classifying-review-findings

Code review standardization assigns findings to five consistent severity levels—CRITICAL, IMPORTANT, DEBT, SUGGESTED, or QUESTION—with hybrid emoji labeling. This ensures uniform triage routing and remediation prioritization across pull requests and review comments, meeting organizational conventions and audit requirements.

Before assigning severity, verify three prerequisites: the execution path is traceable, alternative handlers exist for the issue, and language semantics are clear. These checks ensure classification accuracy and prevent misrouting of findings to inappropriate severity buckets.

Yes. The Skill implements hybrid emoji and text labeling, combining visual indicators with severity text (CRITICAL, IMPORTANT, DEBT, SUGGESTED, QUESTION) to enable fast triage and consistent communication across review comments and pull request feedback.

Standardized severity classification routes findings to the correct remediation action based on risk level. CRITICAL flags urgent blockers, IMPORTANT requires fixes before merge, DEBT defers refactoring, SUGGESTED aids quality, and QUESTION prompts discussion—enabling teams to prioritize review effort.

A finding is ready when it has a traceable execution path demonstrating how the code reaches the issue, confirmation that other handlers or mitigations do not already address it, and clear language semantics so reviewers understand the problem without ambiguity.