Update; [pp] now handles float arguments correctly, [pp f 45 f] will come out as [pack f 45 f] as it should instead of [pack f float f] as it was. Also added a note regarding $n arguments to the [pp] helpfile, all $s must be escaped or they will be expanded, no way around this one that I can see, must be lived with.All three now assume any incorrect type is an insert, [tt b Z f] will expand as [t b Z f] with what was connected to f staying connected to f and Z being an empty outlet. This might change to just ignoring all incorrect types so the previous example would expand to [t b f], nothing changed.

No more swapping wires or disconnecting and reconnecting things when you need to get another bang at the start of your trigger.

Some dynamic patching fun, just rename to the object as required, adjust your arguments to suit and it patches in the new trigger/pack/unpack automatically. This should save some time and headaches. Requires iemguts to use but since it replaces itself with a vanilla object your patches will still be vanilla if the target platform needs vanilla. You can confuse them and get some unexpected (actually logical) results, but easy enough to work around and certainly quicker and easier than the old way.
pp-tt-uu-0.3.zip

It is not letting me update the original post for some reason.

Updated version, no more save required and no more bugs (hopefully). Apparently I forgot about [iemguts/canvasconnections], so no longer need to save and parse the file to get the connections. Abstractions are much simpler, work better and appear to be bug free.

Edit; apparently it did let me edit, just did not update my browser. Oh well.

I think these are now about as good as they can get, I have been using them heavily for the past few days without issue. Only remaining bugs are when leaving off arguments to remove them when also inserting, [t b b b b] to [tt b B b] will connect the third outlet to the second as well as keeping the second connected, [tt b B b - -] solves this. Other bug is related, after the [tt b B b} if you undo and then redo it will end up like you had just done [t b b b] instead of [tt b B b], but this is unlikely to ever be an issue since it would be an unlikely move to do and and really just the undo/redo mechanism of pd doing what it should, it can not undo the self deleting abstraction, just replace it with the [trigger] the abstraction replaced itself with so connections remain as they were since that is all it knows. The bit of logic required to fix this seems to elude me and every fix results in error elsewhere so for now it stays and the user is advised that if you are both inserting arguments and removing trailing arguments remove them with - instead of leaving them off. If you are not inserting than things are fine, [t b b b b b] to [tt b - b] works as expected. Perhaps someone will chime in and point out my error here.

@oid thank you for this!

I was about to study how you made those abstractions but can not open uu.pd tt.pd pp.pd
Opening them with file>open - nothing happens, nothing in the console.

EDIT: Oh, now I see, you can get around this by

starting PD with the -noloadbang flag.

(And there is a name clash with iem_lib's prepend [pp])

uu and tt are working well here in your helpfiles.