Guide-linked quiz · Medium Level
How to Write Code Documentation with AI Quiz
Check how to prompt AI for code docs, validate examples against implementation, and keep generated explanations accurate.
Question 1 of 8
An assistant drafts a docstring for a function that validates a file path. What should the developer verify first?
Preview every question8 questions
Answers and explanations appear as you take the quiz.
- An assistant drafts a docstring for a function that validates a file path. What should the developer verify first?
- The described inputs, return value, side effects and edge cases against the code and tests
- Whether the prose uses the most technical vocabulary
- Whether the comment is longer than the function
- Whether the model sounds confident
- A README says to run a setup command generated by AI. What is a strong validation step?
- Assume it works because the command is familiar
- Remove all installation steps
- Run it in a clean environment and confirm it matches the project scripts and configuration
- Ask the assistant to repeat the command with more confidence
- Which inline comment is most useful?
- A paraphrase of the next line with no added context
- A note explaining a non-obvious invariant or reason for the code
- A claim that the function is secure without evidence
- A list of unrelated TODOs
- A generated API example imports a plausible package name not used in the repository. What should happen?
- Keep it because the package name looks conventional
- Change the dependency file to match the example without review
- Delete the API documentation
- Check the import against actual dependencies and run the example
- Why should developers avoid sending secrets to an unapproved coding assistant?
- The code or context may be processed under data-handling terms that have not been approved
- Secrets make comments harder to read
- The assistant needs every credential to understand a function
- A README cannot describe a credential safely
- A performance paragraph claims an algorithm is O(n), but no analysis or test supports it. What is the responsible edit?
- Keep it because asymptotic notation sounds precise
- Ask the model to invent a benchmark
- Remove or qualify the claim until it is verified
- Replace it with a security guarantee
- A new CLI flag changes behavior. When should its documentation be updated?
- Only after users report that the README is wrong
- With the code change and review, including a checked example
- Never, because flags are discoverable from source
- After deleting the tests
- Why might a code assistant describe an incorrect design reason even when it reads the function?
- Source code contains every team decision automatically
- Tests always explain product intent completely
- Documentation never needs human input
- The rationale or operational context may not be inferable from source code alone
Keep testing yourself
More quizzes picked for your level.