dao-ai 0.0.26__tar.gz → 0.0.27__tar.gz

{dao_ai-0.0.26 → dao_ai-0.0.27}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dao-ai
-Version: 0.0.26
+Version: 0.0.27
 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
@@ -41,7 +41,7 @@ Requires-Dist: langgraph>=0.6.10
 Requires-Dist: langmem>=0.0.29
 Requires-Dist: loguru>=0.7.3
 Requires-Dist: mcp>=1.17.0
-Requires-Dist: mlflow>=3.5.0
+Requires-Dist: mlflow>=3.5.1
 Requires-Dist: nest-asyncio>=1.6.0
 Requires-Dist: openevals>=0.0.19
 Requires-Dist: openpyxl>=3.1.5

{dao_ai-0.0.26 → dao_ai-0.0.27}/config/examples/genie.yaml

@@ -2,11 +2,8 @@
 
 
 schemas:
-  retail_schema: &retail_schema
-    catalog_name: nfleming                    # Unity Catalog name
-    schema_name: retail_ai                    # Schema within the catalog
 
-  quick_serve_retaurant_schema: &quick_serve_retaurant_schema
+  quick_serve_restaurant_schema: &quick_serve_restaurant_schema
     catalog_name: retail_consumer_goods                    # Unity Catalog name
     schema_name: quick_serve_restaurant                    # Schema within the catalog
 
@@ -14,45 +11,58 @@ resources:
   llms:
     # Primary LLM for general tasks
     default_llm: &default_llm
-      name: databricks-meta-llama-3-3-70b-instruct  # Databricks serving endpoint name
+      name: databricks-claude-sonnet-4
       temperature: 0.1                              # Low temperature for consistent responses
       max_tokens: 8192                              # Maximum tokens per response
       on_behalf_of_user: False
 
+  warehouses:
+
+    shared_endpoint_warehouse: 
+      name: "Shared Endpoint Warehouse"             # Human-readable name
+      description: "A warehouse for shared endpoints"  # Description
+      warehouse_id: 148ccb90800933a1                # Databricks warehouse ID
+      on_behalf_of_user: False
+
   tables:
-    product_table: &product_table
-      schema: *quick_serve_retaurant_schema                        # Reference to schema defined above
-      name: products                                # Table name
+
+    all:
+      schema: *quick_serve_restaurant_schema                        # Reference to schema defined above
+    #   name: products         
+
+    # product_table: &product_table
+    #   schema: *quick_serve_restaurant_schema                        # Reference to schema defined above
+    #   name: products                                # Table name
     
-    inventory_table: &inventory_table
-      schema: *quick_serve_retaurant_schema
-      name: inventory
+    # inventory_table: &inventory_table
+    #   schema: *quick_serve_restaurant_schema
+    #   name: inventory
 
-    items_description: &items_description
-      schema: *quick_serve_retaurant_schema
-      name: items_description
+    # items_description: &items_description
+    #   schema: *quick_serve_restaurant_schema
+    #   name: items_description
 
-    fulfil_item_orders: &fulfil_item_orders
-      schema: *quick_serve_retaurant_schema
-      name: fulfil_item_orders
+    # fulfil_item_orders: &fulfil_item_orders
+    #   schema: *quick_serve_restaurant_schema
+    #   name: fulfil_item_orders
 
-    orders: &orders
-      schema: *quick_serve_retaurant_schema
-      name: orders
+    # orders: &orders
+    #   schema: *quick_serve_restaurant_schema
+    #   name: orders
 
   functions:
 
-    lookup_items_by_descriptions: &lookup_items_by_descriptions
-      schema: *quick_serve_retaurant_schema
-      name: lookup_items_by_descriptions
+    all: 
+      schema: *quick_serve_restaurant_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_item_by_description_and_price: &match_item_by_description_and_price
+    #   schema: *quick_serve_restaurant_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     
+    # match_historical_item_order_by_date: &match_historical_item_order_by_date
+    #   schema: *quick_serve_restaurant_schema
+    #   name: match_historical_item_order_by_date     
 
   genie_rooms:
     # Genie space for retail data queries
@@ -86,13 +96,13 @@ agents:
 
 
 app:
-  name: genie_example                      # Application name  
+  name: dao_genie                      # Application name  
   description: "Multi-agent system that talks to genie"
-  log_level: INFO                                   # Logging level for the application
+  log_level: DEBUG                                   # Logging level for the application
   registered_model:                                 # MLflow registered model configuration
-    schema: *retail_schema                          # Schema where model will be registered
-    name: retail_ai_agenie                          # Model name in MLflow registry
-  endpoint_name: retail_ai_genie                    # Model serving endpoint name
+    schema: *quick_serve_restaurant_schema                          # Schema where model will be registered
+    name: dao_genie                          # Model name in MLflow registry
+  endpoint_name: dao_genie                    # Model serving endpoint name
   tags:                                             # Tags for resource organization
     business: rcg                                   # Business unit identifier
     streaming: true                                 # Indicates streaming capabilities

dao_ai-0.0.27/config/examples/genie_and_genie_mcp.yaml

@@ -0,0 +1,185 @@
+# yaml-language-server: $schema=../../schemas/model_config_schema.json
+
+variables:
+  client_id: &client_id                          # Service principal client ID
+    options:
+      - env: RETAIL_AI_DATABRICKS_CLIENT_ID
+      - scope: retail_consumer_goods
+        secret: RETAIL_AI_DATABRICKS_CLIENT_ID
+  client_secret: &client_secret                  # Service principal secret
+    options:
+      - env: RETAIL_AI_DATABRICKS_CLIENT_SECRET
+      - scope: retail_consumer_goods
+        secret: RETAIL_AI_DATABRICKS_CLIENT_SECRET
+  workspace_host: &workspace_host                # Databricks workspace URL
+    options:
+      - env: RETAIL_AI_DATABRICKS_HOST
+      - scope: retail_consumer_goods
+        secret: RETAIL_AI_DATABRICKS_HOST
+
+schemas:
+
+  store_ops_schema: &store_ops_schema
+    catalog_name: retail_consumer_goods                    # Unity Catalog name
+    schema_name: store_ops                    # Schema within the catalog
+
+resources:
+  llms:
+    
+    fast_llm: &fast_llm
+      name: databricks-meta-llama-3-3-70b-instruct
+      on_behalf_of_user: False
+
+    tool_calling_llm: &tool_calling_llm
+      name: databricks-claude-sonnet-4
+      on_behalf_of_user: False
+
+  warehouses:
+
+    shared_endpoint_warehouse: 
+      name: "Shared Endpoint Warehouse"             # Human-readable name
+      description: "A warehouse for shared endpoints"  # Description
+      warehouse_id: 148ccb90800933a1                # Databricks warehouse ID
+      on_behalf_of_user: False
+
+  tables:
+    all:
+      schema: *store_ops_schema                      
+
+
+  functions:
+    all: 
+      schema: *store_ops_schema
+
+
+  genie_rooms:
+    # Genie space for retail data queries
+    retail_genie_room: &retail_genie_room
+      name: "Retail AI Genie Room"                        # Human-readable name
+      description: "A room for Genie agents to interact"  # Description
+      space_id: 01f05dd06c421ad6b522bf7a517cf6d2          # Databricks Genie space ID
+
+prompts:
+  # ---------------------------------------------------------------------------
+  # Employee Agent Prompt
+  # ---------------------------------------------------------------------------
+  employee_agent_prompt: &employee_agent_prompt
+    schema: *store_ops_schema
+    name: employee_prompt
+    description: "The employee management agent prompt for retail operations"
+    default_template: |
+      You are an employee management specialist for retail operations.
+      Your role is to help answer questions about employees, their performance,
+      tasks, and daily activities using the available data sources.
+      
+      You have access to the following data through your tools:
+      - **employee_daily_tasks**: Track daily tasks assigned to employees
+      - **employee_performance**: Monitor employee performance metrics and evaluations
+      - **employee_tasks**: Manage task assignments and completion status
+      
+      When responding:
+      - Be specific when referencing employee data and metrics
+      - Use the genie tool to query the employee databases for current information
+      - Provide insights on performance trends and task completion rates
+      - Respect employee privacy while providing helpful analytics
+      - When discussing tasks, always check current status before responding
+      - Provide actionable recommendations for task management and performance improvement
+      
+      Available tables in retail_consumer_goods.store_ops schema:
+      - employee_daily_tasks
+      - employee_performance  
+      - employee_tasks
+    tags:
+      environment: development
+      agent_type: employee_agent
+
+  # ---------------------------------------------------------------------------
+  # Inventory Agent Prompt
+  # ---------------------------------------------------------------------------
+  inventory_agent_prompt: &inventory_agent_prompt
+    schema: *store_ops_schema
+    name: inventory_prompt
+    description: "The inventory agent prompt for retail operations"
+    default_template: |
+      You are an inventory management specialist for retail operations.
+      Your role is to help answer questions about product inventory, stock levels,
+      and inventory-related queries using the available data sources.
+      
+      When responding:
+      - Be precise with inventory numbers and product details
+      - Always verify information before providing answers
+      - If you need current data, use the genie tool to query the inventory database
+      - Provide actionable insights when discussing stock levels
+    tags:
+      environment: development
+      agent_type: inventory_agent
+
+tools:
+  genie_tool: &genie_tool
+    name: genie_tool
+    function:
+      type: factory                                 # Tool type: factory function
+      name: dao_ai.tools.create_genie_tool       # Factory function path
+      args:                                         # Arguments passed to factory
+        name: genie_tool
+        description: Answers questions about retail products and inventory
+        genie_room: *retail_genie_room              # Reference to Genie room config
+
+  genie_mcp: &genie_mcp                          # Genie MCP tool configuration
+    name: genie_mcp                                  
+    function:
+      type: mcp                                  # MCP function type
+      name: genie_mcp                              
+      genie_room: *retail_genie_room             # Use genie_room convenience object
+      client_id: *client_id
+      client_secret: *client_secret
+      workspace_host: *workspace_host
+
+
+agents:
+  employee_agent: &employee_agent
+    name: employee_agent                                   
+    description: "Employee Agent"
+    model: *tool_calling_llm                          
+    tools:                                      
+      - *genie_tool
+    prompt: *employee_agent_prompt
+    handoff_prompt: |
+      Questions about employees, employee performance, task assignments, daily activities, or workforce management.
+      Examples: "What tasks are assigned to John today?", "Show me employee performance metrics", "How is Sarah performing?", "What are the pending employee tasks?", "Who completed the most tasks this week?"
+
+  inventory_agent: &inventory_agent
+    name: inventory_agent
+    description: "Inventory Management Agent - handles inventory queries via Genie MCP"
+    model: *tool_calling_llm
+    tools:
+      - *genie_mcp                            # Uses Genie MCP server tool
+    prompt: *inventory_agent_prompt           # Uses external prompt from registry
+    handoff_prompt: |
+      Questions about inventory, stock levels, product availability, or inventory management.
+      Examples: "What's the current inventory level?", "Do we have product X in stock?", "Show me low stock items", "What products need reordering?", "What's the inventory status for store 123?"
+      
+
+app:
+  name: dao_pepsi_genie_demo                      # Application name  
+  description: "Multi-agent system that talks to genie"
+  log_level: DEBUG                                   # Logging level for the application
+  registered_model:                                 # MLflow registered model configuration
+    schema: *store_ops_schema                          # Schema where model will be registered
+    name: dao_pepsi_genie_demo                          # Model name in MLflow registry
+  endpoint_name: dao_pepsi_genie_demo                    # Model serving endpoint name
+  tags:                                             # Tags for resource organization
+    business: rcg                                   # Business unit identifier
+    streaming: true                                 # Indicates streaming capabilities
+  permissions:                                      # Model serving permissions
+    - principals: [users]                           # Grant access to all users
+      entitlements:
+        - CAN_QUERY                                # Full management permissions
+  agents:                                           # List of agents included in the system
+    - *employee_agent                               # Employee management agent
+    - *inventory_agent                              # Inventory management agent
+  orchestration:                                    # Agent orchestration configuration
+    supervisor:                                     # Supervisor orchestration pattern
+      model: *fast_llm                           # LLM for routing decisions
+
+

{dao_ai-0.0.26 → dao_ai-0.0.27}/config/examples/mcp.yaml

@@ -19,31 +19,74 @@ variables:
 
 schemas:
   retail_schema: &retail_schema                  # Unity Catalog schema configuration
-    catalog_name: nfleming                       # Unity Catalog name
-    schema_name: retail_ai                       # Schema within the catalog
+    catalog_name: retail_consumer_goods                       # Unity Catalog name
+    schema_name: hardware_store                       # Schema within the catalog
 
 resources:
   llms:
     default_llm: &default_llm                    # Default LLM configuration
       name: databricks-claude-3-7-sonnet        # Databricks serving endpoint name
 
+  genie_rooms:
+    retail_genie_room: &retail_genie_room        # Genie room configuration
+      name: "Retail AI Genie Room"               # Human-readable name
+      description: "Genie room for retail queries"
+      space_id: 01f01c91f1f414d59daaefd2b7ec82ea  # Databricks Genie space ID
+
+  tables:
+    product_table: &product_table                # Table configuration for vector store
+      schema: *retail_schema
+      name: products
+
+  vector_stores:
+    retail_vector_store: &retail_vector_store    # Vector store configuration
+      source_table: *product_table               # Source table for vector search
+      embedding_source_column: description       # Column containing embeddings
+      endpoint: 
+        name: dbdemos_vs_endpoint
+      index:
+        schema: *retail_schema                   # Schema for vector search index
+        name: products_index                      # Vector search index name
+
 tools:
+  # Example 1: Using direct URL (generic approach - works with any MCP server)
+  generic_mcp: &generic_mcp                      # Generic MCP tool with direct URL
+    name: generic_mcp
+    function:
+      type: mcp                                  # MCP function type
+      name: generic_mcp
+      url: https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/vector-search/retail_consumer_goods/hardware_store
+      client_id: *client_id
+      client_secret: *client_secret
+      workspace_host: *workspace_host
+
+  # Example 2: Using convenience object (recommended - automatic URL generation)
   vector_search_mcp: &vector_search_mcp          # Vector search MCP tool configuration
     name: vector_search_mcp                                  
     function:
       type: mcp                                  # MCP function type
       name: vector_search_mcp                              
-      url: https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/vector-search/nfleming/retail_ai
+      vector_search: *retail_vector_store        # Use vector search convenience object
+      client_id: *client_id
+      client_secret: *client_secret
+      workspace_host: *workspace_host
+
+  sql_mcp: &sql_mcp                              # DBSQL MCP tool configuration
+    name: sql_mcp
+    function:
+      type: mcp                                  # MCP function type
+      name: sql_mcp
+      sql: true                                  # Use serverless DBSQL MCP server (no warehouse needed)
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
 
-  unity_catalog_mcp: &unity_catalog_mcp          # Unity Catalog MCP tool configuration
-    name: unity_catalog_mcp                                  
+  functions_mcp: &functions_mcp                  # Unity Catalog Functions MCP tool configuration
+    name: functions_mcp                                  
     function:
       type: mcp                                  # MCP function type
-      name: unity_catalog_mcp                              
-      url:  https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/functions/nfleming/retail_ai
+      name: functions_mcp                              
+      functions: *retail_schema                  # Use functions (UC Functions) convenience object
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
@@ -53,7 +96,7 @@ tools:
     function:
       type: mcp                                  # MCP function type
       name: genie_mcp                              
-      url:  https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/genie/01f01c91f1f414d59daaefd2b7ec82ea
+      genie_room: *retail_genie_room             # Use genie_room convenience object
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
@@ -63,13 +106,17 @@ agents:
     name: mcp_agent                              # Agent identifier
     model: *default_llm                          # Reference to LLM configuration
     tools:                                       # Tools available to this agent
-      - *vector_search_mcp
-      - *unity_catalog_mcp
+      # Note: generic_mcp and vector_search_mcp both point to vector search
+      # Only one should be used in practice (shown here for demonstration)
+      - *generic_mcp                             # Example: Direct URL approach
+      - *vector_search_mcp                       # Example: Convenience object approach
+      - *sql_mcp
+      - *functions_mcp
       - *genie_mcp
     prompt: |                                    # System prompt defining agent behavior
-      You are a multi-purpose agent that can perform vector search, interact with Unity Catalog, and
-      interact with Genie rooms. Use the tools provided to answer questions and perform tasks. 
-      You many need to use the tools in combination to achieve complex tasks.
+      You are a multi-purpose agent that can perform vector search, execute SQL queries via DBSQL,
+      call Unity Catalog functions, and interact with Genie rooms. Use the tools provided to answer
+      questions and perform tasks. You may need to use the tools in combination to achieve complex tasks.
       
 app:
   name: mcp_agent                                # Application name  

{dao_ai-0.0.26 → dao_ai-0.0.27}/config/examples/mcp_with_uc_connection.yaml

@@ -21,6 +21,10 @@ schemas:
   retail_schema: &retail_schema                  # Unity Catalog schema configuration
     catalog_name: nfleming                       # Unity Catalog name
     schema_name: retail_ai                       # Schema within the catalog
+  
+  hardware_store_schema: &hardware_store_schema  # Hardware store schema configuration
+    catalog_name: retail_consumer_goods          # Unity Catalog name
+    schema_name: hardware_store                  # Schema within the catalog
 
 resources:
   llms:
@@ -29,36 +33,76 @@ resources:
   
   connections:
     github_connection: &github_connection       # UC Connection for GitHub
-      name: github_pat_connection_nfleming               # UC Connection name
+      name: github_pat_connection_nfleming      # UC Connection name
+
+  genie_rooms:
+    retail_genie_room: &retail_genie_room        # Genie room configuration
+      name: "Retail AI Genie Room"               # Human-readable name
+      description: "Genie room for retail queries"
+      space_id: 01f01c91f1f414d59daaefd2b7ec82ea  # Databricks Genie space ID
+
+  tables:
+    product_table: &product_table                # Table configuration for vector store
+      schema: *hardware_store_schema
+      name: products
+
+  vector_stores:
+    hardware_vector_store: &hardware_vector_store  # Vector store configuration
+      source_table: *product_table               # Source table for vector search
+      embedding_source_column: description       # Column containing embeddings
+      index:
+        schema: *hardware_store_schema           # Schema for vector search index
+        name: products_index                      # Vector search index name
 
 tools:
-  # Example using UC Connection for GitHub MCP with explicit URL
+  # Example 1: Using UC Connection (automatic URL generation from connection)
   github_mcp_uc: &github_mcp_uc
     name: github_mcp                                # Tool name for agent reference
     function:
       type: mcp                                     # Tool type: MCP server
       name: github_mcp                              # MCP function name
       transport: streamable_http                    # Transport protocol
-      #url: https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/external/github_u2m_connection  # MCP endpoint URL (optional with connection)
       connection: *github_connection                # Reference to UC Connection (provides OAuth auth)
+      # Note: URL is auto-generated from connection, no need to specify
 
+  # Example 2: Using direct URL (generic approach - works with any MCP server)
+  generic_mcp: &generic_mcp                      # Generic MCP tool with direct URL
+    name: generic_mcp
+    function:
+      type: mcp                                  # MCP function type
+      name: generic_mcp
+      url: https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/vector-search/retail_consumer_goods/hardware_store
+      client_id: *client_id
+      client_secret: *client_secret
+      workspace_host: *workspace_host
       
+  # Example 3: Using convenience object (recommended - automatic URL generation)
   vector_search_mcp: &vector_search_mcp          # Vector search MCP tool configuration
     name: vector_search_mcp                                  
     function:
       type: mcp                                  # MCP function type
       name: vector_search_mcp                              
-      url: https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/vector-search/retail_consumer_goods/hardware_store
+      vector_search: *hardware_vector_store      # Use vector search convenience object
+      client_id: *client_id
+      client_secret: *client_secret
+      workspace_host: *workspace_host
+
+  sql_mcp: &sql_mcp                              # DBSQL MCP tool configuration
+    name: sql_mcp
+    function:
+      type: mcp                                  # MCP function type
+      name: sql_mcp
+      sql: true                                  # Use serverless DBSQL MCP server (no warehouse needed)
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
 
-  unity_catalog_mcp: &unity_catalog_mcp          # Unity Catalog MCP tool configuration
-    name: unity_catalog_mcp                                  
+  functions_mcp: &functions_mcp                  # Unity Catalog Functions MCP tool configuration
+    name: functions_mcp                                  
     function:
       type: mcp                                  # MCP function type
-      name: unity_catalog_mcp                              
-      url:  https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/functions/retail_consumer_goods/hardware_store
+      name: functions_mcp                              
+      functions: *hardware_store_schema          # Use functions (UC Functions) convenience object
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
@@ -68,7 +112,7 @@ tools:
     function:
       type: mcp                                  # MCP function type
       name: genie_mcp                              
-      url:  https://adb-984752964297111.11.azuredatabricks.net/api/2.0/mcp/genie/01f01c91f1f414d59daaefd2b7ec82ea
+      genie_room: *retail_genie_room             # Use genie_room convenience object
       client_id: *client_id
       client_secret: *client_secret
       workspace_host: *workspace_host
@@ -80,19 +124,23 @@ prompts:
     name: mcp_agent_prompt
     description: "MCP agent prompt"
     default_template: |
-      You are a multi-purpose agent that can perform vector search, interact with Unity Catalog,
-      interact with Genie rooms, and interact with GitHub. Use the tools provided to answer 
-      questions and perform tasks. You many need to use the tools in combination to achieve 
-      complex tasks.
+      You are a multi-purpose agent that can perform vector search, execute SQL queries via DBSQL,
+      call Unity Catalog functions, interact with Genie rooms, and interact with GitHub via UC Connection.
+      Use the tools provided to answer questions and perform tasks. You may need to use the tools in
+      combination to achieve complex tasks.
 
 agents:
   mcp_agent: &mcp_agent                          # MCP agent configuration
     name: mcp_agent                              # Agent identifier
     model: *default_llm                          # Reference to LLM configuration
     tools:                                       # Tools available to this agent
-      - *github_mcp_uc
-      - *vector_search_mcp
-      - *unity_catalog_mcp
+      - *github_mcp_uc                           # Example: UC Connection approach
+      # Note: generic_mcp and vector_search_mcp both point to vector search
+      # Only one should be used in practice (shown here for demonstration)
+      - *generic_mcp                             # Example: Direct URL approach
+      - *vector_search_mcp                       # Example: Convenience object approach
+      - *sql_mcp
+      - *functions_mcp
       - *genie_mcp
     prompt: *mcp_agent_prompt