Skip to content
AI Security

Interview question · Agent, Tool & MCP Security

In an agent, where may the model reason, and where must deterministic code own the decision?

Separate model judgment from enforceable invariants, then inspect the inputs and state transitions used by the policy code. Deterministic code is useful only when its authority and facts are trustworthy.

mediumNewUpdated Nov 2026

Separate model judgment from enforceable invariants, then inspect the inputs and state transitions used by the policy code. Deterministic code is useful only when its authority and facts are trustworthy.

Make your first attempt

What would you say before reading the answer?

  1. 01 State the assumptions your answer depends on.
  2. 02 Explain the mechanism and choose a control.
  3. 03 Name a condition that would change your decision.

The worked answer includes an approach, scoring criteria and follow-up questions. Compare the reasoning with your own.

10 answers per topic instead of 5, plus saved progress and bookmarks · costs nothingor unlock all 529 remaining answers · $25

Every worked answer in all 17 topics, every lesson in every course, the full concept curriculum, the complete practice-test bank, and the downloadable security field guides.

6 months · One payment · No auto-renewal

What if · the follow-ups in this answer

The answer works through 4 follow-ups like an interviewer would

Alder’s refund gateway checks a model-generated eligibility flag, an amount and a customer ID. Every comparison is deterministic.

  1. Does deterministic evaluation establish a secure decision?
  2. Can two requests both pass a correct balance check?
  3. Must every judgment call be handed to a person?
  4. How would you verify the boundary?

Each rung has a nudge, a model answer, the principle that carries over and the weak answer to avoid. Premium opens every answer and ladder.

UP NEXT ON YOUR JOURNEY
Next in this trackHow do rate limits, budgets and loop limits reduce abuse of an AI agent, and how do you set them?Next in this trackDesign and test an agent’s file-write boundary, including symlinks and concurrent changes.Next in this trackAn AI agent caused financial damage in production. What do you do in the first hour?
DISCUSSION · 0

No comments yet. Share a correction or a different way to answer it.