Module 07 Knowledge Check
5 questions. Pass mark 4 out of 5. Answer every question before checking the answer key below, then retry after reading the feedback.
1. A tool description matters because…
- A. It is documentation
- B. The model chooses tools based on it - a vague description causes wrong tool selection
- C. It affects speed
- D. It sets permissions
2. Arguments produced by the model for a tool call must be…
- A. Trusted
- B. Validated and constrained before execution
- C. Logged only
- D. Reformatted
3. An agent loop needs a step limit because…
- A. Of rate limits
- B. Without one it can loop indefinitely, burning budget without converging
- C. Tools are slow
- D. Models are large
4. Giving an agent a tool that sends email requires…
- A. Nothing special
- B. A confirmation step before execution and least-privilege scoping
- C. A faster model
- D. More examples
5. If a tool returns content from an untrusted source, you must assume…
- A. It is safe
- B. It may contain instructions attempting to steer the agent
- C. It is empty
- D. It is validated
---
Answer Key and Explanations
Check these only after attempting every question.
1. B - The model chooses tools based on it - a vague description causes wrong tool selection
Tool descriptions are prompt surface. Write them for the model's decision, not for a human reader.
2. B - Validated and constrained before execution
The model proposes; your code decides. That boundary is where safety is enforced.
3. B - Without one it can loop indefinitely, burning budget without converging
Bounded iteration plus a terminal condition is what makes an agent operable.
4. B - A confirmation step before execution and least-privilege scoping
Irreversible outbound actions are the canonical case for a human gate.
5. B - It may contain instructions attempting to steer the agent
Indirect prompt injection arrives through tool results as readily as through user input.
Practical Check
Apply this module to your own work: complete the module activity for *Tools and Simple Agents*, then write one sentence naming what your result shows and one naming what it does not.
Strong Answer Pattern
A strong answer names the task, the evidence used, the check performed, and the remaining limitation. It avoids "proved", "guaranteed", or "always" unless the evidence genuinely supports it.