1.10Self-Assessment Quiz

Test your understanding of Chapter 1:

1. What are the five elements of an effective test-generation prompt?
2. Why is pasting an OpenAPI schema more effective than describing it?
3. What is the typical negative-to-positive test ratio in a well-designed suite?
4. Why should you provide a style reference in your prompt?
5. How many iterations does the prompt refinement loop typically require?
6. What happens when you ask for 100 tests in a single prompt?
7. Name three things that strong context tells the LLM that weak context does not.

Answers:

1. Context, Artifact, Constraints, Coverage Goals, Output Format
2. The schema provides exact field names, types, constraints, and response codes -- eliminating ambiguity and hallucination
3. Approximately 2:1 negative to positive
4. The AI replicates naming conventions, fixture usage, assertion style, and code structure from examples
5. 2-3 iterations
6. Quality degrades in later tests; generate in batches of 10-20 instead
7. Technology stack details, external dependencies, data formats, authentication mechanism, business flow, architecture style