Business Insights AI Agent using CDGC Metadata Discovery > Using the Business Insights AI Agent using CDGC Metadata Discovery recipe > Step 4. Configure the agent flow
  

Step 4. Configure the agent flow

To configure the agent flow, configure its variables and deployment properties, then save your changes. After you configure the agent flow, you can deploy and test it.
  1. 1Open the AI Agent for Business Insights agent flow.
  2. 2Configure the agent flow variables. To do this, click Variables on the tool bar and enter values for the following agent flow variables:
  3. Variable
    Type
    Description
    Default Value
    include_relationships
    String
    Flag indicating whether the agent should retrieve metadata about relationships between tables (True) or not (False).
    False
    results_limit
    Integer
    Maximum number of rows that the SQLGenerationAgent will return by including a LIMIT clause in the SQL query.
    100
    catalog_source
    String
    CDGC catalog source used to retrieve metadata.
    -
    retry_number
    Integer
    Maximum number of times the agent will retry generating and validating the SQL query before stopping.
    3
    database_type
    String
    Type of database, for example, Snowflake, used to determine the correct SQL syntax and execution environment.
    Snowflake
    api_url
    String
    URL of the IDMC API gateway to retrieve metadata and interact with the CDGC service.
    Use the following URL:
    https://idmc-api.<cloud provider>-<region>.informaticacloud.com
    -
    host_name_login
    String
    Base URL of the POD to log in to the IDMC organization.
    Use the following URL:
    https://<cloud provider>-<region>.informaticacloud.com
    -
  4. 3Configure the deployment properties.
  5. For more information, see Deployment properties in the AI Agent Engineering guide.
  6. 4Validate and save the agent flow.
For more information about deploying and testing an agent flow, see "Deploy the agent" and "Test the agent" in the AI Agent Engineering guide.