Material Design MCP Server
Enables browsing Material Design 3 documentation by listing top-level sections and retrieving page content as Markdown.
README
Material Design MCP Server
A Cloudflare Worker that exposes an MCP (Model Context Protocol) server for browsing the Material Design 3 documentation.
Features
list_sections— Discover top-level documentation sections (Foundations, Components, Styles, etc.)get_page— Navigate any Material Design page and get its content as clean Markdown
Architecture
- Runtime: Cloudflare Workers (TypeScript,
nodejs_compat) - Router: Hono
- Browser automation: Browser Run (Puppeteer binding)
- Markdown conversion: Workers AI
toMarkdown() - MCP transport: Streamable HTTP (
@modelcontextprotocol/serverv2)
Because m3.material.io is a JavaScript-heavy SPA, the worker uses Puppeteer to wait for the client-side <article> element to render before extracting content.
Prerequisites
- Node.js 18+
- pnpm
- A Cloudflare account with:
- Browser Run enabled
- Workers AI enabled
Setup
# Install dependencies
pnpm install
# Generate TypeScript types from wrangler.jsonc
pnpm run gen:cf-types
Development
# Start local dev server (connects to remote Cloudflare services)
pnpm dev
The MCP endpoint is available at http://localhost:8787/mcp.
Deployment
pnpm deploy
MCP Tools
list_sections
Returns the top-level documentation sections from the Material Design sitemap.
Input: none
Example output:
blog
components
develop
foundations
get-started
styles
get_page
Navigates to a specific Material Design page, extracts the article content, and converts it to Markdown.
Input:
path— URL path onm3.material.io(e.g./foundations/overview)
Example output:
## Accessibility & Material Design
**Accessibility by default** is a core design value for Material...
Connect an MCP Client
Any MCP client that supports Streamable HTTP can connect to:
https://<your-worker>.<your-subdomain>.workers.dev/mcp
For example, with Claude Desktop:
{
"mcpServers": {
"material-design": {
"url": "https://material-mcp.your-account.workers.dev/mcp"
}
}
}
License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。