Agentic AI – MCP Custom Server Implementation
Table Of Contents:
- MCP Server Implementation?
- Math MCP Server Implementation.
- Weather MCP Server Implementation.
- MultiServer MCP Client + LangChain Agent.
- Run The Workflow.
(1) MCP Server Implementation
(2) Math MCP Server Implementation
# math_server.py
from fastmcp import FastMCP
mcp = FastMCP("Math")
@mcp.tool()
def add(a: int, b:int)->int:
"""Add Two Numbers"""
return a * b
@mcp.tool()
def multiply(a:int, b:int)-> int:
"""Multiply Two Numbers"""
return a * b
if __name__ ==== "__main__":
mcp.run(transport="stdio")(3) Weather MCP Server Implementation
# weather_server.py
from fastamcp import FastMCP
mcp = FastMCP("Weather")
@mcp.tool()
async def get_weather(location: str) -> str:
"""Get Weather For Location"""
return f"Its always Sunny in {location}"
if __name__ == "__main__":
mcp.run(transport = "streamable-http")(4) MultiServer MCP Client + LangChain Agent
# client.py
import asyncio
from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.agents import create_agent
async def main():
# Connect to both servers
client = MultiServerMCPClient(
{
"math": {
"transport": "stdio",
"command": "python",
"args": ["/absolute/path/to/math_server.py"],
},
"weather": {
"transport": "http",
"url": "http://localhost:8000/mcp", # Weather server endpoint
}
}
)
# Discover tools from servers
tools = await client.get_tools()
# Create agent with Claude Sonnet + MCP tools
agent = create_agent(
"claude-sonnet-4-6",
tools
)
# Example 1: Math query
math_response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "what's (3 + 5) x 12?"}]}
)
# Example 2: Weather query
weather_response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "what is the weather in nyc?"}]}
)
print("Math Response:", math_response)
print("Weather Response:", weather_response)
# Close connections
await client.close()
if __name__ == "__main__":
asyncio.run(main())
(5) Run The Workflow.
