microsoft/building-ai-agents-from-zero-to-production

Python

8 environment variables found in .env.example

Set these on your host before deploying. Each platform has an “Environment Variables” section in the deploy flow.

FOUNDRY_PROJECT_ENDPOINT Your Microsoft Foundry PROJECT endpoint (note: the /api/projects/<project> path) Example:…
FOUNDRY_MODEL The model deployment name the agents run on. Use a current (non-retired) model such as gpt-5.1, gpt-5-mini, or gpt-5-codex. Do NOT use…
VECTOR_STORE_ID Vector Store ID (run `python setup_vector_store.py` first to get this value)
GITHUB_PERSONAL_ACCESS_TOKEN secret GitHub Personal Access Token (only for the optional task-recommendation agent) Create at…
TOOLBOX_ENDPOINT Lesson 6 (Toolbox): the toolbox CONSUMER MCP endpoint (only for toolbox_agent.py). Create a toolbox first, then set this to:…
A2A_HOST Lesson 7 (Agent-to-Agent): host/port the A2A server binds to, and the URL the A2A client calls. Defaults work for a local demo; override…
A2A_PORT
A2A_SERVER_URL

Best match Fly.io Anything Dockerized — requires the Fly CLI.

macOS

brew install flyctl && fly launch --from https://github.com/microsoft/building-ai-agents-from-zero-to-production

Linux

curl -L https://fly.io/install.sh | sh && fly launch --from https://github.com/microsoft/building-ai-agents-from-zero-to-production

Windows (PowerShell)

iwr https://fly.io/install.ps1 -useb | iex; fly launch --from https://github.com/microsoft/building-ai-agents-from-zero-to-production

Or deploy to a different host