kiro-debug

Root-cause debugging investigates implementation failures by combining local runtime data, configuration inspection, and external documentation to diagnose recurring issues. It produces structured outputs like ROOT_CAUSE and FIX_PLAN to guide remediation instead of guessing patches.

Debugging a blocked implementation involves root-cause-first analysis that examines runtime evidence and configuration states. This approach identifies why verification fails and outputs a structured FIX_PLAN and VERIFICATION step to unblock progress.

To start root-cause analysis for non-converging remediation, provide a failure summary to initiate the investigation. The process inspects local evidence and external documentation to generate a categorized ROOT_CAUSE and NEXT_ACTION for targeted remediation.

Root-cause debugging is applicable to validation failures across software tasks by inspecting local runtime data and configuration. It generates a structured CONFIDENCE score and VERIFICATION steps to ensure the identified failure category is accurately addressed.

You should use root-cause investigation instead of guess-first debugging when an implementer is blocked or repeated remediation loops fail to converge. This method relies on local evidence and runtime inspection rather than generating immediate patches.

A root-cause debugging investigation produces structured outputs including ROOT_CAUSE, CATEGORY, FIX_PLAN, VERIFICATION, NEXT_ACTION, and CONFIDENCE. These structured results guide remediation by providing a clear path from diagnosis to verified resolution.