The configuration file with links to repositories, descriptions, etc can be either a custom toml/json/yaml file or a traitlet (py/json file); the question here boils down to preferences on file format, and:
if we use a traitlet we can re-use the jupyter-server config files (py/json); the python files could read in the PAT token from environment variables and embed it in the repository URL used for cloning
if we use a custom file (toml/yaml/json), we will need to implement variable substitution
There needs to be a way to setup it via nebari config; if it is a traitlet, then maybe nebari should have a sibling entry to jupyterlab.default_settings but for the server-side settings.
The configuration file with links to repositories, descriptions, etc can be either a custom toml/json/yaml file or a traitlet (py/json file); the question here boils down to preferences on file format, and:
Originally posted in https://github.com/nebari-dev/nebari/issues/2456#issuecomment-2112053214
Note: It cannot be placed in a JupyterLab JSON overrides as these are client-side and would expose PATs