Summary
In this module, you learned how to build a Fabric data agent that uses an ontology as its data source. The ontology defines entity types — Departments, Hospitals, Patients, Rooms, and VitalSignEquipment — and the relationships between them, giving the agent the context it needs to translate natural language into GQL queries. You connected an ontology as a data source, wrote agent instructions to guide interpretation, and used the steps dropdown to validate how the agent reasons through a question. You also published the agent and shared it with colleagues.