[Template] MCP Server
Description
# [Template] MCP Server ## How to use this template - Add tool definitions in src/index.ts - Update the README below - Update the placeholder info in package.json - Test building the MCP server with `npm run build`, and using the @modelcontextprotocol/inspector library - Publish…
About
# [Template] MCP Server ## How to use this template - Add tool definitions in src/index.ts - Update the README below - Update the placeholder info in package.json - Test building the MCP server with `npm run build`, and using the @modelcontextprotocol/inspector library - Publish to npm: `npm login`, `npm publish…
Details
- Author
- larryhudson
- Downloads
- 264
- Categories
- Media
Jump to
- Template for creating custom MCP servers
- Instructions for setup with VS Code and Claude
- Environment variable configuration for API keys
- Built with TypeScript and the MCP framework
- Tool definitions added in a single file (src/index.ts)
- Supports publishing to npm
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
[Template] MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Developers add tool definitions in src/index.ts, update the placeholder information in package.json, and build the server with npm run build. They can test the server locally using the @modelcontextprotocol/inspector library. After testing, they publish the package to npm with npm login and npm publish --access public. To run the server, users set an API key as an environment variable (API_KEY_ENV_VAR) and add the MCP configuration to their client (e.g., VS Code or Claude).
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"[template] mcp server": {
"figma-mcp-server-larryhudson": {
"command": "npx",
"args": [
"@modelcontextprotocol/inspector",
"npx",
"tsx",
"src/index.ts"
]
}
}
}
}McpServers
{
"figma-mcp-server-larryhudson": {
"command": "npx",
"args": [
"@modelcontextprotocol/inspector",
"npx",
"tsx",
"src/index.ts"
]
}
}[Template] MCP Server
How to use this template
- Add tool definitions in src/index.ts
- Update the README below
- Update the placeholder info in package.json
- Test building the MCP server with npm run build, and using the @modelcontextprotocol/inspector library
- Publish to npm: npm login, npm publish --access public
Overview
[Insert overview here]
Features
[Insert features here]
Setup
1. Create a API key etc.
2. Set the token as an environment variable:
export API_KEY_ENV_VAR=the_secret_api_key
Usage
Using with VS Code
Add this to your settings JSON file:
{
"mcp": {
"inputs": [
{
"type": "promptString",
"id": "some_secret_api_key",
"description": "Secret API key",
"password": true
}
],
"servers": {
"[library-name]": {
"command": "npx",
"args": [
"-y",
"@larryhudson/[the-library]"
],
"env": {
"API_KEY_ENV_VAR": "${input:some_secret_api_key}"
}
}
}
}
}
Using with Claude or other MCP-compatible applications
Add this to your MCP configuration JSON file:
{
"mcpServers": {
"[the-library]": {
"command": "npx",
"args": ["-y", "@larryhudson/[the-library]"],
"env": {
"API_KEY_ENV_VAR": "<THE_SECRET_API_KEY>"
}
}
}
}
Available Tools
[List the tools here]
How It Works
[Some info about how it works]
Technical Details
Built with:
- Model Context Protocol (MCP): Framework for allowing AI assistants to interact with external tools
- TypeScript: For type safety and better developer experience
Development
You can use the Model Context Protocol inspector to try out the server:
npx @modelcontextprotocol/inspector npx tsx src/index.ts
Limitations and Future Improvements
License
MIT
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
