Another unsolved feature in Wayland is ssh-agent/gpg-agent support. I haven't found a way to make it work yet. It's either multiple agents or many fresh agents running.
Why does Wayland need any sort of ssh-agent/gpg-agent support? I can't think of any reason for Wayland to have anything to do with their functionality, but I'm not at all familiar with Wayland.
The way this works in X11 is that you usually start your X11 env such that the environment variable pointing to the agent (which got started) is shared and the agent is accessible by applications. That way running ssh or gpg in any xterm will just work. This is not possible at the moment in Wayland. Think of it like OSX's keychain.
Huh, I'll try that. I never thought of it like that because due to conventions the agent was always part of xinit. Might need to come up with a different way of running wayland compositors to keep the actually TTY shell free of the agent.