aminvishvam/fluentui-mcp-server
If you are the rightful owner of fluentui-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
The Fluent UI MCP Server is an AI-powered server designed to assist in the development of Fluent UI React components, leveraging the Model Context Protocol for enhanced capabilities.
š Fluent UI MCP Server
A comprehensive Model Context Protocol (MCP) server that provides intelligent assistance for Fluent UI React component development. This server acts as your AI-powered companion for creating, validating, and maintaining Fluent UI applications following v9 design system patterns.
What is MCP? The Model Context Protocol enables AI assistants to securely access external data sources and tools. This server extends AI capabilities with deep Fluent UI knowledge.
⨠Key Features
š§ Intelligent Component Knowledge
- š Comprehensive database of 50+ Fluent UI v9 components
- š Smart component search and discovery
- š Complete props, slots, and usage documentation
- š” Real-world examples and best practices
- šÆ Context-aware recommendations
šØ Design System Mastery
- š 45+ design tokens across 6 categories (colors, typography, spacing, shadows, borders, motion)
- ā Token validation and usage guidance
- š Semantic color system with accessibility compliance
- š Consistent spacing and typography scales
- š¬ Motion and animation guidelines
š ļø Advanced Code Generation
- ā” Generate complete v9 component implementations
- šŖ Create custom hooks following Fluent UI patterns
- š
Generate styles using
makeStylesand design tokens - š TypeScript interfaces with full type safety
- šļø Template-based scaffolding for common patterns
ā Smart Validation & Analysis
- š Real-time component validation against design specs
- āæ Accessibility compliance checking (WCAG 2.1)
- š Pattern analysis with scoring (0-100)
- šØ Anti-pattern detection and warnings
- ļæ½ Actionable improvement suggestions
š Quick Start
Prerequisites
- Node.js 18+
- npm or yarn
- MCP-compatible client
Installation & Setup
# Clone or download the server
cd fluentui-mcp-server
# Install dependencies
npm install
# Build the server
npm run build
# Test the server
npm run test
Running the Server
# Start the MCP server
npm start
# Or run in development mode with hot reload
npm run dev
š§ Available Tools
The server provides 12 powerful tools organized into 4 categories:
š Component Knowledge Tools
get_component_info
Get comprehensive information about any Fluent UI component.
{
"name": "get_component_info",
"arguments": {
"componentName": "Button"
}
}
Returns: Complete component documentation including description, props, slots, examples, and accessibility guidelines.
search_components
Find components based on functionality, category, or keywords.
{
"name": "search_components",
"arguments": {
"query": "form input",
"category": "inputs",
"limit": 5
}
}
get_component_props
Get detailed property information for a specific component.
{
"name": "get_component_props",
"arguments": {
"componentName": "Avatar",
"propName": "size"
}
}
get_component_examples
Retrieve usage examples for components.
{
"name": "get_component_examples",
"arguments": {
"componentName": "Card",
"exampleType": "advanced"
}
}
šØ Design System Tools
get_design_tokens
Access design tokens by category with usage guidance.
{
"name": "get_design_tokens",
"arguments": {
"category": "colors",
"includeUsage": true
}
}
Categories: colors, typography, spacing, shadows, borders, motion
validate_design_tokens
Check code for proper token usage and suggest improvements.
{
"name": "validate_design_tokens",
"arguments": {
"code": "const styles = makeStyles({ color: '#0078d4' });"
}
}
š ļø Code Generation Tools
generate_component
Generate complete component implementations with TypeScript.
{
"name": "generate_component",
"arguments": {
"componentName": "CustomCard",
"componentType": "layout",
"requirements": "A card component for user profiles with avatar and bio"
}
}
Component Types: form-input, layout, feedback, navigation
generate_component_hook
Create custom hooks following Fluent UI patterns.
{
"name": "generate_component_hook",
"arguments": {
"componentName": "CustomButton",
"hookType": "state-management"
}
}
generate_component_styles
Generate makeStyles implementations with design tokens.
{
"name": "generate_component_styles",
"arguments": {
"componentName": "CustomCard",
"slots": ["root", "header", "content", "footer"]
}
}
ā Validation & Analysis Tools
validate_component_design
Comprehensive design validation with scoring.
{
"name": "validate_component_design",
"arguments": {
"componentCode": "const Button = () => { ... }",
"componentName": "Button"
}
}
Returns: Validation score (0-100), issues found, and improvement suggestions.
check_accessibility
Analyze components for accessibility compliance.
{
"name": "check_accessibility",
"arguments": {
"componentCode": "const Component = () => { ... }"
}
}
analyze_component_patterns
Detect patterns and anti-patterns in component code.
{
"name": "analyze_component_patterns",
"arguments": {
"componentCode": "const Component = () => { ... }"
}
}
š¦ Component Database
The server includes comprehensive knowledge for 50+ Fluent UI v9 components across all categories:
š¤ Input Components (15+)
- Button (+ variants: CompoundButton, MenuButton, SplitButton, ToggleButton)
- Checkbox, Radio, Switch - Selection controls
- Input, Textarea, SpinButton - Text input components
- Combobox, Dropdown, Select - Selection components
- Slider, Rating - Value input components
- SearchBox, TagPicker - Advanced input patterns
š Data Display Components (12+)
- Avatar, AvatarGroup - User representation
- Badge, CounterBadge, PresenceBadge - Status indicators
- Text, Label - Typography components
- Image, Skeleton - Media and loading states
- List, DataGrid, Table - Data presentation
- Tree, Accordion - Hierarchical data
šļø Layout & Surface Components (8+)
- Card (+ variants: CardHeader, CardPreview, CardFooter)
- Divider, Drawer - Space division
- Dialog, Modal - Overlay components
- Field, FieldGroup - Form layout
š§ Navigation Components (8+)
- Breadcrumb, Link - Navigation aids
- Menu, MenuList - Action menus
- Nav, TabList - Primary navigation
- Toolbar, CommandBar - Action toolbars
š¬ Feedback Components (10+)
- MessageBar, InfoBar - System messages
- Toast, Notification - Temporary feedback
- Popover, Tooltip - Contextual information
- TeachingPopover - Educational overlays
- ProgressBar, Spinner - Loading states
šØ Design Token System
Access to the complete Fluent UI design token system with 45+ tokens across 6 categories:
š Colors (8 tokens)
colorNeutralForeground1; // #242424 - Primary text
colorNeutralBackground1; // #ffffff - Default background
colorBrandBackground; // #0078d4 - Primary brand
colorStatusSuccessBackground1; // Success states
colorStatusErrorBackground1; // Error states
// ... and more
š Typography (9 tokens)
fontFamilyBase; // 'Segoe UI', system fonts
fontSizeBase100; // 10px
fontSizeBase200; // 12px
fontSizeBase300; // 14px
fontWeightRegular; // 400
fontWeightSemibold; // 600
// ... complete scale
š Spacing (11 tokens)
spacingHorizontalXS; // 2px
spacingHorizontalS; // 4px
spacingHorizontalM; // 8px
spacingHorizontalL; // 12px
spacingHorizontalXL; // 16px
// ... full spacing scale
š«ļø Shadows (4 tokens)
shadow2; // Subtle elevation
shadow4; // Card elevation
shadow8; // Dialog elevation
shadow16; // Maximum elevation
š² Borders (7 tokens)
borderRadiusNone; // 0px
borderRadiusSmall; // 2px
borderRadiusMedium; // 4px
borderRadiusLarge; // 6px
strokeWidthThin; // 1px
strokeWidthThick; // 2px
š¬ Motion (6 tokens)
durationUltraFast; // 50ms
durationFaster; // 100ms
durationFast; // 150ms
durationNormal; // 200ms
curveAccelerateMax; // Easing curves
curveDecelerateMax;
š” Usage Examples
Example 1: Getting Component Information
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "get_component_info", "arguments": {"componentName": "Button"}}}' | node dist/server.js
Example 2: Generating a Custom Component
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "generate_component", "arguments": {"componentName": "UserCard", "componentType": "layout", "requirements": "A card showing user avatar, name, and role"}}}' | node dist/server.js
Example 3: Validating Component Design
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "validate_component_design", "arguments": {"componentCode": "const Button = () => <button>Click</button>", "componentName": "Button"}}}' | node dist/server.js
šļø Architecture
Server Structure
src/
āāā server.ts # Main MCP server implementation
āāā types.ts # TypeScript interfaces
āāā tools/
āāā component-tools.ts # Component knowledge & search
āāā design-tools.ts # Design tokens & validation
āāā generation-tools.ts # Code generation templates
āāā validation-tools.ts # Design & accessibility validation
data/
āāā components.json # Component database
āāā patterns.json # Design patterns & templates
Key Technologies
- MCP SDK: Model Context Protocol implementation
- TypeScript: Full type safety and IntelliSense
- Node.js: Runtime environment
- JSON-RPC 2.0: Communication protocol
š§ Development
Development Workflow
# Install dependencies
npm install
# Development mode with hot reload
npm run dev
# Build for production
npm run build
# Run tests
npm test
# Lint code
npm run lint
# Type checking
npm run type-check
Project Scripts
npm start- Start the MCP servernpm run dev- Development mode with tsxnpm run build- TypeScript compilationnpm test- Test server functionalitynpm run lint- ESLint code checking
Testing the Server
# Test that server starts and lists tools
npm run test
# Manual testing with curl/echo
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list", "params": {}}' | node dist/server.js
š¤ Integration
MCP Client Integration
This server is compatible with any MCP client. Here's how to integrate:
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"fluentui": {
"command": "node",
"args": ["/path/to/fluentui-mcp-server/dist/server.js"]
}
}
}
Custom MCP Client
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";
const client = new Client(
{
name: "my-app",
version: "1.0.0",
},
{
capabilities: {},
}
);
const transport = new StdioClientTransport({
command: "node",
args: ["./dist/server.js"],
});
await client.connect(transport);
š Validation Rules
The server implements comprehensive validation rules:
Design Validation
- ā Proper use of design tokens
- ā Consistent spacing and typography
- ā Color contrast compliance
- ā Component composition patterns
Accessibility Validation
- ā ARIA attributes and roles
- ā Keyboard navigation support
- ā Focus management
- ā Screen reader compatibility
- ā Color contrast ratios (WCAG 2.1)
Code Quality Validation
- ā TypeScript type safety
- ā Fluent UI v9 patterns
- ā Performance best practices
- ā Component composition
š Contributing
We welcome contributions! Here's how to get started:
1. Fork & Clone
git clone https://github.com/yourusername/fluentui-mcp-server.git
cd fluentui-mcp-server
2. Create Feature Branch
git checkout -b feature/new-component-support
3. Make Changes
- Add new components to
data/components.json - Implement new tools in
src/tools/ - Add tests for new functionality
- Update documentation
4. Test Changes
npm run build
npm test
npm run lint
5. Submit Pull Request
- Ensure all tests pass
- Add documentation for new features
- Follow existing code style
Areas for Contribution
- š New Components: Add more Fluent UI components
- š ļø Enhanced Tools: Improve existing tool capabilities
- š Better Validation: Add more validation rules
- šØ Design Patterns: Expand pattern library
- š Documentation: Improve guides and examples
š License
MIT License - see file for details.
š Related Resources
- š Fluent UI React Components - Official documentation
- šØ Design Tokens - Token system guide
- šļø Component Implementation Guide - Development patterns
- š Migration Guide - v8 to v9 migration
- š Model Context Protocol - MCP specification
- š¤ Claude Desktop - MCP-compatible AI assistant
Built with ā¤ļø for the Fluent UI community
Made with Model Context Protocol ⢠Fluent UI v9 ⢠TypeScript