Secure AI Agents and MCP Quiz Answer Key
Table of Contents
Return to the Secure AI Agents and MCP Quiz
Use this key after you answer all ten questions. Compare your reason with the security boundary or evidence named below.
Answer Key
| # | Answer | Reason |
|---|---|---|
| 1 | C | Enforcement belongs outside untrusted content and model output. |
| 2 | C | The server validates a trusted host-held, single-use approval token and bound target state before the narrow write. A caller-supplied digest does not prove approval. |
| 3 | B | Schema validates shape. Server logic and identity controls enforce authority. |
| 4 | B | Approval needs the final action, target, payload, and effect. |
| 5 | C | An independent enforcement event proves the decision with useful context. |
| 6 | B | Environment inheritance often carries credentials and host configuration. |
| 7 | A | Tokens need audience and boundary checks. A token for one service should not flow to another. |
| 8 | B | The host should stop after the defined repeated failure threshold. |
| 9 | C | A controlled retest proves the enforcement change addresses the observed path. |
| 10 | B | Security review follows behavior and authority, not the feature label. |
Passing Gate
You pass when your score reaches 8 out of 10 and questions 1, 2, 4, and 5 are correct.
- Pass both gates: Record your score and start the capstone.
- Miss a core question: Revisit the related lesson, repair the artifact, and answer the question again.
- Score below eight: Review every missed topic, update the related artifacts, and retake the quiz from a clean copy.
Verify Your Review
- The recorded score reaches 8 out of 10
- Questions 1, 2, 4, and 5 are correct
- Every missed answer has a corrected reason
- Every related course artifact reflects the correction
Continue with the Security Capstone after you pass both gates.
