mcd-demo

Testing creation of simple MCP servers and integrating with LangChain agent

Prerequisites

Create a virtual environment

python3 -m venv venv
source venv/bin/activate

Install dependencies

pip install -r requirements.txt

Set Environment variables

Set the following environment variables. Get the values from Azure AI Foundry where the models are deployed:

export AZURE_OPENAI_API_KEY=<your_azure_openai_api_key>
export AZURE_OPENAI_ENDPOINT=<your_azure_openai_endpoint>

Optionally, you can set the following environment variables to configure the MCP servers:

export MCP_MATH_URI=http://<server-uri>:5001/sse

Note: You can also set these variables in a .env file in the root directory of the project.

Running the agent

Start the MCP servers

You have to start all three MCP servers before starting the agent. Each server listens on a separate port. You can start them in separate terminals or run them in the background. To run in the background, do the following:

python weather_server.py &
python math_server.py &
python telemetry_server.py &

Alternatively, you can run the math server in a Docker container. To do this, first build the Docker image:

make build

Then, run the container:

make run-local

If you want to push the Docker image to a registry, tag and push it using the following commands:

make login
make push

Start the agent

python agent.py

Killing the MCP servers

pkill -9 -f weather_server.py
pkill -9 -f math_server.py
pkill -9 -f telemetry_server.py