How do I use gpg-agent in --daemon mode?

I know that in this mode it starts as a network server (sort of like a network server, but intended just to be used on your own computer), but what port does it serve one? And what commands does it recognize when sent over TCPIP from a client? Does it recognize the same commands as when run with the command --server (which makes it work with stdin and stdout), just sent over TCPIP instead of via the console? Does it have an underlying protocol that is needed for TCPIP operation (sort of like how websites aren’t just raw HTML text, and instead they exist within an underlying protocol called HTTP)?

Please let me know. I’m trying to write my own program to communicate with GPG4win, to make my own GUI front end for it, as I’m not satisfied with Kleopatra’s operation (the fact that it leaves some exe files running, even after Kleopatra closes down, is a MAJOR security hole). My front end won’t be as fancy as Kleopatra in terms of its graphics, but it will be very good at what it does, and it won’t leave any security holes. Any info that I might need to create my own GUI, and the easiest mechanism to use for this inter-process communication (the TCPIP version or the command console version), please let me know.

Hi AD!

gpg-agent speaks the assuan protocol, the texinfo documentation coming with the source code of the components has it all documented.

If you believe that leaving applications running is a problem for you (in some other thread the threat-modelling was already mentioned), you could also try to come up with a patch or a script that allows user to terminate the processes. Of course you can also write your own! It is Free Software after all, so experiments, alternativer or any other development is encouraged. :slight_smile:

I suggest you hook up with the gnupg development community on gnupg-devel@ or other places.

Best Regards,
Bernhard
ps.: Flattr Gpg4win at https://flattr.com/thing/2053326,
if you appreciate this answer and my work within the Gpg4win Initiative.