bug-hunter

Root cause analysis for flaky tests requires structured debugging across reproduction, recent changes, and execution-path tracing. You collect observed versus expected behavior and evidence to form a single hypothesis, rule out alternatives, and propose the smallest clean fix with a validation plan.

Debugging asynchronous state failures starts with capturing the complete error signal and deciding if the issue is reproducible. By tracing execution paths and backward data state, you can map the exact failure point and outline a minimal safe fix with validation steps.

Investigating build failures and regressions involves analyzing recent diffs alongside the full error logs. You compare observed behavior against expected outcomes to isolate the triggering change, rule out alternative causes, and plan the smallest safe correction.

Yes, this approach explicitly handles non-reproducible bugs by determining what specific constraints prevent reproduction. It then uses execution-path tracing and backward data or state tracing to build a root-cause hypothesis from the available evidence.

Yes, this debugging method works without external dependencies. It operates as a standalone reasoning process that guides you through reproduction, evidence collection, and hypothesis validation using your existing logs, stack traces, and code diffs.

You should avoid proposing fixes until you have collected complete evidence, traced the execution path, and ruled out alternative hypotheses. Proposing a fix prematurely risks applying an ineffective symptom patch instead of resolving the actual root cause.