Designing the Toolbox You Hand Your Agent

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-agentic-ai-agent-give-it-tools-r10a8-teen

1. Which sentence best captures the main idea of 'Designing the Toolbox You Hand Your Agent'?

   1. Agents should always run without limits or oversight
   2. Tools and goals are unnecessary for agent design
   3. Pick the smallest set of tools that lets the agent finish the job.
   4. Agents and chatbots are the same thing in every way

2. Which of the following is part of 'Some examples'?

   1. Giving an agent only read_file and run_tests
   2. Always run with no oversight
   3. Ignore cost when scaling
   4. Disable safety checks for speed

3. Which of the following is part of 'The rule of thumb'?

   1. fewer, sharper tools beat a giant pile of overlapping ones.
   2. Disable safety checks for speed
   3. Hide tool calls from the operator
   4. Approve all actions automatically

4. Which of the following is part of 'You did it!'?

   1. Use the most expensive model regardless of fit
   2. Nice. You just practiced designing the toolbox you hand your agent. Do it three more times this week and it stops feeling like a trick and starts feeling like a tool.
   3. Always run with no oversight
   4. Hide tool calls from the operator

5. What is 'tool design' in this context?

   1. A way to disable the agent's tools
   2. A trick to bypass approvals
   3. A reason to skip all logging
   4. A core concept covered in Designing the Toolbox You Hand Your Agent

6. What is 'scope' in this context?

   1. A core concept covered in Designing the Toolbox You Hand Your Agent
   2. A reason to skip all logging
   3. A trick to bypass approvals
   4. A way to disable the agent's tools

7. What is 'tool naming' in this context?

   1. A trick to bypass approvals
   2. A reason to skip all logging
   3. A way to disable the agent's tools
   4. A core concept covered in Designing the Toolbox You Hand Your Agent

8. What changes the moment you 'give' a model tools?

   1. The model becomes a different model
   2. It stops being able to chat
   3. It gains the ability to take actions, and you must reason about safety, scope, and cost
   4. It no longer needs prompts

9. Why is keeping a human in the loop valuable for high-stakes agent actions?

   1. It replaces the model entirely
   2. It speeds the agent up
   3. It removes the need for any logging
   4. It catches mistakes before they cause real-world harm

10. What is the safest first place to deploy a brand new agent?

    1. Production, against real customers
    2. A sandbox or low-stakes task with reversible actions
    3. Inside a critical billing system
    4. On a public server with no auth

11. Which budget control most directly prevents runaway costs from an agent loop?

    1. A friendly system prompt
    2. A longer context window
    3. A hard cap on steps, tokens, or dollars per task
    4. A bigger model

12. Why is it dangerous to give an agent access to your email and calendar without scoped permissions?

    1. It will refuse to work
    2. Scopes only matter for paid accounts
    3. Scoped permissions slow the model down
    4. Broad access means a single misstep can send the wrong message or wipe events

13. What is the difference between an agent's memory and its context window?

    1. Context is what the model sees right now; memory persists across runs
    2. Nothing — they are the same thing
    3. Memory is faster but less accurate than context
    4. Context lasts forever; memory is cleared every minute

14. Which is the best way to think about an agent's 'autonomy level'?

    1. It is binary — fully autonomous or not
    2. Autonomy does not affect risk
    3. It is a slider from suggestion only, to ask-then-act, to act-then-report, to fully autonomous
    4. It depends only on the model size

15. Why does an AI agent need 'tools' such as a browser, calendar, or code runner?

    1. Tools let the agent take actions in the world instead of only producing text
    2. Tools shrink the context window
    3. Tools make the model speak more naturally
    4. Tools replace the need for any prompts