FRONT007infoPro

unbounded-ai-output

AI/LLM output rendered in JSX without truncation or overflow handling.

Applies to: JavaScript, TypeScript

Why this matters

AI model responses are unpredictably long — a 'summary' might be one sentence or five paragraphs. Rendering them directly in JSX without truncation (line-clamp, text-ellipsis) or overflow handling (overflow-hidden, max-w-*) can break card layouts, push other content off-screen, and create a poor user experience. Always constrain AI output with CSS truncation or programmatic slicing.

Catch it before it ships

pip install stablestack # or: npx stablestackstablestack # scans your project, FRONT007 includedstablestack explain FRONT007

FRONT007 is part of the Pro rule set. See pricing — the free tier ships 24 checks with no signup.

False positive in your codebase? Suppress a single line with # noqa: FRONT007

More Frontend checks

tabs-url-routing

MUI Tabs should use URL routing via useSearchParams for bookmarkable/shareable tabs.

window-ssr-unsafe

Accessing window/document without SSR check causes server-side errors

inline-jsx-handler

Inline arrow functions in JSX cause unnecessary re-renders

nextjs-redirect-request-url

Redirect uses request.url as URL base, which is wrong behind a reverse proxy.

useeffect-side-effect-unguarded

useEffect contains a mutating HTTP call without a guard against double-firing.

stale-time-dependent-ui

Time-dependent filtering without a re-render timer will show stale data.

loading-returns-null

Returning null during loading causes layout shift and flash of empty content.

eager-queries-no-defer

3+ query hooks all firing on mount with no deferred loading.