Represents the state of an LLM analysis. New analysis requests begin in the 'pending' state, which should then transition to either 'ready' or 'error' after the model response is received. If a user adds an additional question to the chat, the state will be 'pending-followup', followed again by 'ready' or 'error'.

pending pending_followup ready error

Represents the mode of an LLM analysis. 'host' indicates the analysis was started on a host; 'results' indicates the analysis was started on one or more results.

host results

User ID (Sandfly login) of the user who initiated the analysis.

Full name of the user who initiated the analysis.

Brief one-sentence summary of LLM analysis, if available. Will be an empty string if the first response is not yet available, or if no brief was found.

Data that can go along with the status.

Extended explanation of status.

Optional additional details about the status (typically an error) that, if present and not an empty string, should be displayed as pre-formatted, monospaced text.

Short description of status.

Data that can go along with the status.

Extended explanation of status.

Optional additional details about the status (typically an error) that, if present and not an empty string, should be displayed as pre-formatted, monospaced text.

Short description of status.