A fully managed platform in Microsoft Foundry for hosting, scaling, and securing AI agents built with any supported framework or model
Hi Noopur Nanda,
This behavior is expected and not related to any permission or configuration issue.
When you create a Knowledge Base in Azure AI Foundry, the Azure AI Search indexes used in the background for storing, chunking, and retrieving data are fully managed by the Foundry service. These indexes are created and maintained internally as part of the Knowledge Base and are not exposed in the Foundry project UI or APIs.
Because of this design, you will not be able to see or access those indexes directly, even if you have provided all the required permissions on Azure AI Search and Blob Storage. This is intentional, as Foundry handles the entire indexing and retrieval process automatically to simplify the experience and avoid manual configuration.
If your use case requires direct access to indexes for example to customize the schema or run queries manually, you would need to create and manage indexes separately in Azure AI Search instead of relying on the Knowledge Base.