Agents You Already Use

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-agentic-agents-everywhere

1. Which sentence best captures the main idea of 'Agents You Already Use'?

   1. You have already used agents — Alexa, Siri, Google Assistant.
   2. Tools and goals are unnecessary for agent design
   3. Agents should always run without limits or oversight
   4. Agents and chatbots are the same thing in every way

2. Which of the following is part of '2026 vs 2018 agents'?

   1. Never log what the agent did
   2. Hide tool calls from the operator
   3. 2018 was 'Set a timer.' 2026 is 'Find a recipe with ingredients I have, scale it to 6 servings, and start a 25-minute timer for the oven.'
   4. Run unbounded retries on any error

3. Which of the following is part of 'Three categories of agents you know'?

   1. Voice assistants (Alexa, Siri, Google)
   2. Approve all actions automatically
   3. Skip every form of evaluation
   4. Avoid taking any actions in the world

4. Which of the following is part of 'Review date'?

   1. Disable safety checks for speed
   2. Always run with no oversight
   3. Never log what the agent did
   4. Reviewed in 2026. Treat fast-changing product names, prices, availability, and policy details as examples to verify before use.

5. What is 'voice assistant' in this context?

   1. A core concept covered in Agents You Already Use
   2. A way to disable the agent's tools
   3. A reason to skip all logging
   4. A trick to bypass approvals

6. What is 'NPC' in this context?

   1. A reason to skip all logging
   2. A trick to bypass approvals
   3. A core concept covered in Agents You Already Use
   4. A way to disable the agent's tools

7. What is 'recommendation' in this context?

   1. A trick to bypass approvals
   2. A reason to skip all logging
   3. A core concept covered in Agents You Already Use
   4. A way to disable the agent's tools

8. Where are AI agents already showing up in everyday tools?

   1. Customer support, code editors, email assistants, and many SaaS apps
   2. Nowhere yet
   3. Only on supercomputers
   4. Only in games

9. Before letting an agent take a destructive action, what is the safest default?

   1. Hide the action from any log
   2. Approve once and let the agent repeat forever
   3. Skip approvals if the user trusts the agent
   4. Require explicit human approval for the specific action

10. An agent quietly retries a failed payment 50 times overnight. What design principle was missing?

    1. Bounded retries with human notification on repeated failure
    2. A larger model
    3. More creative prompting
    4. A bigger context window

11. Which signal best tells you an agent is stuck in a runaway loop?

    1. It keeps repeating the same tool call with no new progress
    2. It finishes the task in one step
    3. It asks one clarifying question
    4. It returns a short summary and stops

12. What does an 'eval' for an agent measure?

    1. The exact wording of every prompt
    2. The temperature setting
    3. How polite the model sounds
    4. Whether the agent reliably completes a defined task end to end

13. What is the best response when an agent suggests an action you do not understand?

    1. Ask the agent to explain the action and its expected effect before approving
    2. Approve it to keep things moving
    3. Reject everything and stop using the agent
    4. Run it twice to be sure

14. Why is logging every tool call an agent makes a baseline requirement?

    1. Logs replace the need for testing
    2. Logs make the model run faster
    3. Logs are only for legal teams
    4. Logs are needed to debug, audit, and explain agent behavior to users

15. Why are clear success criteria critical when building an agent?

    1. They make the agent sound smarter
    2. They reduce the number of tokens used
    3. They are required by law
    4. Without them you cannot tell whether the agent worked or guess