vibecodng.com

Free MCP betaIndex and tools live

Measured entry

hextaui-ai-streaming-response

chatminimal-flatmotion on file

A client-side React renderer that reveals whitespace-delimited tokens with punctuation-aware pauses, then hands the text to HextaUI's AI Message for Markdown, GFM, copy actions, and Shiki highlighting. It ships as a shadcn registry item and relies on Tailwind utilities, theme support, registry button and menu primitives, and its declared Markdown and Radix packages.

Specimen
Motion capture of the mounted demo, recorded by the harness. No audio track.
Harness screenshot of hextaui-ai-streaming-response at 375 pixels wide
Harness screenshot at 375 pixels wide. This viewport scrolled sideways.
Harness screenshot of hextaui-ai-streaming-response at 768 pixels wide
Harness screenshot at 768 pixels wide. No horizontal overflow recorded.
Harness screenshot of hextaui-ai-streaming-response at 1440 pixels wide
Harness screenshot at 1440 pixels wide. No horizontal overflow recorded.

Measured

Every value below was written by the harness on the run that produced the frames above. None of it is derived from the prose.

Harness measurements recorded for hextaui-ai-streaming-response
BuildOK
MountOK
Console errors0
Uncaught page errors0
Horizontal overflowflagged at 375
Axe critical0
Axe serious0
Axe moderate0
Axe minor0
Gzip marginal1938.3 KB (2003.8 KB build, 65.5 KB baseline)
Bare modeFAIL
Provider requiredNO
Demo usedwritten from the component source API

What the harness recorded

BARE BUILD

✗ Build failed in 1.95s
error during build:
[vite]: Rollup failed to resolve import "lucide-react" from "<sandbox>re-3e61b8e7c1b5/src/cut/blocks/ai/ai-message.tsx".

Verdict

A capable fit for shadcn chat surfaces that need polished Markdown playback from complete text. The prepared build mounted cleanly and produced no automated accessibility findings, but narrow overflow and the heavy highlighting stack limit its appeal. It needs a shadcn-style setup, and the bare install failed because lucide-react was unresolved.

Use it if

  • You need a simulated assistant response from complete text rather than transport-level token streaming.
  • Your chat already uses shadcn conventions and needs Markdown, GFM, copy actions, and highlighted code.
  • You want punctuation-aware pacing and an automatic reduced-motion fallback.

Skip it if

  • You need to append real server tokens as they arrive; this component animates a finished content string.
  • Bundle restraint matters more than Markdown and code highlighting, since the Shiki-based message dependency is substantial.
  • Your chat must fit narrow containers without adjustment; the verified narrow view overflowed.
  • You want a framework-neutral drop-in with no Tailwind, path-alias, theme, or registry wiring.

Install

Installation record for hextaui-ai-streaming-response
Declared dependenciesreact-markdown, remark-gfm, shiki, next-themes, @radix-ui/react-slot, @radix-ui/react-dropdown-menu, clsx, tailwind-merge
Missing on installnone
Bare installfails once the shadcn-style helpers are removed, so it needs that setup
Providerno provider required in the harness

In a non-shadcn app, add Tailwind with the expected design tokens, the cn utility, and the path alias; copy in AI Message, Button, and Dropdown Menu registry files; install React Markdown, remark GFM, Shiki, next-themes, lucide-react, Radix Slot and Dropdown Menu, clsx, and tailwind-merge; then wire theme support.

Source and credit

This page is a review, not a copy. The component source lives with its author and this site does not republish it. Read the license before you use anything.

Author repository

github.com/preetsuthar17/HextaUI
License
MIT
Record published
2026-08-17

Alternatives

Other published entries for the chat pattern, ordered by what they measured: mounts first, then clean axe scans, then bare-mode passes, then the smaller bundle.