AI-103: Developing AI Apps and Agents on Azure

A Microsoft Foundry project connection stores an authentication detail such as an API key or endpoint once at the project level, and any agent or tool inside that project can reference the same connection instead of each agent holding a separate copy of the Azure AI Search credential. Adding a connection to the Azure AI Search resource is therefore the mechanism that centralizes credential storage and management and makes it reusable across every agent in Project1. Enabling RBAC and disabling key-based access are good complementary hardening steps but do not by themselves centralize credential management, and a managed private endpoint addresses network-level connectivity to the resource, not how agents obtain or share the credential.