dao-ai 0.1.8__tar.gz → 0.1.9__tar.gz

{dao_ai-0.1.8 → dao_ai-0.1.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dao-ai
-Version: 0.1.8
+Version: 0.1.9
 Summary: DAO AI: A modular, multi-agent orchestration framework for complex AI workflows. Supports agent handoff, tool integration, and dynamic configuration via YAML.
 Project-URL: Homepage, https://github.com/natefleming/dao-ai
 Project-URL: Documentation, https://natefleming.github.io/dao-ai
@@ -43,7 +43,7 @@ Requires-Dist: langgraph>=1.0.5
 Requires-Dist: langmem>=0.0.30
 Requires-Dist: loguru>=0.7.3
 Requires-Dist: mcp>=1.24.0
-Requires-Dist: mlflow>=3.8.1
+Requires-Dist: mlflow[databricks]>=3.8.1
 Requires-Dist: nest-asyncio>=1.6.0
 Requires-Dist: openevals>=0.1.3
 Requires-Dist: openpyxl>=3.1.5

dao_ai-0.1.9/app.yaml

@@ -0,0 +1,12 @@
+command: ["uv", "run", "python", "-m", "dao_ai.app_server"]
+# Databricks Apps listen by default on port 8000
+
+env:
+  - name: MLFLOW_TRACKING_URI
+    value: "databricks"
+  - name: MLFLOW_REGISTRY_URI
+    value: "databricks-uc"
+  # Path to the dao-ai configuration file
+  # This can be set to a workspace path or local file path
+  - name: DAO_AI_CONFIG_PATH
+    value: "model_config.yaml"

{dao_ai-0.1.8 → dao_ai-0.1.9}/config/examples/02_mcp/filtered_mcp.yaml

@@ -85,8 +85,8 @@ tools:
         - execute_query        # Exact match
         - list_tables          # Exact match
         - describe_table       # Exact match
-        - get_*                # Pattern: all getters
-        - show_*               # Pattern: all show operations
+        - "get_*"              # Pattern: all getters
+        - "show_*"             # Pattern: all show operations
 
   # ---------------------------------------------------------------------------
   # Example 2: Exclude Dangerous Tools (Denylist)
@@ -104,11 +104,11 @@ tools:
       workspace_host: *workspace_host
       # Load all tools EXCEPT these dangerous ones
       exclude_tools:
-        - drop_*               # Block all drop operations
-        - delete_*             # Block all delete operations
-        - truncate_*           # Block all truncate operations
+        - "drop_*"             # Block all drop operations
+        - "delete_*"           # Block all delete operations
+        - "truncate_*"         # Block all truncate operations
         - execute_ddl          # Block DDL execution
-        - alter_*              # Block all alter operations
+        - "alter_*"            # Block all alter operations
 
   # ---------------------------------------------------------------------------
   # Example 3: Hybrid Filtering (Include + Exclude)
@@ -126,14 +126,14 @@ tools:
       workspace_host: *workspace_host
       # Start with these categories
       include_tools:
-        - query_*              # All query functions
-        - get_*                # All getter functions
-        - list_*               # All list functions
+        - "query_*"            # All query functions
+        - "get_*"              # All getter functions
+        - "list_*"             # All list functions
       # But exclude sensitive ones
       exclude_tools:
-        - *_sensitive          # Exclude anything with "_sensitive"
-        - *_admin              # Exclude admin functions
-        - get_secret_*         # Exclude secret getters
+        - "*_sensitive"        # Exclude anything with "_sensitive"
+        - "*_admin"            # Exclude admin functions
+        - "get_secret_*"       # Exclude secret getters
 
   # ---------------------------------------------------------------------------
   # Example 4: Pattern-Based Inclusion
@@ -151,10 +151,10 @@ tools:
       workspace_host: *workspace_host
       # Only read operations with patterns
       include_tools:
-        - query_*              # All queries
-        - list_*               # All lists
-        - describe_*           # All describe operations
-        - show_*               # All show operations
+        - "query_*"            # All queries
+        - "list_*"             # All lists
+        - "describe_*"         # All describe operations
+        - "show_*"             # All show operations
 
   # ---------------------------------------------------------------------------
   # Example 5: Maximum Security (Very Restrictive)
@@ -192,8 +192,8 @@ tools:
       workspace_host: *workspace_host
       # Allow everything except the really dangerous stuff
       exclude_tools:
-        - drop_*               # Can't drop anything
-        - truncate_*           # Can't truncate
+        - "drop_*"             # Can't drop anything
+        - "truncate_*"         # Can't truncate
         - execute_ddl          # Can't run arbitrary DDL
 
 # =============================================================================
@@ -295,17 +295,44 @@ agents:
 # APPLICATION CONFIGURATION
 # =============================================================================
 
-app_name: filtered_mcp_example
-entry_agent: safe_sql_agent
+app:
+  name: filtered_mcp_example              # Application name
+  log_level: INFO                         # Logging level (DEBUG, INFO, WARNING, ERROR)
+  registered_model:                       # MLflow registered model configuration
+    schema: *retail_schema                # Schema where model will be registered
+    name: filtered_mcp_example            # Model name in MLflow registry
+  agents:                                 # List of agents included in the system
+    - *safe_sql_agent                     # Safe SQL agent (read-only)
+    - *analyst_agent                      # Data analyst agent
+    - *dev_agent                          # Development agent
+    - *secure_agent                       # High-security agent
+    - *functions_agent                    # Functions agent
+  orchestration:                          # Agent orchestration configuration
+    swarm:                                # Supervisor orchestration pattern
+      model: *default_llm                 # LLM for routing decisions
+      # Agents will be automatically routed based on their capabilities
 
 # =============================================================================
 # USAGE NOTES
 # =============================================================================
 #
+# Deployment:
+#   databricks apps deploy dao-ai-filtered-mcp --source-dir . --config-file config/examples/02_mcp/filtered_mcp.yaml
+#
+# The app uses swarm orchestration to automatically route user queries to the
+# most appropriate agent based on their capabilities and tool access:
+#
+# - safe_sql_agent: Read-only SQL queries (allowlist of 5 safe operations)
+# - analyst_agent: Data analysis queries (query_*, list_*, describe_*, show_*)
+# - dev_agent: Development queries (all tools except dangerous DDL operations)
+# - secure_agent: Maximum security (only 3 specific tools)
+# - functions_agent: UC Functions access (filtered to exclude sensitive/admin)
+#
 # Testing Filters:
-# 1. Run with safe_sql_agent - try to call drop_table (should not be available)
-# 2. Run with dev_agent - verify you can query but not drop
-# 3. Run with secure_agent - verify only 3 tools are available
+# 1. Ask to query data → Routes to analyst_agent or safe_sql_agent
+# 2. Ask to drop a table → Routes to dev_agent, which will refuse (blocked)
+# 3. Ask in secure mode → Routes to secure_agent (only 3 tools available)
+# 4. Ask to call a function → Routes to functions_agent (sensitive functions blocked)
 #
 # Pattern Examples:
 # - "query_*" matches: query_sales, query_inventory, query_anything

{dao_ai-0.1.8 → dao_ai-0.1.9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "dao-ai"
-version = "0.1.8"
+version = "0.1.9"
 description = "DAO AI: A modular, multi-agent orchestration framework for complex AI workflows. Supports agent handoff, tool integration, and dynamic configuration via YAML."
 readme = "README.md"
 license = { text = "MIT" }
@@ -61,7 +61,7 @@ dependencies = [
     "langmem>=0.0.30",
     "loguru>=0.7.3",
     "mcp>=1.24.0",
-    "mlflow>=3.8.1",
+    "mlflow[databricks]>=3.8.1",
     "nest-asyncio>=1.6.0",
     "openevals>=0.1.3",
     "openpyxl>=3.1.5",

{dao_ai-0.1.8 → dao_ai-0.1.9}/requirements.txt

@@ -9,6 +9,9 @@ anthropic==0.75.0
 anyio==4.12.0
 asyncer==0.0.8
 attrs==25.4.0
+azure-core==1.37.0
+azure-storage-blob==12.28.0
+azure-storage-file-datalake==12.23.0
 backoff==2.2.1
 blinker==1.9.0
 boto3==1.42.16
@@ -57,7 +60,12 @@ fsspec==2025.10.0
 gepa==0.0.22
 gitdb==4.0.12
 gitpython==3.1.45
+google-api-core==2.28.1
 google-auth==2.45.0
+google-cloud-core==2.5.0
+google-cloud-storage==3.7.0
+google-crc32c==1.8.0
+google-resumable-media==2.8.0
 googleapis-common-protos==1.72.0
 grandalf==0.8
 graphene==3.4.3
@@ -81,6 +89,7 @@ identify==2.6.15
 idna==3.11
 importlib-metadata==8.7.1
 iniconfig==2.3.0
+isodate==0.7.2
 itsdangerous==2.2.0
 jinja2==3.1.6
 jiter==0.12.0
@@ -151,6 +160,7 @@ pluggy==1.6.0
 pre-commit==4.5.1
 primp==0.15.0
 propcache==0.4.1
+proto-plus==1.27.0
 protobuf==6.33.2
 psycopg==3.3.2
 psycopg-binary==3.3.2

{dao_ai-0.1.8 → dao_ai-0.1.9}/schemas/model_config_schema.json

@@ -5150,6 +5150,18 @@
   },
   "additionalProperties": false,
   "properties": {
+    "version": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "title": "Version"
+    },
     "variables": {
       "additionalProperties": {
         "anyOf": [

dao_ai-0.1.9/src/dao_ai/app_server.py

@@ -0,0 +1,103 @@
+"""
+App server module for running dao-ai agents as Databricks Apps.
+
+This module provides the entry point for deploying dao-ai agents as Databricks Apps
+using MLflow's AgentServer. It follows the same pattern as agent_as_code.py but
+uses the AgentServer for the Databricks Apps runtime.
+
+Configuration Loading:
+    The config path is specified via the DAO_AI_CONFIG_PATH environment variable,
+    or defaults to model_config.yaml in the current directory.
+
+Usage:
+    # With environment variable
+    DAO_AI_CONFIG_PATH=/path/to/config.yaml python -m dao_ai.app_server
+
+    # With default model_config.yaml in current directory
+    python -m dao_ai.app_server
+"""
+
+import os
+from typing import AsyncGenerator
+
+import mlflow
+from dotenv import load_dotenv
+from mlflow.genai.agent_server import AgentServer, invoke, stream
+from mlflow.pyfunc import ResponsesAgent
+from mlflow.types.responses import (
+    ResponsesAgentRequest,
+    ResponsesAgentResponse,
+    ResponsesAgentStreamEvent,
+)
+
+from dao_ai.config import AppConfig
+from dao_ai.logging import configure_logging
+
+# Load environment variables from .env.local if it exists
+load_dotenv(dotenv_path=".env.local", override=True)
+
+# Configure MLflow
+mlflow.set_registry_uri("databricks-uc")
+mlflow.set_tracking_uri("databricks")
+mlflow.langchain.autolog()
+
+# Get config path from environment or use default
+config_path: str = os.environ.get("DAO_AI_CONFIG_PATH", "model_config.yaml")
+
+# Load configuration using AppConfig.from_file (consistent with CLI, notebook, builder)
+config: AppConfig = AppConfig.from_file(config_path)
+
+# Configure logging
+if config.app and config.app.log_level:
+    configure_logging(level=config.app.log_level)
+
+# Create the ResponsesAgent
+_responses_agent: ResponsesAgent = config.as_responses_agent()
+
+
+@invoke()
+def non_streaming(request: ResponsesAgentRequest) -> ResponsesAgentResponse:
+    """
+    Handle non-streaming requests by delegating to the ResponsesAgent.
+
+    Args:
+        request: The incoming ResponsesAgentRequest
+
+    Returns:
+        ResponsesAgentResponse with the complete output
+    """
+    return _responses_agent.predict(request)
+
+
+@stream()
+def streaming(
+    request: ResponsesAgentRequest,
+) -> AsyncGenerator[ResponsesAgentStreamEvent, None]:
+    """
+    Handle streaming requests by delegating to the ResponsesAgent.
+
+    Args:
+        request: The incoming ResponsesAgentRequest
+
+    Yields:
+        ResponsesAgentStreamEvent objects as they are generated
+    """
+    # The predict_stream method returns a generator, convert to async generator
+    for event in _responses_agent.predict_stream(request):
+        yield event
+
+
+# Create the AgentServer instance
+agent_server = AgentServer("ResponsesAgent", enable_chat_proxy=True)
+
+# Define the app as a module level variable to enable multiple workers
+app = agent_server.app
+
+
+def main() -> None:
+    """Entry point for running the agent server."""
+    agent_server.run(app_import_string="dao_ai.app_server:app")
+
+
+if __name__ == "__main__":
+    main()

{dao_ai-0.1.8 → dao_ai-0.1.9}/src/dao_ai/cli.py

@@ -309,6 +309,14 @@ Examples:
         metavar="FILE",
         help="Path to the model configuration file to validate",
     )
+    deploy_parser.add_argument(
+        "-t",
+        "--target",
+        type=str,
+        choices=["model_serving", "apps"],
+        default="model_serving",
+        help="Deployment target: 'model_serving' (default) or 'apps'",
+    )
 
     # List MCP tools command
     list_mcp_parser: ArgumentParser = subparsers.add_parser(
@@ -729,11 +737,17 @@ def handle_graph_command(options: Namespace) -> None:
 
 
 def handle_deploy_command(options: Namespace) -> None:
+    from dao_ai.config import DeploymentTarget
+
     logger.debug(f"Validating configuration from {options.config}...")
     try:
         config: AppConfig = AppConfig.from_file(options.config)
+
+        # Convert target string to enum
+        target: DeploymentTarget = DeploymentTarget(options.target)
+
         config.create_agent()
-        config.deploy_agent()
+        config.deploy_agent(target=target)
         sys.exit(0)
     except Exception as e:
         logger.error(f"Deployment failed: {e}")

{dao_ai-0.1.8 → dao_ai-0.1.9}/src/dao_ai/config.py

@@ -208,7 +208,9 @@ class IsDatabricksResource(ABC, BaseModel):
     Authentication Options:
     ----------------------
     1. **On-Behalf-Of User (OBO)**: Set on_behalf_of_user=True to use the
-       calling user's identity via ModelServingUserCredentials.
+       calling user's identity. Implementation varies by deployment:
+       - Databricks Apps: Uses X-Forwarded-Access-Token from request headers
+       - Model Serving: Uses ModelServingUserCredentials
 
     2. **Service Principal (OAuth M2M)**: Provide service_principal or
        (client_id + client_secret + workspace_host) for service principal auth.
@@ -221,9 +223,17 @@ class IsDatabricksResource(ABC, BaseModel):
 
     Authentication Priority:
     1. OBO (on_behalf_of_user=True)
+       - Checks for forwarded headers (Databricks Apps)
+       - Falls back to ModelServingUserCredentials (Model Serving)
     2. Service Principal (client_id + client_secret + workspace_host)
     3. PAT (pat + workspace_host)
     4. Ambient/default authentication
+
+    Note: When on_behalf_of_user=True, the agent acts as the calling user regardless
+    of deployment target. In Databricks Apps, this uses X-Forwarded-Access-Token
+    automatically captured by MLflow AgentServer. In Model Serving, this uses
+    ModelServingUserCredentials. Forwarded headers are ONLY used when
+    on_behalf_of_user=True.
     """
 
     model_config = ConfigDict(use_enum_values=True)
@@ -235,9 +245,6 @@ class IsDatabricksResource(ABC, BaseModel):
     workspace_host: Optional[AnyVariable] = None
     pat: Optional[AnyVariable] = None
 
-    # Private attribute to cache the workspace client (lazy instantiation)
-    _workspace_client: Optional[WorkspaceClient] = PrivateAttr(default=None)
-
     @abstractmethod
     def as_resources(self) -> Sequence[DatabricksResource]: ...
 
@@ -273,32 +280,56 @@ class IsDatabricksResource(ABC, BaseModel):
         """
         Get a WorkspaceClient configured with the appropriate authentication.
 
-        The client is lazily instantiated on first access and cached for subsequent calls.
+        A new client is created on each access.
 
         Authentication priority:
-        1. If on_behalf_of_user is True, uses ModelServingUserCredentials (OBO)
-        2. If service principal credentials are configured (client_id, client_secret,
-           workspace_host), uses OAuth M2M
-        3. If PAT is configured, uses token authentication
-        4. Otherwise, uses default/ambient authentication
+        1. On-Behalf-Of User (on_behalf_of_user=True):
+           - Forwarded headers (Databricks Apps)
+           - ModelServingUserCredentials (Model Serving)
+        2. Service Principal (client_id + client_secret + workspace_host)
+        3. PAT (pat + workspace_host)
+        4. Ambient/default authentication
         """
-        # Return cached client if already instantiated
-        if self._workspace_client is not None:
-            return self._workspace_client
-
         from dao_ai.utils import normalize_host
 
         # Check for OBO first (highest priority)
         if self.on_behalf_of_user:
+            # NEW: In Databricks Apps, use forwarded headers for per-user auth
+            try:
+                from mlflow.genai.agent_server import get_request_headers
+
+                headers = get_request_headers()
+                forwarded_token = headers.get("x-forwarded-access-token")
+
+                if forwarded_token:
+                    forwarded_user = headers.get("x-forwarded-user", "unknown")
+                    logger.debug(
+                        f"Creating WorkspaceClient for {self.__class__.__name__} "
+                        f"with OBO using forwarded token from Databricks Apps",
+                        forwarded_user=forwarded_user,
+                    )
+                    # Use workspace_host if configured, otherwise SDK will auto-detect
+                    workspace_host_value: str | None = (
+                        normalize_host(value_of(self.workspace_host))
+                        if self.workspace_host
+                        else None
+                    )
+                    return WorkspaceClient(
+                        host=workspace_host_value,
+                        token=forwarded_token,
+                        auth_type="pat",
+                    )
+            except (ImportError, LookupError):
+                # mlflow not available or headers not set - fall through to Model Serving
+                pass
+
+            # Fall back to Model Serving OBO (existing behavior)
             credentials_strategy: CredentialsStrategy = ModelServingUserCredentials()
             logger.debug(
                 f"Creating WorkspaceClient for {self.__class__.__name__} "
-                f"with OBO credentials strategy"
-            )
-            self._workspace_client = WorkspaceClient(
-                credentials_strategy=credentials_strategy
+                f"with OBO credentials strategy (Model Serving)"
             )
-            return self._workspace_client
+            return WorkspaceClient(credentials_strategy=credentials_strategy)
 
         # Check for service principal credentials
         client_id_value: str | None = (
@@ -318,13 +349,12 @@ class IsDatabricksResource(ABC, BaseModel):
                 f"Creating WorkspaceClient for {self.__class__.__name__} with service principal: "
                 f"client_id={client_id_value}, host={workspace_host_value}"
             )
-            self._workspace_client = WorkspaceClient(
+            return WorkspaceClient(
                 host=workspace_host_value,
                 client_id=client_id_value,
                 client_secret=client_secret_value,
                 auth_type="oauth-m2m",
             )
-            return self._workspace_client
 
         # Check for PAT authentication
         pat_value: str | None = value_of(self.pat) if self.pat else None
@@ -332,20 +362,28 @@ class IsDatabricksResource(ABC, BaseModel):
             logger.debug(
                 f"Creating WorkspaceClient for {self.__class__.__name__} with PAT"
             )
-            self._workspace_client = WorkspaceClient(
+            return WorkspaceClient(
                 host=workspace_host_value,
                 token=pat_value,
                 auth_type="pat",
             )
-            return self._workspace_client
 
         # Default: use ambient authentication
         logger.debug(
             f"Creating WorkspaceClient for {self.__class__.__name__} "
             "with default/ambient authentication"
         )
-        self._workspace_client = WorkspaceClient()
-        return self._workspace_client
+        return WorkspaceClient()
+
+
+class DeploymentTarget(str, Enum):
+    """Target platform for agent deployment."""
+
+    MODEL_SERVING = "model_serving"
+    """Deploy to Databricks Model Serving endpoint."""
+
+    APPS = "apps"
+    """Deploy as a Databricks App."""
 
 
 class Privilege(str, Enum):
@@ -865,10 +903,6 @@ class GenieRoomModel(IsDatabricksResource):
                 pat=self.pat,
             )
 
-            # Share the cached workspace client if available
-            if self._workspace_client is not None:
-                warehouse_model._workspace_client = self._workspace_client
-
             return warehouse_model
         except Exception as e:
             logger.warning(
@@ -912,9 +946,6 @@ class GenieRoomModel(IsDatabricksResource):
                                 workspace_host=self.workspace_host,
                                 pat=self.pat,
                             )
-                            # Share the cached workspace client if available
-                            if self._workspace_client is not None:
-                                table_model._workspace_client = self._workspace_client
 
                             # Verify the table exists before adding
                             if not table_model.exists():
@@ -952,9 +983,6 @@ class GenieRoomModel(IsDatabricksResource):
                 workspace_host=self.workspace_host,
                 pat=self.pat,
             )
-            # Share the cached workspace client if available
-            if self._workspace_client is not None:
-                function_model._workspace_client = self._workspace_client
 
             # Verify the function exists before adding
             if not function_model.exists():
@@ -3255,6 +3283,7 @@ class ResourcesModel(BaseModel):
 
 class AppConfig(BaseModel):
     model_config = ConfigDict(use_enum_values=True, extra="forbid")
+    version: Optional[str] = None
     variables: dict[str, AnyVariable] = Field(default_factory=dict)
     service_principals: dict[str, ServicePrincipalModel] = Field(default_factory=dict)
     schemas: dict[str, SchemaModel] = Field(default_factory=dict)
@@ -3275,6 +3304,9 @@ class AppConfig(BaseModel):
     )
     providers: Optional[dict[type | str, Any]] = None
 
+    # Private attribute to track the source config file path (set by from_file)
+    _source_config_path: str | None = None
+
     @classmethod
     def from_file(cls, path: PathLike) -> "AppConfig":
         path = Path(path).as_posix()
@@ -3282,12 +3314,20 @@ class AppConfig(BaseModel):
         model_config: ModelConfig = ModelConfig(development_config=path)
         config: AppConfig = AppConfig(**model_config.to_dict())
 
+        # Store the source config path for later use (e.g., Apps deployment)
+        config._source_config_path = path
+
         config.initialize()
 
         atexit.register(config.shutdown)
 
         return config
 
+    @property
+    def source_config_path(self) -> str | None:
+        """Get the source config file path if loaded via from_file."""
+        return self._source_config_path
+
     def initialize(self) -> None:
         from dao_ai.hooks.core import create_hooks
         from dao_ai.logging import configure_logging
@@ -3358,6 +3398,7 @@ class AppConfig(BaseModel):
 
     def deploy_agent(
         self,
+        target: DeploymentTarget = DeploymentTarget.MODEL_SERVING,
         w: WorkspaceClient | None = None,
         vsc: "VectorSearchClient | None" = None,
         pat: str | None = None,
@@ -3365,6 +3406,18 @@ class AppConfig(BaseModel):
         client_secret: str | None = None,
         workspace_host: str | None = None,
     ) -> None:
+        """
+        Deploy the agent to the specified target.
+
+        Args:
+            target: The deployment target (MODEL_SERVING or APPS). Defaults to MODEL_SERVING.
+            w: Optional WorkspaceClient instance
+            vsc: Optional VectorSearchClient instance
+            pat: Optional personal access token for authentication
+            client_id: Optional client ID for service principal authentication
+            client_secret: Optional client secret for service principal authentication
+            workspace_host: Optional workspace host URL
+        """
         from dao_ai.providers.base import ServiceProvider
         from dao_ai.providers.databricks import DatabricksProvider
 
@@ -3376,7 +3429,7 @@ class AppConfig(BaseModel):
             client_secret=client_secret,
             workspace_host=workspace_host,
         )
-        provider.deploy_agent(self)
+        provider.deploy_agent(self, target=target)
 
     def find_agents(
         self, predicate: Callable[[AgentModel], bool] | None = None

{dao_ai-0.1.8 → dao_ai-0.1.9}/src/dao_ai/providers/base.py

@@ -1,15 +1,19 @@
 from abc import ABC, abstractmethod
-from typing import Any, Sequence
+from typing import TYPE_CHECKING, Any, Sequence
 
 from dao_ai.config import (
     AppModel,
     DatasetModel,
+    DeploymentTarget,
     SchemaModel,
     UnityCatalogFunctionSqlModel,
     VectorStoreModel,
     VolumeModel,
 )
 
+if TYPE_CHECKING:
+    from dao_ai.config import AppConfig
+
 
 class ServiceProvider(ABC):
     @abstractmethod
@@ -52,4 +56,26 @@ class ServiceProvider(ABC):
     ) -> Any: ...
 
     @abstractmethod
-    def deploy_agent(self, config: AppModel) -> Any: ...
+    def deploy_model_serving_agent(self, config: "AppConfig") -> Any:
+        """Deploy agent to Databricks Model Serving endpoint."""
+        ...
+
+    @abstractmethod
+    def deploy_apps_agent(self, config: "AppConfig") -> Any:
+        """Deploy agent as a Databricks App."""
+        ...
+
+    @abstractmethod
+    def deploy_agent(
+        self,
+        config: "AppConfig",
+        target: DeploymentTarget = DeploymentTarget.MODEL_SERVING,
+    ) -> Any:
+        """
+        Deploy agent to the specified target.
+
+        Args:
+            config: The AppConfig containing deployment configuration
+            target: The deployment target (MODEL_SERVING or APPS)
+        """
+        ...