Silk currently creates a 'conf' folder in the deployed site to house the nginx.conf and supervisord.conf includes.
For nginx that's already changing, as the zero-downtime fix means that we'll have multiple copies of the site in /srv at a time, and two or more of them running under supervisord, but nginx should point to only one. This implies that it would be better to put that one nginx include in /etc/nginx/conf.d/ or /etc/nginx/sites-enabled (probably the latter).
While we're doing that, it may also make sense to use /etc/supervisor/conf.d/ for the supervisord includes. The benefit to doing this is that we would no longer need special versions of the nginx or supervisord master config files (to include /srv/<site>/conf/). As long as you have nginx and supervisord installed with Ubuntu's default configuration, we'd be good to go. I think that's a win.