code-reviewer

To catch code quality issues that static analysis tools miss, review code for judgment-call problems like poor naming, unclear error handling, nested logic, and misleading comments to ensure alignment with CTO profile standards before merge.

To review AI-generated code for quality, scrutinize it for verbose boilerplate, generic variable names, and idiom mismatches that commonly pass automated linting checks but degrade readability and maintainability in Python, TypeScript, or Java codebases.

Yes, you can enforce consistent code review standards across Python, TypeScript, C#, Java, C, C++, and SQL codebases by applying uniform rules for naming conventions, error handling, and anti-patterns to catch issues automated tools miss.

To validate test code and prevent silent CI failures, block empty catch blocks, identify tests without assertions, and catch silent test failures that hide bugs from automated quality gates.

To block pull request phase advancement on critical issues, emit severity-calibrated findings during the refinement loop, enforcing anti-patterns and CTO profile red lines before allowing a merge.

Automated linting passes code with poor error handling and nested logic because these are judgment-call issues impacting readability and maintainability that require semantic code review to detect and evaluate against best practices.