completeness-review

To scan your codebase for stubs and TODOs before deployment, run a completeness check to detect empty bodies, placeholders, and unfinished implementations, then review the generated severity report to resolve critical issues. It cross-references implemented code against feature definitions to ensure full functionality.

A completeness check in code review is an automated scan that identifies markers of incomplete code like TODOs, FIXMEs, placeholders, and debug statements. It validates that features marked as complete are genuinely implemented by verifying feature definitions and project deliverables against the actual codebase.

You verify feature completeness against a features.md file by scanning the codebase to cross-reference implemented code with the defined feature specifications. This process identifies and reports any critical, high, medium, or low severity incompletions where functionality is missing or only stubbed out.

Yes, you can detect empty function bodies and placeholder code automatically through pattern scanning. The scan targets common markers of incomplete code, catching functions that only return dummy data or contain unfinished implementations before they slip into production.

You validate project deliverables against a project-plan.md file by running an automated check that ensures all planned items have corresponding, complete code implementations. It flags any deliverables lacking functional code, preventing unfinished features from reaching deployment.

When finding unfinished code implementations, the completeness review reports incompletions across four severity levels: critical, high, medium, and low. This categorization helps prioritize which stubs, TODOs, or missing feature implementations must be resolved before a robust release.