Dot in AI Database Agents

Dot can find a complained-about exchange from a rough time and the member who asked it, then open it for inspection. It locates the chat in History by user and time, narrows it with a membership search, and shows the question and reply.

Dot did combine a headline figure with its breakdown. It gave an August sales sentence with a monthly chart, then a contribution chart and driver sentences; Furniture was the main driver.

Dot updated its answer when the customer corrected July to June. The corrected reply gave 11 June orders and the query limited the count to June; the earlier July turn was 18 against a reference of 19 and was not the graded one.

Dot passed this scenario: it refused to answer because the returns table was turned off, and it did not use the table as the answer. The logs show only a search step, with no data fetch or SQL, and the reply pointed to Model for the switch.

Dot assembled revenue, category, and cancellation into one report. Revenue and category matched earlier chat answers; the cancellation figure came after a clarification and matched the PDF.

Dot kept the answer available for reopening. Saving it as an app and reopening it in a new session showed the same title, date range, and June/July/August values, and the saved app re-ran the same query.