Research and defect analysis¶
Two capabilities often stand out in practice: disciplined research and disciplined defect analysis.
Research¶
/research is designed to move from question to defensible answer with explicit evidence handling and bounded experimentation when needed.
Good use cases¶
- technology option assessment;
- control or design trade-off analysis;
- root-cause hypotheses requiring validation;
- questions where the organisation needs a structured argument rather than a quick opinion.
Defect analysis¶
/diagnose forces the investigation to respect software engineering reality:
- compare with the last known good state;
- compare with a working analogue;
- treat timing/order effects as first-class possibilities;
- produce a regression-proof fix contract.
This is a materially stronger pattern than “guess a patch from the failing symptom”.