Name/Nym: Orthzar
Summary of What You’re Building: A program/protocol that sends SSH traffic via Nostr relays as ephemeral events. The primary use-case is to allow you to remotely manage your servers without:
  • knowing their public IPs
  • changing firewall rules
  • setuping up a bounce server or paying for an RMM service
What Kind of Help Are You Looking For? A software engineer, ideally with a deep understanding of SSH. But a good understanding of cryptography and network protocols are probably enough. If you don't know anything about Nostr, you'll obviously have to learn that.
Purpose: This program/protocol would solve a non-trivial problem for me at my job, but I don't have time to work on it. But I know that many other people have to deal with the same problem. I have no idea how to make a business out of this program/protocol, especially since it would be open-source.
How to Contact You: For the time being, @ me on SN and we can SYN-ACK some private comms.
Nostr would add unnecessary overhead to that, but you could wrap in the elements you want (like the identity keys) with regular sockets. Tunneling SSH over WS is not entirely uncommon.
RMM
lol, you must be in the MSP world... my condolences
reply
Nostr relays are like rendesvous points on the internet, which could replace the need for many centralized services, such as RMM services.
Nostr would add unnecessary overhead to that
Tor imposes a lot of overhead, but SSH works well when connecting to a server via a Hidden Service, which involves multiple TLS tunnels.
This program/protocol would only use ephmeral events, so only a few NIPs need to be implemented.
lol, you must be in the MSP world... my condolences
Thankfully I am not really in the MSP world. But, a part of my job requires that I use an RMM service, which is very rough around the edges, due to some serious design flaws.
I have looked at all other options, and making a new protocol is the only option -- well, other than quitting this job.
reply
Nostr relays are like rendesvous points on the internet
I was just comparing to a more general socat-like server, but yea compared to Tor it'd still be a massive improvement and might have added benefits by being able to hide amongst other traffic on public infrastructure.
ephmeral events, so only a few NIPs need to be implemented
You probably wouldn't need any in that case as the relay will just dump them, your protocol would just be an application that uses nip01 as it defines kind ranges for ephemeral events... then yours is just an application with a specific kind, there's a few like that including ours: https://github.com/nostr-protocol/nips#event-kinds
reply