code-review-and-quality

Code review for merge readiness evaluates correctness, readability, architecture, security, and performance in a single pass. It checks tests first, verifies implementation logic, and labels severity findings to block defects before entering the main branch.

Reviewing AI-generated code applies the same five-axis discipline: correctness, readability, architecture, security, and performance. It enforces test-first evaluation and dead code detection to ensure the generated change clearly improves code health before merge.

Security and performance review on a feature branch checks tests, implementation, and verification hygiene. It identifies regressions and risks, applying severity-labeled findings to ensure the branch meets practical approval standards without demanding perfection.

Code quality checks enforce review discipline by evaluating tests first, then implementation, then verification and dead code hygiene. This test-first approach ensures the change is validated before deeper architectural and security analysis begins.

Automated code review detects dead code and enforces dependency discipline during the verification phase. It ensures changes do not introduce unused code or rogue dependencies before approving the pull request for merge.

A full five-axis code review is unnecessary for trivial changes that do not affect architecture or security. However, for feature implementations, bug fixes, and refactors, applying the complete review lifecycle ensures merge readiness and prevents regressions.