Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@SSE Calculator MCP Serveradd 15 and 27, then multiply the result by 3"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
SSE Calculator MCP Server
A stateful calculator server that uses Server-Sent Events (SSE) for real-time communication. This server implements the Model Context Protocol (MCP) for integration with AI assistants.
Features
Real-time calculator operations using SSE
Stateful sessions for ongoing calculations
History tracking for all calculations
Web-based calculator interface
MCP integration for AI assistants
Architecture
The server is built using NestJS and implements the following components:
Session Service: Manages calculator sessions and state
Calculator Service: Performs calculation operations
SSE Controller: Handles client connections and event streaming
Getting Started
Prerequisites
Node.js (v14 or higher)
npm (v6 or higher)
Installation
Clone the repository
Install dependencies:
npm installStart the server:
npm run start:devOpen the calculator interface in your browser:
http://localhost:3000API Endpoints
SSE Stream
GET /calculator/streamEstablishes an SSE connection and creates a new calculator session.
Calculate
POST /calculator/calculate/:sessionIdPerforms a calculation operation on the specified session.
Request body:
{
"operation": "add", // add, subtract, multiply, divide, clear
"a": 5,
"b": 3
}History
GET /calculator/history/:sessionIdRetrieves the calculation history for the specified session.
Sessions
GET /calculator/sessionsRetrieves all active calculator sessions.
MCP Integration
To use this server as an MCP server, add the following configuration to your MCP settings file:
{
"mcpServers": {
"remote-calculator": {
"url": "http://localhost:3000/calculator/stream",
"disabled": false,
"alwaysAllow": [],
"timeout": 60
}
}
}Deployment Options
For production deployment, you can use the following options:
Docker: A Dockerfile is provided for containerized deployment
See
Dockerfileanddocker-compose.ymlfor configuration
Smithery.ai: Deploy directly from GitHub
See
SMITHERY_DEPLOYMENT.mdfor detailed instructionsConfiguration provided in
smithery.json
Heroku/DigitalOcean: Cloud platform deployment
See
DEPLOYMENT.mdfor step-by-step instructions
Self-hosted: Use NGINX or similar for SSL termination and load balancing
Hosting Your Own MCP Server
After deploying your server, update your MCP client configuration to point to your hosted URL:
{
"mcpServers": {
"remote-calculator": {
"url": "https://your-deployed-url.com/calculator/stream",
"disabled": false,
"alwaysAllow": [],
"timeout": 60
}
}
}License
MIT