~15 MIN · DO THIS AT HOME

Prepare your laptop

A cold bootstrap takes about two minutes on a good connection. Thirty-five of them on conference Wi-Fi takes the entire first block, so we move that work to your kitchen table. Everything below is the whole ask.

00Before you start

You need a machine with at least 8 GB of free RAM and ~10 GB of free disk, and you need to be comfortable with kubectl — reading a manifest, reading a rollout. No machine-learning background is required; none of this is model training. Budget about fifteen minutes.

On a corporate laptop, check that Docker runs and that api.anthropic.com is reachable before you travel. Proxies and MDM are the two things we cannot debug for you in the room.

01Install the tools

Git, Docker (or Podman/Colima), kubectl, k3d, make and Python 3.11 or newer.

$ brew install --cask docker$ brew install kubectl k3d python@3.12$ open -a Docker # start it, and leave it running

02Clone the repository

Everything you run on the day lives here, including the Makefile that drives the labs.

$ git clone https://github.com/leffen/agentic-cloud-ops.git$ cd agentic-cloud-ops

03Install the Python dependencies

A virtual environment inside the repo. The eval harness and the docs build both use it.

$ python3 -m venv .venv$ source .venv/bin/activate$ pip install -r requirements.txt

04Run the preflight

This checks that every tool is present. Every line must say OK.

$ make preflight

$ make preflightexit 0

OKdocker 27.1runningOKkubectl 1.31foundOKk3d 5.7foundOKpython3 3.12foundOKdisk14.2 GB freeOKimagescached (3.8 GB)6/6checks passed.

Known gap: preflight currently checks for tools only — it does not yet pre-pull the container images, despite what its help text says. Until it does, run `make cluster-up` once at home as well, so the images land on disk while you still have good bandwidth.

05Agent credentials

The agent needs a model endpoint. An Anthropic API key is the best-tested path; OpenAI works. Ollama is not viable alongside a cluster on an 8 GB machine — do not plan on it. Workshop keys with a spend cap will be handed out on the day if you would rather not use your own.

$ cp .env.example .env$ $EDITOR .env # ANTHROPIC_API_KEY=sk-ant-...

.env is git-ignored. Never commit it, and use a key you can rotate afterwards.

--Troubleshooting

The five you are most likely to meet, in the form you will actually meet them.

Cannot connect to the Docker daemon at unix:///var/run/docker.sockDocker Desktop is installed but not running.$ Start Docker and wait for the whale to stop animating.

error: no configuration has been provided, try setting KUBERNETES_MASTERThe cluster is not up, or your kubectl context points somewhere else.$ make cluster-up && kubectl config current-context

nodes "k3d-agentic-ops-server-0" not readyNormal for the first ~30 seconds after cluster-up.$ kubectl get nodes -w # wait for Ready, then carry on

zsh: no matches found: agentic-ops:agent-hardenzsh expands $NS:agent-harden using the :a path modifier and mangles it.$ Always brace it: ${NS}:agent-harden

ImagePullBackOffIn Lab 1 this is deliberate — it is the fault the agent has to find.$ Leave it. That one is the exercise.

--Day-of checklist

☐Laptop charged, and a charger in the bag
☐make preflight says OK on every line
☐make cluster-up has been run at least once at home, so the images are cached
☐An API key in .env, or a plan to collect a workshop key on the day
☐~10 GB of disk still free after all of the above
☐A terminal you are comfortable in, and an editor

Stuck? The full documentation is at/docs/workshop/prerequisites, and we will be in the room 15 minutes early.