🏔️ Summit 2026 Feature
Let AI chat bots and agents answer dynamic questions with approved query results.
🏗️ Provisional document
This feature is pending release, and this document may change over time. Check What’s New for the latest releases.
AI Chat Bot Query Knowledge extends Slate AI Query Knowledge to AI-enabled chat bots and agents. Instead of relying only on crawled web pages or static instructions, a bot can consult a selected query knowledge source when the conversation requires current structured data.
Use query knowledge when a bot needs to answer questions from data that changes often, such as upcoming visits, active events, available programs, or office-specific service options.
How query knowledge works in bot conversations
A query knowledge source starts with a Slate query. The query controls the available rows, exports, and filters. The source context tells Slate AI how to interpret those rows.
After the source is selected in an AI chat bot or agent, the bot can use the query when its instructions or the conversation indicate that the data is relevant. In the campus visit example, the source returns active upcoming form events. The bot uses exports such as form URL, date and time, location, and title to suggest a registration option.
An admissions bot might use an upcoming-events query to answer questions about campus visits. The query returns only confirmed future events and includes a registration URL, date and time, city, location, and event title.
A student success bot might use a services query to help students find available advising sessions, workshops, or support appointments. The query can keep answers focused on currently active options instead of outdated schedule information.
An advancement bot might use an events query to help alumni find upcoming receptions, reunions, or campaign events. The bot can answer with the event name, date, location, and registration link returned by the query.
Preparing the query knowledge source
Before adding query knowledge to a bot, create a query knowledge source in Database → Bots. For the full setup pattern, see Slate AI Query Knowledge.
Keep the query focused on the data the bot should use in conversation. A focused source gives the bot clearer context and limits the response to the intended information.
Exports: Include the fields the bot should be able to cite, summarize, or link to in its response.
Filters: Limit rows to the records that are relevant for live answers, such as active records or future-dated events.
Context: Explain what the source represents and how important exports should be used.
Adding query knowledge to an AI-enabled bot
Go to Database → Bots.
Open the AI chat bot or agent that should use query-based knowledge.
Select Edit.
In Knowledge Sources, select the query knowledge source.
Update System Instructions to explain when the bot should query the source.
Select Save.
Use Preview Bot to test a question that should require the query knowledge source.
Related articles
📖 Agents