I'm building a method to use Twitter (via API) as a configuration management repository, storing encrypted configuration files in twitter, then using API calls

I'm building a method to use Twitter (via API) as a configuration management repository, storing encrypted configuration files in twitter, then using API calls from new network clients to grab the role-appropriate configuration files for a given service. So, "puppet, but with Twitter as the storage component". It initially grew out of a test to build a shell-based encrypted method for posting and reading tweets, compounded by an underlying desire to figure out how to make twitter "useful" for me somehow. Discovering that stego data does not get stripped was an eye opener on the size of data chunks that could then be stored as tweets, which allowed me to push larger configuration files, while still having them be largely useless to random viewers.