It has nothing to do with the protocol of Matrix.

* OP has an iphone and some kind of non-iphone messaging interface accessed through ssh which can somehow send messages to iphone users through iMessage

* OP prefers to use the non-iphone messaging interface to send messages to iMessage recipients (I'm assuming both OP and recipients are using a modern generation of the iphone)

I'm saying that-- as it stands today-- OP is accepting a decrease in security by sending to the iMessage recipients from the non-iphone setup instead of from OP's iphone.

Regardless of the security protocol of Matrix and regardless of which client OP is using, what I'm claiming will be true.

It's also trivial to refute-- if I'm wrong then the messages OP is sending from their custom box should show up with a blue background in iMessage on the recipient's iphone. Is that the case or not?

>It's also trivial to refute-- if I'm wrong then the messages OP is sending from their custom box should show up with a blue background in iMessage on the recipient's iphone. Is that the case or not?

The distinction that the iOS messaging app uses to set the color of the message bubbles is whether messages are sent over SMS/MMS or iMessage. Since they're sent over iMessage, the message bubbles will be blue.

But yes, adding Matrix to the mix does add more potential failure points, since there are now two messaging protocols involved instead of one. That's just the nature of bridging messaging protocols though.

iMessage + Matrix definitely has a larger attack surface than just iMessage