Agent Permission Revocation: When Trust Breaks

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-agentic-agent-permission-revocation-creators

1. A system uses the same credential for all agents to simplify management. Why is this approach risky when a security incident occurs?

   1. Revoking that single credential would disable every agent, including trusted ones
   2. It makes agent tracking more difficult for administrators
   3. It slows down the authentication process for all agents
   4. The credential will automatically expire too quickly

2. What is the security advantage of short TTL (time-to-live) credentials in an agentic system?

   1. They make authentication faster for users
   2. They reduce the window of time an attacker can use stolen credentials
   3. They require less storage space in the database
   4. They eliminate the need for password policies

3. What does a global kill switch allow an administrator to do in an emergency?

   1. Reset all agent passwords to default values
   2. Revoke all agent permissions instantly across the entire system
   3. Upgrade all agents to the latest version simultaneously
   4. Temporarily pause individual agents for maintenance

4. An organization has never tested its revocation procedures. What is the most likely consequence if a real incident occurs?

   1. The revocation will work perfectly because the system is well-designed
   2. The process will be slow, confused, and may fail to contain the breach
   3. The AI will automatically fix any procedural mistakes
   4. The attackers will be automatically identified and removed

5. Which statement accurately describes what AI systems cannot do regarding permission revocation?

   1. AI cannot generate new credentials for agents
   2. AI cannot detect when an agent has been compromised
   3. AI cannot recover from incidents without pre-existing revocation infrastructure
   4. AI cannot authenticate agents using standard protocols

6. What is the primary purpose of conducting an impact assessment before revoking agent permissions?

   1. To identify what legitimate business functions will be disrupted
   2. To determine which agents were involved in the incident
   3. To generate a report for regulatory compliance
   4. To calculate the financial cost of the incident

7. A company implements credential rotation every 90 days. An attacker steals a credential on day 1, and the breach is detected on day 45. How many days could the attacker potentially use the stolen credential?

   1. 44 days
   2. 45 days
   3. 1 day
   4. 90 days

8. An organization has three agents: Agent A (trusted), Agent B (suspicious), Agent C (trusted). All three share the same credential. If Agent B is compromised, what is the primary problem with revocation?

   1. The system cannot determine which agent is suspicious
   2. Credential databases will become corrupted
   3. Revoking the shared credential affects A and C unnecessarily
   4. The attackers will know which agent was revoked

9. A global kill switch is activated during an incident. What should be prepared in advance to restore legitimate services afterward?

   1. A report for the company board of directors
   2. A backup of all agent machine learning models
   3. A post-revocation recovery plan for non-incident scenarios
   4. A list of all attackers who triggered the kill switch

10. A system uses one-year TTL credentials. During a breach detected after 6 months, the security team revokes the credential. How does this compare to using 24-hour TTL credentials?

    1. The longer TTL automatically rotates credentials faster
    2. The longer TTL requires less administrative oversight
    3. The longer TTL provides stronger encryption
    4. The longer TTL means the attacker had 6 months of access versus potentially 24 hours

11. Why is drilling the revocation process important even when the infrastructure is well-designed?

    1. Drilling prevents future attacks from occurring
    2. Drilling ensures human operators can execute procedures quickly under stress
    3. Drilling automatically updates the system's threat database
    4. Drilling makes the AI learn from past incidents

12. A financial services company has an agent that processes daily transactions. The security team must revoke its permissions due to a suspected compromise. What is the most important consideration before executing revocation?

    1. If the agent's name appears in incident logs
    2. Which executive approved the agent
    3. Whether the agent has recently downloaded new data
    4. How many transactions will be delayed

13. An organization implements unique credentials per agent but has no kill switch. A sophisticated attacker compromises multiple agents simultaneously. What is the primary challenge?

    1. Credential databases will become unsynchronized
    2. Identifying which credentials belong to which agent
    3. Manually revoking each agent's credentials one-by-one takes time during an active attack
    4. The agents will refuse to communicate with the system

14. After revoking permissions for a compromised agent, a legitimate agent suddenly stops working. What likely happened?

    1. The legitimate agent was also compromised
    2. The impact assessment was incomplete and legitimate dependencies were not identified
    3. The system requires all agents to be revoked simultaneously
    4. The AI decided to revoke all agents automatically

15. A company decides to use extremely short TTL credentials (every 15 minutes). What is the most significant operational challenge?

    1. Users will need to re-authenticate too frequently
    2. Systems must handle continuous credential refresh without service interruption
    3. Attackers can easily steal credentials within 15 minutes
    4. The company will run out of valid credentials