Integrating An Existing MCP With LangChain

The plan for this section is straightforward: we take the two tools from the previous chapter and move them into a FastMCP server. The agent no longer defines tools itself — it asks MultiServerMCPClient to fetch them from the running server, then passes them to create_agent exactly as before. Everything else — memory, summarization, human-in-the-loop — stays the same.

We will end up with two files:

server.py hosts the tools as MCP tools.
agent.py connects to that server and runs the conversation loop.

Step 1: Create the Project

mkdir -p $HOME/workspace/langchain/langchain_agent_with_mcp
cd $HOME/workspace/langchain/langchain_agent_with_mcp

uv init --bare --python 3.12

Step 2: Install Dependencies

Two packages are new compared to the previous chapter:

fastmcp is the framework we use to write the MCP server.
langchain-mcp-adapters is the official bridge that converts MCP tool definitions into LangChain BaseTool instances the agent can call.

uv add \
    "fastmcp==3.0.2" \
    "httpx==0.28.1" \
    "langchain==1.2.10" \
    "langchain-mcp-adapters==0.2.1" \
    "langchain-openai==1.1.10" \
    "langgraph==1.0.9" \
    "python-dotenv==1.2.1"

Step 3: Add Your API Key

cat > .env <
OPENAI_API_KEY=your_openai_api_key_here
EOF

Step 4: Write the MCP Server

Create server.py. The tool logic is identical to the previous chapter — the only difference is the decorator. Instead of LangChain's @tool, we use FastMCP's @mcp.tool(). From the agent's perspective the tools look exactly the same; the MCP protocol handles the translation.

Imports and Server Instance

import httpx
from fastmcp import FastMCP

mcp = FastMCP("Weather & Air Quality")

FastMCP("Weather & Air Quality") creates the server and gives it a human-readable name that appears in the MCP handshake.

Coordinate Helper and Tools

The _get_coordinates helper and the two tool functions are unchanged from the previous chapter. The only edit is swapping @tool for @mcp.tool().

def _get_coordinates(location: str) -> tuple[float, float]:
    """Resolve a place name to (latitude, longitude) via the Open-Meteo Geocoding API."""
    response = httpx.get(
        "https://geocoding-api.open-meteo.com/v1/search",
        params={"name": location, "count": 1, "language": "en", "format": "json"},
    )
    data = response.json()
    if "results" in data and len(data["results"]) > 0:
        latitude = data["results"][0]["latitude"]
        longitude = data["results"][0]["longitude"]
        return latitude, longitude
    else:
        raise ValueError(f"Could not find coordinates for location: {location}")


@mcp.tool()
def get_air_quality(location: str) -> str:
    """Get air quality information based on a location."""
    latitude, longitude = _get_coordinates(location)
    response = httpx.get(
        "https://air-quality-api.open-meteo.com/v1/air-quality",
        params={
            "latitude": latitude,
            "longitude": longitude,
            "hourly": "pm10,pm2_5",
            "forecast_days": 1,
        },
    )
    data = response.json()
    if "hourly" in data and "pm10" in data["hourly"] and "pm2_5" in data["hourly"]:
        pm10 = data["hourly"]["pm10"][0]
        pm2_5 = data["hourly"]["pm2_5"][0]
        result = f"PM10: {pm10} μg/m³, PM2.5: {pm2_5} μg/m³"
    else:
        result = "Air quality data not available"
    return f"Air quality in {location}: {result}"


@mcp.tool()
def get_temperature(location: str) -> str:
    """Get the current temperature for a location."""
    latitude, longitude = _get_coordinates(location)
    response = httpx.get(
        "https://api.open-meteo.com/v1/forecast",
        params={
            "latitude": latitude,
            "longitude": longitude,
            "hourly": "temperature_2m",
            "forecast_days": 1,
        },
    )
    data = response.json()
    if "hourly" in data and "temperature_2m" in data["hourly"]:
        temperature = data["hourly"]["temperature_2m"][0]
        result = f"Temperature: {temperature} °C"
    else:
        result = "Temperature data not available"
    return f"Temperature in {location}: {result}"

Entry Point

We used the HTTP transport in most if not all previous servers, let's switch to stdio for this one for the sake of variety.

if __name__ == "__main__":
    mcp.run(transport="stdio", show_banner=False)

transport="stdio" tells FastMCP to communicate over standard input/output. The agent will launch this script as a subprocess and talk to it over stdio — no network port needed.
show_banner=False suppresses the startup log so the stdio stream stays clean.

Complete Server Code

cat > $HOME/workspace/langchain/langchain_agent_with_mcp/server.py <<EOF
# server.py
import httpx
from fastmcp import FastMCP

mcp = FastMCP("Weather & Air Quality")


def _get_coordinates(location: str) -> tuple[float, float]:
    """Resolve a place name to (latitude, longitude) via the Open-Meteo Geocoding API."""
    response = httpx.get(
        "https://geocoding-api.open-meteo.com/v1/search",
        params={"name": location, "count": 1, "language": "en", "format": "json"},
    )
    data = response.json()
    if "results" in data and len(data["results"]) > 0:
        latitude = data["results"][0]["latitude"]
        longitude = data["results"][0]["longitude"]
        return latitude, longitude
    else:
        raise ValueError(f"Could not find coordinates for location: {location}")


@mcp.tool()
def get_air_quality(location: str) -> str:
    """Get air quality information based on a location."""
    latitude, longitude = _get_coordinates(location)
    response = httpx.get(
        "https://air-quality-api.open-meteo.com/v1/air-quality",
        params={
            "latitude": latitude,
            "longitude": longitude,
            "hourly": "pm10,pm2_5",
            "forecast_days": 1,
        },
    )
    data = response.json()
    if "hourly" in data and "pm10" in data["hourly"] and "pm2_5" in data["hourly"]:
        pm10 = data["hourly"]["pm10"][0]
        pm2_5 = data["hourly"]["pm2_5"][0]
        result = f"PM10: {pm10} μg/m³, PM2.5: {pm2_5} μg/m³"
    else:
        result = "Air quality data not available"
    return f"Air quality in {location}: {result}"


@mcp.tool()
def get_temperature(location: str) -> str:
    """Get the current temperature for a location."""
    latitude, longitude = _get_coordinates(location)
    response = httpx.get(
        "https://api.open-meteo.com/v1/forecast",
        params={
            "latitude": latitude,
            "longitude": longitude,
            "hourly":

Practical MCP with FastMCP & LangChain

Engineering the Agentic Experience

Enroll now to unlock current content and receive all future updates for free. Your purchase supports the author and fuels the creation of more exciting content. Act fast, as the price will rise as the course nears completion!

Unlock now $26.99 Learn More

Previous Next