Skip to content
AI Security

Interview question · Agent, Tool & MCP Security

How can one compromised or faulty agent cascade through a system, and what circuit breakers stop it?

Separate false-information spread, borrowed authority and runaway work. Put shared limits and stop checks at the propagation paths, then recover without repeating completed effects.

hardNewUpdated Nov 2026

Separate false-information spread, borrowed authority and runaway work. Put shared limits and stop checks at the propagation paths, then recover without repeating completed effects.

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 research task fans out to four workers. A timeout causes replacement workers to start while the original workers may still be running.

  1. Does giving each replacement the original worker’s full budget preserve the task cap?
  2. What if the operator cancels while children remain queued?
  3. Can the operator rerun every task that read a withdrawn supplier note?
  4. What should a resilience demonstration prove?

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 trackA support bot can look up orders and send emails. Where do you start testing and hardening it?Next in this trackHow would you red-team a multi-agent system?Next in this trackDesign an agentic security operations platform and its trust boundaries.
DISCUSSION · 0

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