Running any project can be as easy as:

$ pod run

By wrapping Podman in a command-line interface that gets out of the way, you can easily containerise your development environment, removing dependency conflicts, and making managing a virtual environment unnecessary.

pod simplifies the Podman CLI by defining targets to build and run in the pods.yaml file, so you never forget to include a flag. It still exposes the whole Podman CLI, so you don’t have to learn a huge new set of options and how they map back to Podman. It can be as simple as:

# in pods.yaml
    name: pod-example
    interactive: true
      - sh
# with the pod cli
$ pod run alpine-shell

pod scales up in complexity to add ports, mounts, and any other Podman feature1. Once you’ve built the image you want to run in “production”2, you can deploy it:

$ pod update -d
update: pods-website (arguments changed)
-   --publish=4300:80
+   --publish=4301:80
Container started at 2023-06-08 08:06:19 UTC (up 42m)
update? [y/N] y

pod will compare the configs of your running containers and the config in pods.yaml, and update anything that has changed.


pod can be installed by building an image with all the necessary dependencies in it. You might need to change the base image, depending on your Linux flavour.

$ git clone
$ cd pod
$ ./ ~/.local/bin
... installer will run and copy executable

Quick Usage

$ pod init
# Run through guided setup
$ pod build
$ pod run

See Getting Started for more detail, and get more info on the config file.


pod can also be used to run simple scripts without setting up a whole project, see pod script.

  1. Probably, I haven’t tested them all. 

  2. pod is not intended for 100% robust production setups. Use Kubernetes instead. I am not an expert.