Hello,
1) What is your original issue/pain point you want to solve?
We already have several AI agent types available (ticket classifier, prioritizer, title rewriter, group dispatcher), but none of them can generate an automatic reply to the customer, on a ticket or on chat, based on a knowledge source.
2) Which are one or two concrete situations where this problem hurts the most?
- On chat, customers expect an immediate answer, including outside business hours or during high-traffic periods.
- On tickets, a large volume of recurring, simple requests (order tracking, delays, basic information) currently has to be handled manually by an agent, even though they could be resolved at first level.
3) Why is it not solvable with the Zammad standard?
Existing AI agents classify, prioritize, or route tickets, but none of them generate reply content for the customer based on an external knowledge source (FAQ-type).
4) What is your expectation/what do you want to achieve?
A new AI agent type able to:
- read a knowledge source (a question/answer file),
- search it for the most relevant answer to the customer’s request,
- automatically generate/send a first-level reply on the ticket and/or chat, with escalation to a human agent when no relevant answer is found.
Your Zammad environment:
- Average concurrent agent count: 3
- Average tickets a day: ~25/agent on average
- What roles/people are involved: customer support team
Anything else which you think is useful to understand your use case:
We’re attaching a screenshot of the AI agent creation screen (list of available types) to illustrate where this new option could fit in.
Thanks you
Dan
