It is possible using the `devcontainer.json` file which is inside the `.devcontainer` folder.

You can fill the property `extensions` with an array of extension IDs that specify the extensions that should be installed inside the container when it is created.

You can find an example in the formation I made to use IRIS and create a production in full Python here https://github.com/LucasEnard/formation-template-python/blob/main/.devco... at the end of the file.

Exactly, for instance in Python, if you wish to use this client outside the container you need to `pip install fhirpy` and `pip install fhir.resources` on your own computer.

Using the container, the requierments.txt file will be read by docker which will call `pip install requirements.txt` inside the container and install the right modules inside the container and not on your computer.
 

Same goes for java and c# !

Thank you for your feedback!

From what I know, you can use it locally without docker at all or use it with docker-compose and mapping if you wish to.
I did it like this because I wanted it to work without the need for the user to install anything on their computer (as I've made a c# and a java version too, it seemed easier to preinstall everything to have a clone and go github).

You can use VSCode with IRIS and IRIS Python / Embedded Python, for that I recommend you to check https://github.com/grongierisc/interoperability-embedded-python which is a proof of concept aiming to show how the iris interoperability framework can be used with embedded python while coding inside a container.

If you have any more questions don't hesitate !