Server MCP Agent

Example Use Cases

• 1

  Perform mathematical operations via a local MCP tool.

• 2

  Retrieve weather information from a remote MCP server.

• 3

  That require structured communication with external tools using MCP protocol.

Description

This project demonstrates how to use the Model Context Protocol (MCP) to connect an AI agent based on LangChain and LangGraph with external tools, specifically math and weather services. It supports two transport modes: stdio for local tools and streamable_http for remote servers, enabling structured communication between the LLM and external services. The server also integrates observability features through LangSmith and Langfuse for tracing and analytics, enhancing monitoring and evaluation of the agent's interactions.