andresnator Load this skill when reviewing refactor plans for naming and readability in languages without a dedicated naming skill.
Review goals
- Prefer names that reveal intent and domain meaning.
- Flag misleading abbreviations, overloaded terms, and hidden units or formats.
- Distinguish readability issues from purely stylistic preferences.
- Avoid mass renames unless the evidence shows concrete maintenance benefit.
Confidence rules
- Use high confidence only when the issue is directly visible from the code or symbol names.
- Lower confidence when the recommendation depends on ecosystem or language-specific naming conventions that are not proven in the repository.
- If the code is too small or names are already clear, return no findings instead of inventing polish work.
Output notes
- Evidence must cite concrete file/line/symbol examples.
- Cosmetic-only rename suggestions belong out of executable work unless maintainability value is explicit.