1.2k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill chat-ui- _meta.json266 B
- SKILL.md3.1 KB
Overview
This skill provides ready-made chat UI building blocks for React and Next.js based on ui.inference.sh. It includes a container, message list, input, typing indicators, and avatars to assemble messaging interfaces quickly. The components are Tailwind-ready and designed for streaming, real-time, and AI assistant use cases.
How this skill works
Install the chat components and import the pieces you need: ChatContainer, ChatMessage, ChatInput, TypingIndicator, and avatars. Use ChatContainer as the layout, render ChatMessage items for user/assistant/system roles, and wire ChatInput to your send handler. The components support streaming responses and simple className overrides for styling with Tailwind/shadcn tokens.
When to use it
- Building a custom chat or messaging UI in a React or Next.js app
- Integrating an AI assistant or conversational agent front end
- Prototyping conversational flows with streaming responses
- Adding consistent message layout and role-based styling
- Creating a chat widget that follows shadcn/ui and Tailwind conventions
Best practices
- Keep messages as a simple array of {role, content} and map them to ChatMessage for predictable rendering
- Show TypingIndicator during streaming or long-running API calls to improve UX
- Disable ChatInput while awaiting responses to prevent duplicate submissions
- Use className to extend styling with Tailwind tokens instead of editing components directly
- Handle system messages separately (centered) for notices or context changes
Example use cases
- A support chat widget in a web app that streams assistant replies
- An AI coding assistant embedded in a developer dashboard
- A conversational onboarding flow that displays system prompts and tips
- A multi-role chat viewer that shows user, assistant, and system messages
- A prototype for testing streaming LLM responses with live typing indicators
FAQ
Yes. The components are designed to work with streaming APIs; show TypingIndicator during streaming and append partial messages to the list.
Can I customize styling?
Yes. Components accept className props and use Tailwind/shadcn design tokens so you can override or extend styles without changing internals.