Code Review Checklist

Risk — low

0 / 11 checked

Low risk: a focused single pass on the critical items should be sufficient. Budget roughly 32 minutes.

• Does the change do what the description says, and nothing else?criticalScope creep hidden inside a large diff is the most common source of surprise regressions.
• Is every new branch covered by a test that fails without the change?criticalA test that passes on the old code proves nothing.
• Are error paths handled, not just the happy path?criticalMost production incidents happen on the branch nobody exercised locally.
• Are names accurate to what the code now does?Stale names cost the next reader more time than the change saved.
• Is the feature behind a flag or otherwise reversible?Reversibility turns an outage into a toggle.
• What happens on the second call — is it idempotent?criticalRetries and double clicks are guaranteed, not hypothetical.
• Any await inside a loop that should be Promise.all?Serial awaits are the usual cause of a slow endpoint.
• Are useEffect dependency arrays complete and cleanup functions present?Missing cleanup leaks listeners and fires state updates after unmount.
• Is user-supplied content rendered as text, never as raw HTML?criticaldangerouslySetInnerHTML on user input is a direct XSS route.
• Keyboard reachable, with visible focus and sensible labels?Accessibility regressions are cheap now and expensive later.
• What does this render while loading and on failure?Empty and error states are where unfinished features show.