Plus your comment is just silly, whoever implemented this obviously knew about Unix shells... It takes knowledge of Unix shells to implement another Unix-y shell.
This isn't reinventing Unix, this is a (cut down part of) Unix. (Unix the family, not the original Bell Labs OS, of course).
And "non-portable" because it's only portable to Unix? That's... Approximately everything. Linux family and Darwin naively, NT via WSL/cygwin/etc., BSDs obviously, Android sorta-natively or via Termux, Chrome OS via Termux, crouton, or crostani(sp?). What is Unix not portable to?
I can't run C on a 8-bit calculator either. At least not using GCC or Clang or any other mainstream compiler. Sure, I could use some toy compiler for that, but then 99% of the libraries I'd like to use won't work there anyways, due to a variety of issues.
I was about to object, but actually it looks like you're right; tigcc appears to only target the TI-89 et al. which run on the 16-bit Motorola 68000 series. A quick search doesn't give me any C compiler for the TI-84 after all.
Your actual point, of course, is valid regardless:)
I absolutely abhor this quote, I think it's taken out of context.
First of all, it denotes hubris, which I hate.
Secondly, it's totally anti-scientific, it assumes that Unix is some paragon of OS design, the zenith of Computer Science.
I like Unix as much as the next IT guy, but let's cut it out with the cliché quotes. There's plenty of things Unix doesn't do that well (everything is a file, until it isn't; filenames are basically binary blobs, etc.: https://dwheeler.com/essays/fixing-unix-linux-filenames.html).
Plus your comment is just silly, whoever implemented this obviously knew about Unix shells... It takes knowledge of Unix shells to implement another Unix-y shell.