Quality Engineering & QA Agent FAQs

Question 1

How does the sub-agent architecture work?

Accepted Answer

When you request a QA task, the skill spawns a scoped sub-agent that only loads the specific reference files needed (e.g., test-strategy.md). This keeps unrelated testing knowledge out of your main conversation, saving tokens and improving accuracy.

Question 2

Can I use it to design test data?

Accepted Answer

Yes, it has a dedicated 'test-data' task type that generates specifications for synthetic data, fixtures, valid/invalid values, and setup/teardown dependencies.

Question 3

Does it support different testing methodologies?

Accepted Answer

Absolutely. It includes specific routing for exploratory testing, smoke tests, sanity tests, boundary value analysis, and session-based testing (SBET).

Question 4

Can this skill help determine if a task is actually 'Done'?

Accepted Answer

Yes. It uses an Empirical Validation protocol to distinguish between 'DONE' and 'CODE_COMPLETE'. It identifies if code is structurally sound but still requires runtime validation, such as UI animations or performance profiling.

Question 5

What automation frameworks does it recommend?

Accepted Answer

The skill provides framework-agnostic strategies but includes an 'automation-strategy' routing that evaluates your specific tech stack to recommend the best tools for CI/CD integration and ROI.

Quality Engineering & QA Agent

主要功能

使用场景

Quality Engineering & QA Agent

主要功能

使用场景