Webpixels MCP Server
Enables AI assistants to search, retrieve, and assemble Bootstrap UI components from the Webpixels library. It provides tools to explore component categories, fetch HTML snippets, and build complete web pages.
README
Webpixels MCP Server
MCP (Model Context Protocol) server for Webpixels Bootstrap components. This server enables AI assistants like Claude to search, retrieve, and assemble Bootstrap UI components from the Webpixels library.
Installation
Via npx (Recommended)
Add to your Claude desktop configuration (~/.config/claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"webpixels": {
"command": "npx",
"args": ["-y", "@webpixels/mcp"]
}
}
}
Via Global Install
npm install -g @webpixels/mcp
Then add to your configuration:
{
"mcpServers": {
"webpixels": {
"command": "webpixels-mcp"
}
}
}
Available Tools
search_components
Search Bootstrap components by name, category, type, or description.
Parameters:
query(string): Search query (matches name, description, tags)type(string): Filter by component type (component, section, card, screen, layout, page, form, chart)category(string): Category or subcategory slug (e.g., 'sections', 'hero', 'pricing')is_free(boolean): Filter for free components onlylimit(number): Maximum number of results (default: 20)
get_component
Get the HTML code and metadata for a specific component.
Parameters:
id(string, required): Component UUID or slug (e.g., "section-hero-1")
list_categories
List all component categories with component counts.
Parameters:
type(string): Filter categories by component type
assemble_page
Assemble multiple components into a complete HTML page.
Parameters:
components(array, required): Array of component IDs or slugs in display orderlayout(string): Optional layout component ID to wrap the contentincludeAssets(boolean): Include CSS/JS links in a full HTML document (default: true)
get_component_dependencies
Get the dependency tree for a component.
Parameters:
id(string, required): Component UUID or slugdirection(string): "dependencies" (what this uses), "dependents" (what uses this), or "both"
Component Categories
The library includes components across several categories:
- Components: Banners, Cards, Dropdowns, Headers, Modals, Navbars, Sidebars, Tables, etc.
- Forms: Form groups, textareas, form layouts, form examples
- Sections: Headers, Hero, Features, CTA, Pricing, FAQ, Team, Testimonials, Contact, Gallery, Logos, Blog, Footers
- Layouts: Horizontal, Sidebar, Columns
- Pages: Landings, Dashboard, CRUD, Settings, Auth, Profile, Messaging, etc.
Example Usage
User: I need a hero section for a SaaS landing page
Claude: [Uses search_components with query "hero saas"]
[Returns matching hero sections]
User: Show me the code for section-hero-1
Claude: [Uses get_component with id "section-hero-1"]
[Returns HTML and metadata]
User: Create a page with a hero and pricing section
Claude: [Uses assemble_page with components ["section-hero-1", "section-pricing-1"]]
[Returns complete HTML page]
Development
Build
npm install
npm run build
Run locally
npm start
Generate component data
The component data is generated from the Webpixels library's compiled HTML snippets:
# From the library repository
# Option 1: Generate snippets first, then MCP data
npm run snippets
npm run mcp:generate
# Option 2: All-in-one command
npm run mcp:generate:full
The script reads compiled HTML from library/snippets/ (not raw .njk files) to ensure all template includes are resolved.
License
MIT
Links
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。