Student Policy Q&A Agent · Use case
How can a university use an AI agent to answer staff questions about its regulations and policies on its own infrastructure?
A student policy Q&A agent answers a staff member's question about the institution's regulations and policies from the documents themselves, cites the clause it relied on, and says when the documents are silent. It is used by staff, not by students. Any exception goes to the policy owner. The agent reads no student record, decides nothing about a student and sends nothing to one.
The answer is in the regulations.
Finding the clause takes the time.
- ✗ 01Regulations, policies and programme handbooks run to hundreds of pages, are revised each year, and are read under time pressure by staff with a case in front of them.
- ✗ 02The same question gets different answers in different offices, and an answer given from memory can be a year out of date.
Where it stops for a person
When the documents are silent, and at any exception. The policy owner decides.It informs. It grants no exception, rules on no individual case and decides nothing about a student. A question the documents do not answer goes to the policy owner, and a member of staff, not the agent, speaks to the student.
It answers with the clause.
The policy owner decides any exception.
Step · 01
A member of staff asks a question in their own words.
Step · 02
The agent searches the regulations, policies and handbooks that person is allowed to read.
Step · 03
It answers from the passages it found and cites each clause, with the document and its version.
Step · 04
Where the documents are silent or disagree, it says so and names the policy owner.
Step · 05
The member of staff reads the clause and acts on it. Any exception goes to the policy owner.
Reads from · connectors
The regulations, policies, programme handbooks and procedures you index as a knowledge source, with each reader's permissions kept. No student 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- Index the current regulations, policies and handbooks as a knowledge source, and take out superseded versions.
- Give the agent the owner of each policy, so it can say who decides an exception.
- Build the agent in Katonic Studio and test it on questions your registry has already answered.
- Publish it to one office 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. Draft policies and internal procedures stay inside the institution.
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.
Is it for students?
No. It is built for staff. A member of staff reads the answer and speaks to the student.
What does it do when the regulations are silent?
It says so and names the policy owner. It does not guess.
Does it read student records?
No. It reads the documents you index. It has no tool that reaches a student record.