analyze-problem

Structured debugging identifies a software incident root cause by separating verified facts from assumptions and testing competing hypotheses. A strict 10-step framework extracts context, evaluates evidence, and recommends the smallest safe fix with a verification plan.

Evidence-based problem analysis for complex regressions avoids premature fixes and confirmation bias by explicitly stating uncertainty. It generates a report distinguishing facts from hypotheses to pinpoint the most likely root cause before proposing changes.

Incident triage with multiple hypotheses uses a fixed 10-step framework to evaluate competing explanations against available context. It separates facts from assumptions, handles uncertainty explicitly, and outputs a structured analysis report.

Planning safe fixes after technical risk assessment involves recommending the smallest viable solution while listing rejected alternatives. It requires generating a verification plan to ensure the proposed change resolves the root cause safely.

Avoid hypothesis testing for software issue triage when the problem is trivial or the root cause is already visually obvious. It is designed for complex incidents and regressions requiring strict evidence evaluation and uncertainty handling.