dao-ai 0.0.12__tar.gz → 0.0.14__tar.gz

{dao_ai-0.0.12 → dao_ai-0.0.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dao-ai
-Version: 0.0.12
+Version: 0.0.14
 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
@@ -38,7 +38,7 @@ Requires-Dist: langgraph>=0.6.5
 Requires-Dist: langmem>=0.0.29
 Requires-Dist: loguru>=0.7.3
 Requires-Dist: mcp>=1.9.1
-Requires-Dist: mlflow>=3.1.1
+Requires-Dist: mlflow>=3.3.2
 Requires-Dist: nest-asyncio>=1.6.0
 Requires-Dist: openevals>=0.0.19
 Requires-Dist: openpyxl>=3.1.5

{dao_ai-0.0.12 → dao_ai-0.0.14}/config/examples/genie.yaml

@@ -5,11 +5,10 @@ schemas:
   retail_schema: &retail_schema
     catalog_name: nfleming                    # Unity Catalog name
     schema_name: retail_ai                    # Schema within the catalog
-    # permissions:                              # Access permissions for the schema
-    #   - principals: [users]                   # Grant access to all users
-    #     privileges:
-    #       - ALL_PRIVILEGES                    # Full permissions for demo purposes
 
+  quick_serve_retaurant_schema: &quick_serve_retaurant_schema
+    catalog_name: retail_consumer_goods                    # Unity Catalog name
+    schema_name: quick_serve_restaurant                    # Schema within the catalog
 
 resources:
   llms:
@@ -20,6 +19,40 @@ resources:
       max_tokens: 8192                              # Maximum tokens per response
       on_behalf_of_user: False
 
+  tables:
+    product_table: &product_table
+      schema: *quick_serve_retaurant_schema                        # Reference to schema defined above
+      name: products                                # Table name
+    
+    inventory_table: &inventory_table
+      schema: *quick_serve_retaurant_schema
+      name: inventory
+
+    items_description: &items_description
+      schema: *quick_serve_retaurant_schema
+      name: items_description
+
+    fulfil_item_orders: &fulfil_item_orders
+      schema: *quick_serve_retaurant_schema
+      name: fulfil_item_orders
+
+    orders: &orders
+      schema: *quick_serve_retaurant_schema
+      name: orders
+
+  functions:
+
+    lookup_items_by_descriptions: &lookup_items_by_descriptions
+      schema: *quick_serve_retaurant_schema
+      name: lookup_items_by_descriptions
+
+    match_item_by_description_and_price: &match_item_by_description_and_price
+      schema: *quick_serve_retaurant_schema
+      name: match_item_by_description_and_price     
+    
+    match_historical_item_order_by_date: &match_historical_item_order_by_date
+      schema: *quick_serve_retaurant_schema
+      name: match_historical_item_order_by_date     
 
   genie_rooms:
     # Genie space for retail data queries
@@ -37,7 +70,7 @@ tools:
       name: dao_ai.tools.create_genie_tool       # Factory function path
       args:                                         # Arguments passed to factory
         name: my_genie_tool
-        description: My Answers questions about genie
+        description: Answers questions about retail products and inventory
         genie_room: *retail_genie_room              # Reference to Genie room config
 
 
@@ -49,9 +82,8 @@ agents:
     tools:                                          # Tools available to this agent
       - *genie_tool
     prompt: |                                       # System prompt defining agent behavior
-      Answers foobar questions about genie
-    handoff_prompt: |                               # Conditions for routing to this agent
-      Answers all foobar questions
+      Answers questions about retail products and inventory
+
 
 app:
   name: genie_example                      # Application name  

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/01_ingest_and_transform.py

@@ -1,5 +1,5 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
 # MAGIC %restart_python
 
 # COMMAND ----------

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/02_provision_vector_search.py

@@ -1,7 +1,7 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/03_generate_evaluation_data.py

@@ -1,7 +1,7 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------
@@ -65,26 +65,43 @@ for _, vector_store in config.resources.vector_stores.items():
     spark.table(vector_store.source_table.full_name)
     .withColumn("id", F.col(vector_store.primary_key))
     .withColumn("content", F.col(vector_store.embedding_source_column))
-    .withColumn("doc_uri", F.lit("source"))
+    .withColumn("doc_uri", doc_uri)
   )
   parsed_docs_pdf: pd.DataFrame = parsed_docs_df.toPandas()
 
   display(parsed_docs_pdf)
 
-  agent_description = f"""
-  The agent is a RAG chatbot that answers questions about retail hardware and gives recommendations for purchases. 
-  """
-  question_guidelines = f"""
-  # User personas
-  - An employee or client asking about products and inventory
-
-
-  # Example questions
-  - What grills do you have in stock?
-  - Can you recommend a accessories for my Toro lawn mower?
-
-  # Additional Guidelines
-  - Questions should be succinct, and human-like
+  agent_description: str = evaluation.agent_description
+  if not agent_description:
+      agent_description = """
+  A general-purpose chatbot AI agent is designed to engage in natural conversations 
+  across diverse topics and tasks, drawing from broad knowledge to answer questions, 
+  assist with writing, solve problems, and provide explanations while maintaining 
+  context throughout interactions. It aims to be a versatile, adaptable assistant 
+  that can help with the wide spectrum of things people encounter in daily life, 
+  adjusting its communication style and level of detail based on user needs.
+      """
+
+  question_guidelines: str = evaluation.question_guidelines
+  if not question_guidelines:
+      question_guidelines = f"""
+# User personas
+- A curious individual seeking information or explanations
+- A student looking for homework help or learning assistance  
+- A professional needing quick research or writing support
+- A creative person brainstorming ideas or seeking inspiration
+
+# Example questions
+- Can you explain how photosynthesis works?
+- Help me write a professional email to my boss
+- What are some good books similar to Harry Potter?
+- How do I fix a leaky faucet?
+
+# Additional Guidelines  
+- Questions should be conversational and natural
+- Users may ask follow-up questions to dig deeper into topics
+- Requests can range from simple facts to complex multi-step tasks
+- Tone can vary from casual chat to formal assistance
   """
 
   evals_pdf: pd.DataFrame = generate_evals_df(

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/04_unity_catalog_tools.py

@@ -1,7 +1,7 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/05_agent_as_code_driver.py

@@ -1,7 +1,7 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/06_run_evaluation.py

@@ -1,8 +1,8 @@
 # Databricks notebook source
 # DBTITLE 1,Install Required Python Packages and Restart Kernel
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------
@@ -127,11 +127,6 @@ from mlflow.genai.scorers import scorer, Safety, Guidelines
 from mlflow.entities import Feedback, Trace
 
 
-clarity = Guidelines(
-    name="clarity",
-    guidelines=["The response must be clear, coherent, and concise"],
-)
-
 @scorer
 def response_completeness(outputs: dict[str, Any]) -> Feedback:
 
@@ -225,7 +220,7 @@ eval_df: pd.DataFrame = spark.read.table(payload_table).toPandas()
 
 evaluation_table_name: str = config.evaluation.table.full_name
 
-scorers = [Safety(), clarity, response_completeness, tool_call_efficiency]
+scorers = [Safety(), response_completeness, tool_call_efficiency]
 custom_scorers = []
 
 if config.evaluation.guidelines:

{dao_ai-0.0.12 → dao_ai-0.0.14}/notebooks/08_run_examples.py

@@ -1,7 +1,7 @@
 # Databricks notebook source
-# MAGIC %pip install --quiet -r ../requirements.txt
-# MAGIC %pip uninstall -y databricks-connect pyspark pyspark-connect
-# MAGIC %pip install databricks-connect
+# MAGIC %pip install --quiet --upgrade -r ../requirements.txt
+# MAGIC %pip uninstall --quiet -y databricks-connect pyspark pyspark-connect
+# MAGIC %pip install --quiet databricks-connect
 # MAGIC %restart_python
 
 # COMMAND ----------
@@ -50,15 +50,7 @@ mlflow.langchain.autolog()
 
 config: AppConfig = AppConfig.from_file(path=config_path)
 
-log_level: str = config.app.log_level
-
-logger.remove()
-logger.add(sys.stderr, level=log_level)
-
-graph: CompiledStateGraph = create_dao_ai_graph(config=config)
-
-app: ChatModel = create_agent(graph)
-
+app: ChatModel = config.as_chat_model()
 
 # COMMAND ----------
 

{dao_ai-0.0.12 → dao_ai-0.0.14}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "dao-ai"
-version = "0.0.12"
+version = "0.0.14"
 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" }
@@ -56,7 +56,7 @@ dependencies = [
     "langmem>=0.0.29",
     "loguru>=0.7.3",
     "mcp>=1.9.1",
-    "mlflow>=3.1.1",
+    "mlflow>=3.3.2",
     "nest-asyncio>=1.6.0",
     "openevals>=0.0.19",
     "openpyxl>=3.1.5",

{dao_ai-0.0.12 → dao_ai-0.0.14}/requirements.txt

@@ -12,11 +12,13 @@ boto3==1.40.1
 botocore==1.40.1
 cachetools==5.5.2
 certifi==2025.6.15
+cffi==1.17.1
 cfgv==3.4.0
 charset-normalizer==3.4.2
 click==8.2.1
 cloudpickle==3.1.1
 contourpy==1.3.2
+cryptography==45.0.7
 cycler==0.12.1
 databricks-agents==1.2.0
 databricks-ai-bridge==0.5.1
@@ -88,8 +90,9 @@ marshmallow==3.26.1
 matplotlib==3.10.3
 mcp==1.9.4
 mdurl==0.1.2
-mlflow==3.1.1
-mlflow-skinny==3.1.1
+mlflow==3.3.2
+mlflow-skinny==3.3.2
+mlflow-tracing==3.3.2
 multidict==6.5.0
 mypy==1.16.1
 mypy-extensions==1.1.0
@@ -121,6 +124,7 @@ py4j==0.10.9.7
 pyarrow==20.0.0
 pyasn1==0.6.1
 pyasn1-modules==0.4.2
+pycparser==2.22
 pydantic==2.11.7
 pydantic-core==2.33.2
 pydantic-settings==2.10.0

{dao_ai-0.0.12 → dao_ai-0.0.14}/schemas/model_config_schema.json

@@ -1086,6 +1086,30 @@
           "title": "Num Evals",
           "type": "integer"
         },
+        "agent_description": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null,
+          "title": "Agent Description"
+        },
+        "question_guidelines": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null,
+          "title": "Question Guidelines"
+        },
         "custom_inputs": {
           "additionalProperties": true,
           "title": "Custom Inputs",
@@ -2547,6 +2571,28 @@
         "source_table": {
           "$ref": "#/$defs/TableModel"
         },
+        "source_path": {
+          "anyOf": [
+            {
+              "$ref": "#/$defs/VolumePathModel"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null
+        },
+        "checkpoint_path": {
+          "anyOf": [
+            {
+              "$ref": "#/$defs/VolumePathModel"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null
+        },
         "primary_key": {
           "anyOf": [
             {
@@ -2626,17 +2672,29 @@
       "additionalProperties": false,
       "properties": {
         "volume": {
-          "$ref": "#/$defs/VolumeModel"
+          "anyOf": [
+            {
+              "$ref": "#/$defs/VolumeModel"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null
         },
         "path": {
-          "title": "Path",
-          "type": "string"
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "default": null,
+          "title": "Path"
         }
       },
-      "required": [
-        "volume",
-        "path"
-      ],
       "title": "VolumePathModel",
       "type": "object"
     },

{dao_ai-0.0.12 → dao_ai-0.0.14}/src/dao_ai/agent_as_code.py

@@ -1,14 +1,11 @@
 import sys
 
 import mlflow
-from langgraph.graph.state import CompiledStateGraph
 from loguru import logger
 from mlflow.models import ModelConfig
 from mlflow.pyfunc import ChatModel
 
 from dao_ai.config import AppConfig
-from dao_ai.graph import create_dao_ai_graph
-from dao_ai.models import create_agent
 
 mlflow.langchain.autolog()
 
@@ -20,8 +17,6 @@ log_level: str = config.app.log_level
 logger.remove()
 logger.add(sys.stderr, level=log_level)
 
-graph: CompiledStateGraph = create_dao_ai_graph(config=config)
-
-app: ChatModel = create_agent(graph)
+app: ChatModel = config.as_chat_model()
 
 mlflow.models.set_model(app)

{dao_ai-0.0.12 → dao_ai-0.0.14}/src/dao_ai/config.py

@@ -351,12 +351,85 @@ class IndexModel(BaseModel, HasFullName, IsDatabricksResource):
         )
 
 
+class GenieRoomModel(BaseModel, IsDatabricksResource):
+    model_config = ConfigDict(use_enum_values=True, extra="forbid")
+    name: str
+    description: Optional[str] = None
+    space_id: str
+
+    @property
+    def api_scopes(self) -> Sequence[str]:
+        return [
+            "dashboards.genie",
+        ]
+
+    def as_resource(self) -> DatabricksResource:
+        return DatabricksGenieSpace(
+            genie_space_id=self.space_id, on_behalf_of_user=self.on_behalf_of_user
+        )
+
+
+class VolumeModel(BaseModel, HasFullName):
+    model_config = ConfigDict(use_enum_values=True, extra="forbid")
+    schema_model: Optional[SchemaModel] = Field(default=None, alias="schema")
+    name: str
+
+    @property
+    def full_name(self) -> str:
+        if self.schema_model:
+            return f"{self.schema_model.catalog_name}.{self.schema_model.schema_name}.{self.name}"
+        return self.name
+
+    def create(self, w: WorkspaceClient | None = None) -> None:
+        from dao_ai.providers.base import ServiceProvider
+        from dao_ai.providers.databricks import DatabricksProvider
+
+        provider: ServiceProvider = DatabricksProvider(w=w)
+        provider.create_volume(self)
+
+
+class VolumePathModel(BaseModel, HasFullName):
+    model_config = ConfigDict(use_enum_values=True, extra="forbid")
+    volume: Optional[VolumeModel] = None
+    path: Optional[str] = None
+
+    @model_validator(mode="after")
+    def validate_path_or_volume(self):
+        if not self.volume and not self.path:
+            raise ValueError("Either 'volume' or 'path' must be provided")
+        return self
+
+    @property
+    def full_name(self) -> str:
+        if self.volume and self.volume.schema_model:
+            catalog_name: str = self.volume.schema_model.catalog_name
+            schema_name: str = self.volume.schema_model.schema_name
+            volume_name: str = self.volume.name
+            path = f"/{self.path}" if self.path else ""
+            return f"/Volumes/{catalog_name}/{schema_name}/{volume_name}{path}"
+        return self.path
+
+    def as_path(self) -> Path:
+        return Path(self.full_name)
+
+    def create(self, w: WorkspaceClient | None = None) -> None:
+        from dao_ai.providers.databricks import DatabricksProvider
+
+        if self.volume:
+            self.volume.create(w=w)
+
+        provider: DatabricksProvider = DatabricksProvider(w=w)
+        provider.create_path(self)
+
+
 class VectorStoreModel(BaseModel, IsDatabricksResource):
     model_config = ConfigDict(use_enum_values=True, extra="forbid")
     embedding_model: Optional[LLMModel] = None
     index: Optional[IndexModel] = None
     endpoint: Optional[VectorSearchEndpoint] = None
     source_table: TableModel
+    source_path: Optional[VolumePathModel] = None
+    checkpoint_path: Optional[VolumePathModel] = None
     primary_key: Optional[str] = None
     columns: Optional[list[str]] = Field(default_factory=list)
     doc_uri: Optional[str] = None
@@ -448,43 +521,6 @@ class VectorStoreModel(BaseModel, IsDatabricksResource):
         provider.create_vector_store(self)
 
 
-class GenieRoomModel(BaseModel, IsDatabricksResource):
-    model_config = ConfigDict(use_enum_values=True, extra="forbid")
-    name: str
-    description: Optional[str] = None
-    space_id: str
-
-    @property
-    def api_scopes(self) -> Sequence[str]:
-        return [
-            "dashboards.genie",
-        ]
-
-    def as_resource(self) -> DatabricksResource:
-        return DatabricksGenieSpace(
-            genie_space_id=self.space_id, on_behalf_of_user=self.on_behalf_of_user
-        )
-
-
-class VolumeModel(BaseModel, HasFullName):
-    model_config = ConfigDict(use_enum_values=True, extra="forbid")
-    schema_model: Optional[SchemaModel] = Field(default=None, alias="schema")
-    name: str
-
-    @property
-    def full_name(self) -> str:
-        if self.schema_model:
-            return f"{self.schema_model.catalog_name}.{self.schema_model.schema_name}.{self.name}"
-        return self.name
-
-    def create(self, w: WorkspaceClient | None = None) -> None:
-        from dao_ai.providers.base import ServiceProvider
-        from dao_ai.providers.databricks import DatabricksProvider
-
-        provider: ServiceProvider = DatabricksProvider(w=w)
-        provider.create_volume(self)
-
-
 class FunctionModel(BaseModel, HasFullName, IsDatabricksResource):
     model_config = ConfigDict()
     schema_model: Optional[SchemaModel] = Field(default=None, alias="schema")
@@ -1093,6 +1129,8 @@ class EvaluationModel(BaseModel):
     model: LLMModel
     table: TableModel
     num_evals: int
+    agent_description: Optional[str] = None
+    question_guidelines: Optional[str] = None
     custom_inputs: dict[str, Any] = Field(default_factory=dict)
     guidelines: list[GuidelineModel] = Field(default_factory=list)
 
@@ -1107,21 +1145,6 @@ class DatasetFormat(str, Enum):
     EXCEL = "excel"
 
 
-class VolumePathModel(BaseModel, HasFullName):
-    model_config = ConfigDict(use_enum_values=True, extra="forbid")
-    volume: VolumeModel
-    path: str
-
-    @property
-    def full_name(self) -> str:
-        if self.volume.schema_model:
-            catalog_name: str = self.volume.schema_model.catalog_name
-            schema_name: str = self.volume.schema_model.schema_name
-            volume_name: str = self.volume.name
-            return f"/Volumes/{catalog_name}/{schema_name}/{volume_name}/{self.path}"
-        return f"/Volumes/{volume_name}/{self.path}"
-
-
 class DatasetModel(BaseModel):
     model_config = ConfigDict(use_enum_values=True, extra="forbid")
     table: Optional[TableModel] = None
@@ -1211,6 +1234,10 @@ class AppConfig(BaseModel):
     def initialize(self) -> None:
         from dao_ai.hooks.core import create_hooks
 
+        if self.app and self.app.log_level:
+            logger.remove()
+            logger.add(sys.stderr, level=self.app.log_level)
+
         logger.debug("Calling initialization hooks...")
         initialization_functions: Sequence[Callable[..., Any]] = create_hooks(
             self.app.initialization_hooks

{dao_ai-0.0.12 → dao_ai-0.0.14}/src/dao_ai/providers/databricks.py

@@ -54,6 +54,7 @@ from dao_ai.config import (
     UnityCatalogFunctionSqlModel,
     VectorStoreModel,
     VolumeModel,
+    VolumePathModel,
     WarehouseModel,
 )
 from dao_ai.models import get_latest_model_version
@@ -441,6 +442,12 @@ class DatabricksProvider(ServiceProvider):
             )
         return volume_info
 
+    def create_path(self, volume_path: VolumePathModel) -> Path:
+        path: Path = volume_path.full_name
+        logger.info(f"Creating volume path: {path}")
+        self.w.files.create_directory(path)
+        return path
+
     def create_dataset(self, dataset: DatasetModel) -> None:
         current_dir: Path = "file:///" / Path.cwd().relative_to("/")
 

{dao_ai-0.0.12 → dao_ai-0.0.14}/src/dao_ai/tools/vector_search.py

@@ -51,7 +51,9 @@ def create_vector_search_tool(
 
     vector_search_tool: BaseTool = VectorSearchRetrieverTool(
         name=name,
+        tool_name=name,
         description=description,
+        tool_description=description,
         index_name=index_name,
         columns=columns,
         **search_parameters,