Plan and manage an Azure AI solution 1.3 Manage, monitor, and secure AI systems
A backend service principal must only call an existing agent in a Foundry project (for…
Study guide skill: Configure security, including managed identity, private networking, keyless credentials, and role policies
Question B1-007
A backend service principal must only call an existing agent in a Foundry project (for example, through the Responses API). It must not create or modify agents. Which built-in role follows least privilege?
Select one answer.
Answer: D
Foundry Agent Consumer grants access to interact with agent endpoints and is the least-privilege role for principals that only call agents. Foundry User lets developers build and test in the project, which is more than needed. Foundry Owner is highly privileged, and Azure AI Developer applies to Azure Machine Learning workspaces and hubs, not Foundry projects.
Deep dive
How it works
Foundry has a built-in role made for exactly this case: Foundry Agent Consumer. It grants access to interact with agent endpoints in a Foundry project, and the docs describe it as the least-privilege role for principals that only need to interact with agents. The RBAC article also gives a direct tip: if a user or service principal only needs to interact with agents (for example, by calling the Responses API) without creating or modifying them, assign Foundry Agent Consumer instead of Foundry User.
That matches the question: a backend service principal that calls an existing agent but must never build or change one. Least privilege means giving it the narrowest role that still lets it do its job.
A quick map of the Foundry roles you're likely to see:
- Foundry Agent Consumer: calls agent endpoints.
- Foundry User: reader access to the project and resource plus data actions; the least-privilege role for developers building and testing agents.
- Foundry Project Manager: management actions on projects.
- Foundry Owner: full access to manage projects and resources.
Note the rename: Foundry User, Foundry Owner, Foundry Account Owner and Foundry Project Manager used to be called Azure AI User, Azure AI Owner, Azure AI Account Owner and Azure AI Project Manager. The role IDs and core permissions didn't change.
The other options, and when you'd use them
- Foundry Owner: It grants full access to manage projects and resources, and the docs call it a highly privileged role. That's far more than a caller needs.
- Azure AI Developer: Despite the name, this role is scoped to Azure Machine Learning workspaces and Foundry hubs, not Foundry projects or hosted agents. The docs say not to use it for Foundry work.
- Foundry User (formerly Azure AI User): It's meant for developers building and testing agents. It works, but it isn't least privilege for a principal that only calls agents.
Exam tip
"Only calls agents, never creates them" means Foundry Agent Consumer. "Builds and tests agents" means Foundry User. If you see "Azure AI Developer" for a Foundry project, it's the trap.
References
Last reviewed: