review-code

To review pull request code for architecture and design issues that linters miss, use a design-focused code review tool to audit changed files for single-responsibility violations, poor modularity, and structural smells. It outputs structured findings without applying fixes.

The best way to find files that need to be split or simplified before merging a working tree diff is to run a design audit on the changed files. It identifies dense files, mixed responsibilities, and duplication, helping you clean up code before merge.

Yes, you can generate structured code review findings for downstream automation workflows. This audit process emits results in a strict schema format, allowing downstream review and fix workflows to consume and orchestrate the identified design issues.

Yes, this code review approach works with explicit file lists and excludes generated or vendored files. It performs read-only analysis on targeted files, maintains repository convention awareness, and focuses only on the structural quality of the provided source files.

You should use a design-focused code review instead of an automated linter when your pull request passes linting but still feels messy or overgrown. It catches judgment-based issues like weak modular boundaries, poor naming, and dead code that automated formatters miss.

Your pull request passes automated linting but still has readability and modularity problems because linters do not reliably catch judgment-based design flaws. A dedicated code design audit is required to identify responsibility boundary violations, layering issues, and structural smells.