In early integration work with the QI infrastructure, we have been copying sample Vagrantfile content from a central repo, for example, this one:
As the intent seems to be for people to put the changeable settings in .qi.yml, it seems like most people shouldn't need a local Vagrantfile at all.
For local testing, there are options for configuring vagrant to read its config from an alternate location. They could add the package containing the "stock" vagrantfile to their dev dependencies and use if from within node_modules.
In the QI environment, you have total control over the path to the Vagrantfile. You could keep the current behavior with some smart failover options, for example:
1. If there's a Vagrantfile in the root of the repo, use that.
2. If not, check to see if they have the QI package in their dev deps and use that Vagrantfile.
3. If not, use the stock file from the QI environment.
Something like that strikes a balance between letting people with complex needs change behavior and avoiding duplicated "stock" code that will quickly get out of date.
cc: Antranig Basman