-4"
    assert llm_config.provider_name == "openai"
    assert llm_config.context_window == 8192

    # Test that we can retrieve EmbeddingConfig from handle
    embedding_config = await server.provider_manager.get_embedding_config_from_handle(
        handle="openai/text-embedding-ada-002",
        actor=default_user,
    )
    assert embedding_config.embedding_model == "text-embedding-ada-002"
    assert embedding_config.handle == "openai/text-embedding-ada-002"
    assert embedding_config.embedding_dim == 1536


@pytest.mark.asyncio
async def test_server_startup_handles_disabled_providers(default_user, default_organization, monkeypatch):
    """Test that server startup properly handles providers that are no longer enabled.

    This test verifies that:
    1. Base providers that are no longer enabled (env vars removed) are deleted
    2. BYOK providers that are no longer enabled are NOT deleted (user-created)
    3. The sync process handles providers gracefully when API calls fail
    """
    from letta.schemas.providers import OpenAIProvider, ProviderCreate
    from letta.server.server import SyncServer

    # First, manually create providers in the database
    provider_manager = ProviderManager()

    # Create a base OpenAI provider (simulating it was synced before)
    base_openai_create = ProviderCreate(
        name="openai",
        provider_type=ProviderType.openai,
        api_key="sk-old-key",
        base_url="https://api.openai.com/v1",
    )
    base_openai = await provider_manager.create_provider_async(
        base_openai_create,
        actor=default_user,
        is_byok=False,  # This is a base provider
    )

    # Create a BYOK provider (user-created)
    byok_provider_create = ProviderCreate(
        name="my-custom-openai",
        provider_type=ProviderType.openai,
        api_key="sk-my-key",
        base_url="https://api.openai.com/v1",
    )
    byok_provider = await provider_manager.create_provider_async(
        byok_provider_create,
        actor=default_user,
        is_byok=True,
    )
    assert byok_provider.provider_category == ProviderCategory.byok

    # Now create server with NO environment variables set (all base providers disabled)
    monkeypatch.delenv("OPENAI_API_KEY", raising=False)
    monkeypatch.delenv("ANTHROPIC_API_KEY", raising=False)

    from letta.settings import model_settings

    monkeypatch.setattr(model_settings, "openai_api_key", None)
    monkeypatch.setattr(model_settings, "anthropic_api_key", None)

    # Create server instance
    server = SyncServer(init_with_default_org_and_user=False)
    server.default_user = default_user
    server.default_org = default_organization

    # Verify only letta provider is enabled (no openai)
    enabled_names = [p.name for p in server._enabled_providers]
    assert "letta" in enabled_names
    assert "openai" not in enabled_names

    # Sync base providers (should not include openai anymore)
    await server.provider_manager.sync_base_providers(
        base_providers=server._enabled_providers,
        actor=default_user,
    )

    # Call _sync_provider_models_async
    await server._sync_provider_models_async()

    # Verify base OpenAI provider was deleted (no longer enabled)
    try:
        await server.provider_manager.get_provider_async(base_openai.id, actor=default_user)
        assert False, "Base OpenAI provider should have been deleted"
    except Exception:
        # Expected - provider should not exist
        pass

    # Verify BYOK provider still exists (should NOT be deleted)
    byok_still_exists = await server.provider_manager.get_provider_async(
        byok_provider.id,
        actor=default_user,
    )
    assert byok_still_exists is not None
    assert byok_still_exists.name == "my-custom-openai"
    assert byok_still_exists.provider_category == ProviderCategory.byok


@pytest.mark.asyncio
async def test_server_startup_handles_api_errors_gracefully(default_user, default_organization, monkeypatch):
    """Test that server startup handles API errors gracefully without crashing.

    This test verifies that:
    1. If a provider's API call fails during sync, it logs an error but continues
    2. Other providers can still sync successfully
    3. The server startup completes without crashing
    """
    from letta.schemas.providers import AnthropicProvider, OpenAIProvider
    from letta.server.server import SyncServer

    # Mock OpenAI to fail
    async def mock_openai_fail(*args, **kwargs):
        raise Exception("OpenAI API is down")

    # Mock Anthropic to succeed
    from unittest.mock import MagicMock

    mock_anthropic_response = MagicMock()
    mock_anthropic_response.model_dump.return_value = {
        "data": [
            {
                "id": "claude-3-5-sonnet-20241022",
                "type": "model",
                "display_name": "Claude 3.5 Sonnet",
                "created_at": "2024-10-22T00:00:00Z",
            }
        ]
    }

    class MockAnthropicModels:
        async def list(self):
            return mock_anthropic_response

    class MockAsyncAnthropic:
        def __init__(self, *args, **kwargs):
            self.models = MockAnthropicModels()

    monkeypatch.setattr(
        "letta.llm_api.openai.openai_get_model_list_async",
        mock_openai_fail,
    )
    monkeypatch.setattr(
        "anthropic.AsyncAnthropic",
        MockAsyncAnthropic,
    )

    # Set environment variables
    monkeypatch.setenv("OPENAI_API_KEY", "sk-test-key")
    monkeypatch.setenv("ANTHROPIC_API_KEY", "sk-ant-test-key")

    from letta.settings import model_settings

    monkeypatch.setattr(model_settings, "openai_api_key", "sk-test-key")
    monkeypatch.setattr(model_settings, "anthropic_api_key", "sk-ant-test-key")

    # Create server
    server = SyncServer(init_with_default_org_and_user=False)
    server.default_user = default_user
    server.default_org = default_organization

    # Sync base providers
    await server.provider_manager.sync_base_providers(
        base_providers=server._enabled_providers,
        actor=default_user,
    )

    # This should NOT crash even though OpenAI fails
    await server._sync_provider_models_async()

    # Verify Anthropic still synced successfully
    anthropic_providers = await server.provider_manager.list_providers_async(
        name="anthropic",
        actor=default_user,
    )
    assert len(anthropic_providers) == 1

    anthropic_models = await server.provider_manager.list_models_async(
        actor=default_user,
        provider_id=anthropic_providers[0].id,
        model_type="llm",
    )
    assert len(anthropic_models) >= 1, "Anthropic models should have synced despite OpenAI failure"

    # OpenAI should have no models (sync failed)
    openai_providers = await server.provider_manager.list_providers_async(
        name="openai",
        actor=default_user,
    )
    if len(openai_providers) > 0:
        openai_models = await server.provider_manager.list_models_async(
            actor=default_user,
            provider_id=openai_providers[0].id,
        )
        # Models might exist from previous runs, but the sync attempt should have been logged as failed
        # The key is that the server didn't crash