Policy Q&A Agent · Use case
How can an airline or a hotel group use an AI agent to answer staff questions about fare rules and policies on its own infrastructure?
A policy Q&A agent answers a colleague's question from the company's own fare rules, conditions of carriage, cancellation, baggage and rate policies, cites the clause it used, and says when the policies are silent. The colleague reads the clause and answers the guest or the passenger. The agent reads no booking, grants no exception, changes nothing and sends nothing to a customer.
The answer is in the rules.
Finding the clause takes the time.
- ✗ 01Fare rules, conditions of carriage, rate terms and policies run to hundreds of pages, change often, and differ by fare, by rate and by route or property.
- ✗ 02A service agent answers with a guest or a passenger waiting, and the company is held to what that customer is told, so an answer from memory is a risk.
Where it stops for a person
Always. It informs, and any exception goes to a supervisor.Always. It informs a colleague, it does not decide for a customer: it applies no rule to a real booking, waives no charge and grants no exception. What a guest or a passenger is told is said by a person, who has read the clause, and an exception is a supervisor's decision. The policy, not the answer, is what the company relies on.
It answers with the clause cited.
A supervisor decides any exception.
Step · 01
A service agent, a reservations agent or a front desk colleague asks a question in plain words.
Step · 02
The agent searches the indexed policies the colleague is allowed to read.
Step · 03
It answers from the text, with the document, the clause and the date of the version it used.
Step · 04
Where the policies are silent, out of date or disagree with each other, it says so and does not guess.
Step · 05
The colleague reads the clause and answers the customer, and takes any exception to a supervisor.
Reads from · connectors
Fare rules, conditions of carriage, cancellation, baggage and rate policies, indexed as a knowledge source with each reader's permissions kept. No booking, no guest record and no operational system.
Acts through · MCP servers
Nothing. It has no tool that changes anything, and it writes to no system.
What you set up
- Index the current fare rules, conditions of carriage and policies as a knowledge source, and remove the versions no longer in force.
- Set which teams may read which documents.
- Build the agent in Katonic Studio and test it on questions your supervisors have already answered.
- Publish it to one contact centre team or one property first.
Where it runs
The agent, its knowledge store and the model run on infrastructure you own or control. You choose the model and serve it yourself, so questions, documents and answers are not sent to a public AI service. Fare rules, negotiated rate terms and internal policies stay inside the company.
How you get it today
This page is labelled Use case: the agent is built in Katonic Studio. You describe the job, connect what is named above, add test questions and release the agent once it passes them. When the Catalog holds a tested template for this job, this page will carry the Template label and show how to take it.
Does it answer guests or passengers directly?
No. It answers colleagues. A person reads the clause and answers the customer.
What does it do when the policies do not cover the question?
It says so, and says which documents it searched. It does not guess, and the question goes to a supervisor.
Can it grant an exception or waive a charge?
No. It has no tool that changes a booking or a charge. An exception is a supervisor's decision.