templui-mcp-server
Enables AI assistants to access TemplUI components for building Go applications with the templ templating language.
README
TemplUI MCP Server
🚀 The fastest way to integrate TemplUI components into your AI workflow
A Model Context Protocol (MCP) server that provides AI assistants with comprehensive access to TemplUI components. Get component source code, documentation, examples, and metadata for building Go applications with the templ templating language.
✨ Key Features
- 🎯 Go Templ Support - Native support for Go's templ templating language
- 📦 Component Source Code - Latest TemplUI component implementations
- 🎨 Component Demos - Example implementations and usage patterns
- 📋 Documentation Access - Complete component documentation from templui.io
- 🔧 JavaScript Integration - Client-side functionality for interactive components
- 📊 Metadata Extraction - Component properties, variants, and configuration details
- ⚡ Smart Caching - Efficient GitHub API integration with rate limit handling
🚀 Quick Start
# Basic usage (60 requests/hour)
npx templui-mcp-server
# With GitHub token (5000 requests/hour) - Recommended
npx templui-mcp-server --github-api-key ghp_your_token_here
# Use cached documentation
npx templui-mcp-server --use-cached-docs
🎯 Get your GitHub token in 2 minutes: GitHub Personal Access Tokens
📚 Available Tools
| Tool | Description |
|---|---|
get_component |
Get the source code (.templ file) for a specific component |
get_component_demo |
Get demo/example code showing component usage |
get_component_docs |
Get complete documentation for a component |
get_component_javascript |
Get JavaScript code for interactive components |
get_component_metadata |
Get detailed metadata including props and features |
list_components |
List all available components, optionally by category |
🎨 Supported Components
TemplUI provides 42 components across multiple categories:
Form Components
- Button, Input, Textarea, Checkbox, Radio, Select Box
- Date Picker, Time Picker, Input OTP, Tags Input
- Form, Label, Switch, Rating, Slider
Layout Components
- Card, Separator, Aspect Ratio, Accordion, Table
- Collapsible, Sheet, Sidebar
Navigation Components
- Breadcrumb, Pagination, Tabs
Overlay Components
- Dialog, Dropdown, Popover, Tooltip
Feedback Components
- Alert, Toast, Progress, Skeleton, Copy Button
Display Components
- Avatar, Badge, Icon, Code, Chart, Carousel
🛠️ Installation & Setup
1. Install the Package
# Global installation (optional)
npm install -g templui-mcp-server
# Or use npx (recommended)
npx templui-mcp-server --help
2. Get GitHub Token (Recommended)
# Visit: https://github.com/settings/tokens
# Generate token with no scopes needed for public repositories
export GITHUB_PERSONAL_ACCESS_TOKEN=ghp_your_token_here
3. Test the Server
# Run server
templui-mcp-server --github-api-key ghp_your_token_here
# Server will start and wait for MCP client connections
🔌 Integration with AI Tools
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"templui": {
"command": "npx",
"args": ["templui-mcp-server"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
}
}
}
}
VS Code with Continue.dev
Add to your Continue configuration:
{
"mcpServers": [
{
"name": "templui",
"command": "npx",
"args": ["templui-mcp-server", "--github-api-key", "ghp_your_token_here"]
}
]
}
Cursor
Configure in Cursor's MCP settings:
{
"templui": {
"command": "templui-mcp-server",
"args": ["--github-api-key", "ghp_your_token_here"]
}
}
📖 Usage Examples
Get Component Source Code
// AI Assistant can request:
use_tool("get_component", { "componentName": "button" })
// Returns Go templ source code with helpful comments
Get Component Documentation
// AI Assistant can request:
use_tool("get_component_docs", { "componentName": "dialog" })
// Returns formatted documentation with installation and examples
List Components by Category
// AI Assistant can request:
use_tool("list_components", { "category": "form" })
// Returns all form-related components
Get Component Examples
// AI Assistant can request:
use_tool("get_component_demo", { "componentName": "carousel" })
// Returns example implementations from the showcase
⚙️ Configuration
Command Line Options
templui-mcp-server [options]
Options:
--github-api-key <token> GitHub personal access token
--use-cached-docs Use cached documentation
--log-level <level> Set log level (error, warn, info, debug)
--help Show help message
Environment Variables
GITHUB_PERSONAL_ACCESS_TOKEN # GitHub token for API access
TEMPLUI_DOCS_PATH # Path to documentation files
TEMPLUI_CACHE_DIR # Cache directory location
LOG_LEVEL # Logging level
🎯 Use Cases
- 🤖 AI-Powered Development - Let AI assistants build UIs with TemplUI components
- 📚 Component Discovery - Explore available components and their usage patterns
- 🚀 Rapid Prototyping - Get complete component implementations quickly
- 📝 Code Generation - Generate Go templ code with proper imports and structure
- 🎓 Learning Go Templ - Understand component patterns and best practices
📁 Project Structure
templui-mcp-server/
├── src/
│ ├── index.ts # CLI entry point
│ ├── server/ # MCP server implementation
│ ├── tools/ # Tool implementations
│ ├── utils/ # Utilities (GitHub, docs, cache)
│ ├── data/ # Component metadata
│ └── cli/ # Command line interface
├── samples/ # Sample data (documentation)
├── package.json
└── README.md
🔧 Development
Building from Source
# Clone repository
git clone https://github.com/tggo/templui-mcp-server.git
cd templui-mcp-server
# Install dependencies
npm install
# Build TypeScript
npm run build
# Run development version
npm run dev
Updating Components
To sync with the latest TemplUI components:
# Automated update (recommended)
npm run update:components
# Dry run to see changes
npm run update:components:dry-run
# Update without publishing to npm
npm run update:components:no-publish
The update script will:
- Discover new components from TemplUI repository
- Update static component list
- Bump version number
- Update documentation
- Commit, tag, and push to GitHub
- Publish to npm (if logged in)
Adding Documentation
Place TemplUI documentation files in samples/templui-site-doc/:
samples/templui-site-doc/
├── templui.io_docs_components_button.md
├── templui.io_docs_components_dialog.md
└── ...
🐛 Troubleshooting
Common Issues
-
Rate Limit Exceeded
Error: API rate limit exceeded Solution: Provide GitHub token with --github-api-key -
Documentation Not Found
Error: Documentation directory not found Solution: Ensure samples/templui-site-doc exists with markdown files -
Component Not Found
Error: Component "xyz" not found Solution: Use list_components to see available components
Debug Mode
# Enable debug logging
templui-mcp-server --log-level debug --github-api-key ghp_your_token
📄 API Reference
Tool Schemas
All tools follow consistent parameter patterns:
interface ComponentRequest {
componentName: string; // Required for most tools
}
interface ListRequest {
category?: string; // Optional category filter
}
Response Format
Tools return content in MCP-standard format:
{
content: [
{
type: "text",
text: "Component source code or documentation"
}
]
}
🤝 Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
📄 License
MIT License - see LICENSE for details.
🙏 Acknowledgments
- TemplUI Team - For the amazing Go templ component library
- Go Templ - For the excellent templating language
- Anthropic - For the Model Context Protocol specification
Made with ❤️ for the Go and TemplUI community
Star ⭐ this repo if you find it helpful!
🔗 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。