As I understand it, Kloudless uses a middleware to convert the data. So data flows from your app trough the Kloudless cloud to e.g. Dropbox. Means that you really have to trust them. We are doing all the conversion in the client. No data ever flows via our servers and no third party can read your transferred files ;)
Congrats on the beta release! I'd suggest making the docs available without having to sign up.
I noticed the docs indicate unsupported features for each service, such as creating Dropbox links and downloading OneDrive files. A benefit to managing the integration is that you would be able to provide full functionality for all the APIs you support. This is one of the primary reasons we chose to adopt the architecture we currently have in place.
This would also let you fill in missing features such as Event Notifications and Search even if the underlying storage service doesn't provide it. For example, we support event notifications for SharePoint.
A side-effect of this is that developers would not have to worry about OAuth tokens expiring or having to be refreshed, since token management is handled for them. UI tools such as authenticator widgets and file explorers ( https://github.com/kloudless/file-explorer ) can also be
integrated pretty easily since authentication is managed.
As you pointed out, client-side solutions provide additional security guarantees. This is one of the reasons our enterprise plan features an on-premise version of Kloudless.
I'm excited to see more development in this space, and looking forward to your launch!
