Bug Root Cause Fix Copilot

Problem It Addresses

When a stack trace, failed test, regression, or functional anomaly appears, the bottleneck is often not visibility but fragmented evidence, over-speculation, and overly broad fixes. This skill constrains debugging into an executable path: confirm reproduction boundaries, identify a root cause with evidence, then land a minimal change and verification checklist, reducing the risk of creating another regression.

How It Works

• Inputs: uses error messages, reproduction steps, relevant code paths, and constraints; if critical context is missing, asks at most three focused questions before proceeding with reasonable assumptions.
• Diagnosis: prioritizes high-probability causes such as null values, types, boundaries, state synchronization, timing, and configuration, and requires an evidence chain instead of a generic list of possibilities.
• Fix scope: prefers the smallest viable change, identifies affected files or logic, explains why impact should not expand, and notes rollback needs when relevant.
• Verification: produces manual test steps and the minimum necessary automated cases, turning fixed into a checkable outcome.

Boundaries

It fits engineering debugging with observable errors or clear anomalies, especially when a fast convergence to a conclusion is needed. It is not intended for open-ended architecture exploration, pure capacity planning, or issues with no observable symptom. When evidence is insufficient, it marks assumptions and open questions rather than making destructive suggestions.