"""
MCP Server Template
"""
# NOTE: This MCP server uses `requests` and `numpy` to call an external MATLAB
# production server and parse the returned matrix. Ensure `requests` and
# `numpy` are installed (they're listed in requirements.txt). To change the
# target MATLAB service, edit the MATLAB_SERVICE_URL variable inside
# `calculate_magic_matrix`.
from mcp.server.fastmcp import FastMCP
from pydantic import Field
from typing import Union
mcp = FastMCP("Echo Server", stateless_http=True)
@mcp.tool(
title="Echo Tool",
description="Echo the input text",
)
def echo(text: str = Field(description="The text to echo")) -> str:
return text
@mcp.tool(title="Magic Matrix", description="Calculate Magic Matrix")
def calculate_magic_matrix(in_param: Union[int, float]) -> dict:
"""Call the external MATLAB service to compute a magic matrix.
The function sends a JSON payload like {"nargout":1, "rhs": [in_param]}
to the /mymagic/mymagic endpoint and parses the MATLAB production server
response into a native Python structure. Returns a dict containing:
- magic_square: list[list[int]] (rows of the matrix)
- rows: int
- cols: int
- raw: original JSON response (for debugging)
"""
import json
import requests
import numpy as np
MATLAB_SERVICE_URL = "https://matlab-0j1h.onrender.com"
# Convert in_param to integer
try:
n = int(in_param)
except (ValueError, TypeError):
return {"error": f"Invalid parameter: {in_param}. Must be a number."}
payload = {"nargout": 1, "rhs": [n]}
headers = {"Content-Type": "application/json"}
resp = requests.post(f"{MATLAB_SERVICE_URL}/mymagic/mymagic", json=payload, headers=headers, timeout=10)
resp.raise_for_status()
j = resp.json()
# The MATLAB service returns a structure with lhs[0] containing mwdata and mwsize
try:
lhs = j.get("lhs", [])[0]
flat = lhs.get("mwdata")
rows, cols = lhs.get("mwsize", [None, None])
if flat is None or rows is None or cols is None:
# If the response doesn't match expected shape, return raw JSON
return {"magic_square": None, "rows": None, "cols": None, "raw": j}
# reshape into nested list for JSON-serializable output
arr = np.array(flat).reshape(rows, cols)
magic_list = arr.tolist()
return {"magic_square": magic_list, "rows": int(rows), "cols": int(cols), "raw": j}
except Exception as e:
# Fallback: return raw JSON on any parsing error
return {"magic_square": None, "rows": None, "cols": None, "raw": j, "error": str(e)}
@mcp.resource(
uri="greeting://{name}",
description="Get a personalized greeting",
name="Greeting Resource",
)
def get_greeting(
name: str,
) -> str:
return f"Hello, {name}!"
@mcp.prompt("")
def greet_user(
name: str = Field(description="The name of the person to greet"),
style: str = Field(description="The style of the greeting", default="friendly"),
) -> str:
"""Generate a greeting prompt"""
styles = {
"friendly": "Please write a warm, friendly greeting",
"formal": "Please write a formal, professional greeting",
"casual": "Please write a casual, relaxed greeting",
}
return f"{styles.get(style, styles['friendly'])} for someone named {name}."
if __name__ == "__main__":
mcp.run(transport="stdio")
