Calculator MCP Server
Exposes add, subtract, multiply, and divide as MCP tools, implemented with the FastMCP SDK and a raw stdio version.
README
Calculator MCP Server
A minimal MCP server exposing four tools —
add, subtract, multiply, divide — in two versions:
calculator_server.py— built with the PythonmcpSDK (FastMCP).@mcp.tool()derives the JSON schema from type hints/docstrings and handles the JSON-RPC protocol for you.calculator_server_raw.py— the same server implemented from scratch with only the standard library (json+sys), nomcpdependency. It hand-rolls the JSON-RPC 2.0 message handling to show what FastMCP is doing underneath: read a newline-delimited JSON request from stdin, dispatch onmethod(initialize,tools/list,tools/call,ping), write a JSON-RPC response to stdout. See the file's top-of-file docstring for the framing details (no LSP-style headers, just one JSON object per line).
Setup
python -m venv .venv
.venv\Scripts\pip install -r requirements.txt
Run it directly (stdio)
MCP servers talk over stdio (or HTTP), not a normal CLI — you don't "run and see output," a client connects to it. Two ways to try it:
1. MCP Inspector (visual, interactive) — requires Node.js/npx:
.venv\Scripts\mcp dev calculator_server.py
Opens a browser UI where you can call add, multiply, etc. by hand.
If the page loads but says "Connect to an MCP server to start inspecting" and there's no Tools tab, the sidebar didn't pick up the command automatically (can happen if a stale/previous session config is cached). Fill in the sidebar manually and click Connect:
- Transport Type:
STDIO - Command:
C:\Users\simon\mcpserversample\.venv\Scripts\python.exe - Arguments:
C:\Users\simon\mcpserversample\calculator_server.py
Once connected, a Tools tab appears in the top nav where you can click "List Tools" and run each one.
2. test_client.py — a scripted client included in this repo that
connects, lists tools, and calls each one. Defaults to calculator_server.py,
or pass a path to test the other version:
.venv\Scripts\python test_client.py
.venv\Scripts\python test_client.py calculator_server_raw.py
Wiring it into Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"calculator": {
"command": "C:\\Users\\simon\\mcpserversample\\.venv\\Scripts\\python.exe",
"args": ["C:\\Users\\simon\\mcpserversample\\calculator_server.py"]
}
}
}
How it works
calculator_server.pycreates aFastMCPserver and registers each Python function as a tool via the@mcp.tool()decorator. The function's type hints and docstring become the tool's JSON schema and description, which is what the calling client (Claude, the Inspector, ortest_client.py) sees.mcp.run()starts the server listening on stdin/stdout using the MCP stdio transport — the standard way a client launches and talks to a local MCP server as a subprocess.- A client connects, calls
list_tools()to discover what's available, thencall_tool(name, args)to invoke one and get the result back.
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.