Dashboard Search
The dashboard search agent retrieves relevant dashboards from the catalog.
How it works
Section titled “How it works”The agent follows this workflow:
- Receives a natural language question from the user
- Extracts search terms and identifies filter fields and values to search with
- Executes the search to obtain dashboards
- Summarizes the search results and includes relevant dashboards to the user’s request in the final answer
Input parameters
Section titled “Input parameters”Required:
message(string): The natural language question or request to send to the agent
Output format
Section titled “Output format”The agent produces a series of thinking, tool call, tool return, and text blocks as it works through the user request. The final message, assuming no errors, is a json object with these parameters:
message(string): The final answer to the user’s requestrelevant_objects(array): A list of BIReportObject, each containing:url(string): The URL to access the dashboardtitle(string): The name of the dashboarddescription(string): The description of the dashboardbreadcrumbs(array): The list of folder names leading to the dashboard in the catalogtype(object): The dashboard type.
Available tools
Section titled “Available tools”The agent is configured with 3 tools:
BI report search
Section titled “BI report search”Searches for dashboards in the catalog relevant to the user’s question.
Key parameters:
search_term: The main search term extracted from the user’s questionfilters: Optional filters extracted from the user’s question to narrow down the search (e.g., folder names, tags)limit: The maximum number of dashboards to return
Search filter fields
Section titled “Search filter fields”Identifies relevant fields which are available in the system to refine the dashboard search.
Key parameters:
search_term: The filter field search term extracted from the user’s questionlimit: The maximum number of filter fields to return
Search filter fields
Section titled “Search filter fields”Search over values of a specific field to identify relevant values to refine the dashboard search.
Key parameters:
field_id: The ID of the field to search values insearch_term: The filter value search term extracted from the user’s questionlimit: The maximum number of filter values to return
Behavior notes
Section titled “Behavior notes”- The agent returns a structured object as its final response, including both the final message and relevant objects