Fabkit MCP Server
Provides AI agents with comprehensive documentation and code patterns for the Fabkit Svelte 5 UI library. It enables users to search for components, icons, and theming APIs to ensure accurate implementation and prevent hallucinations.
README
Fabkit MCP Server
A Model Context Protocol server for the Fabkit Svelte 5 UI library, deployed as a Cloudflare Worker.
Agents (Claude, Cursor, Copilot…) can call this server to get accurate, up-to-date Fabkit documentation without hallucinating component names, props, or icon names.
Tools exposed
| Tool | Description |
|---|---|
list_components |
All components grouped by category, with optional filter |
get_component |
Full docs for a single component (props, examples, notes) |
search_components |
Full-text search across all component docs |
get_theming |
Complete theming API: initTheme, CSS variables, dark mode |
search_icons |
Search 1500+ Phosphor icons by keyword → exact Ph* names |
get_pattern |
Ready-to-use Svelte code patterns (app-shell, dashboard, …) |
list_patterns |
List all available patterns |
Quick start
1. Install dependencies
npm install
2. Run locally
npm run dev
# → http://localhost:5173
3. Deploy to Cloudflare Workers
npm run deploy
# → https://fabkit-mcp.fabricators.dev
Connecting to Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"fabkit": {
"url": "https://fabkit-mcp.fabricators.dev/mcp",
"transport": "http"
}
}
}
Connecting to Cursor
In .cursor/mcp.json:
{
"mcpServers": {
"fabkit": {
"url": "https://fabkit-mcp.fabricators.dev/mcp"
}
}
}
Connecting to Gemini CLI
In ~/.gemini/settings.json:
{
"mcpServers": {
"fabkit": {
"url": "https://fabkit-mcp.fabricators.dev/mcp"
}
}
}
Note: Gemini CLI does not support the
transportkey — use onlyurl.
Connecting to GitHub Copilot CLI
In ~/.copilot/mcp-config.json:
{
"mcpServers": {
"fabkit": {
"type": "http",
"url": "https://fabkit-mcp.fabricators.dev/mcp",
"tools": ["*"]
}
}
}
Alternatively, run /mcp add inside the CLI and fill in the fields interactively.
Connecting to VS Code
In .vscode/mcp.json (workspace) or ~/.config/Code/User/mcp.json (global):
{
"servers": {
"fabkit": {
"type": "http",
"url": "https://fabkit-mcp.fabricators.dev/mcp"
}
}
}
Note: VS Code uses
"servers"(not"mcpServers") as the root key.
Connecting to Continue.dev
In ~/.continue/config.json:
{
"mcpServers": [
{
"name": "fabkit",
"type": "streamable-http",
"url": "https://fabkit-mcp.fabricators.dev/mcp"
}
]
}
Connecting to Windsurf
In ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"fabkit": {
"serverUrl": "https://fabkit-mcp.fabricators.dev/mcp"
}
}
}
Note: Windsurf uses
"serverUrl"(not"url") for remote HTTP servers.
MCP Protocol
The server implements MCP 2024-11-05 over Streamable HTTP (POST /mcp).
All requests are JSON-RPC 2.0:
# Initialize
curl -X POST https://fabkit-mcp.fabricators.dev/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{}},"id":1}'
# List tools
curl -X POST https://fabkit-mcp.fabricators.dev/mcp \
-d '{"jsonrpc":"2.0","method":"tools/list","id":2}'
# Call a tool
curl -X POST https://fabkit-mcp.fabricators.dev/mcp \
-d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"get_component","arguments":{"name":"Button"}},"id":3}'
Extending the knowledge base
Edit src/knowledge.ts — add entries to the COMPONENTS array or extend ALL_ICONS.
Edit src/tools.ts — add new entries to the PATTERNS object.
Then redeploy: npm run deploy.
Architecture
src/
index.ts ← Cloudflare Worker entry, routing, CORS
mcp.ts ← JSON-RPC 2.0 dispatch, tool registry
knowledge.ts ← All Fabkit docs embedded as typed data
tools.ts ← Tool implementations (pure functions)
Everything is stateless — no KV, no R2, no external APIs.
The Worker cold-starts in < 5ms because all knowledge is bundled inline.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.