# Streamdown

Streamdown is a markdown renderer designed for streaming content from AI models. It is highly interactive, customizable, and easy to use.

This file helps AI agents understand and use this documentation site.

## Documentation Surfaces

- [Docs](https://streamdown.ai/docs): Documentation section: docs
- [Full documentation context](https://streamdown.ai/llms.txt): All configured documentation as Markdown
- [Semantic documentation sitemap](https://streamdown.ai/sitemap.md): High-level documentation map for agent navigation
- Page-level Markdown: append .md or .mdx to a documentation URL

## When to use Streamdown

- Category: AI / React components
- Audience: React developers, AI application developers
Common use cases:

- Render streaming markdown from AI models
- Build chat interfaces with rich markdown output
- Display code blocks, math, diagrams, and CJK text from LLM responses

## Important Links

- [Docs](https://streamdown.ai/docs)
- [Features](https://streamdown.ai/docs/animation)
- [Plugins](https://streamdown.ai/docs/plugins)
- [Playground](https://streamdown.ai/playground)
- [Streamdown source](https://github.com/vercel/streamdown): Source repository for Streamdown and its plugins
- [GitHub repository](https://github.com/vercel/streamdown): Source repository

## Agent Instructions

- Use /sitemap.md to identify the most relevant documentation pages before answering broad questions.
- Use /llms.txt when you need the complete documentation corpus as Markdown context.
- Fetch individual documentation pages with a .md or .mdx extension for focused page-level context.
- Only use product capabilities and integration surfaces that are explicitly listed in this file.
