uudecode in awk for use on Plan 9.  TL;DR:  It works, but is slow.

PoC||GTFO makes polyglot PDFs, all of which double as zip files.
Cool!  Phrack, though, is all text files, and embedded data is
uuencoded.  That's also cool, but it turns out Plan 9 has no uudecode.
It is probably trivial to port the C version, but it is easy to do in
awk and that is more fun anyway (despite, or because of, the lack of
binary operators).

...And it worked on the first try!

Unfortunately, the second try involved confirming my fears: `printf
"%s"` handles UTF-8 correclty, which makes it impossible to emit
binary data correctly, because awk is for strings of characters, not
arrays of bytes.  So we write out bytes as we get them.  For details,
see the giant comment block at uudecode:/^function.printdec .

Differences from the version shipped with sharutils:
· Slower!
· Portable:  tested working on Plan 9's awk (written by bwk himself),
  gawk, nawk, mawk, and recent versions of busybox's awk.  (busybox
  1.21 failed because it appears to elide null bytes, but busybox 1.38
  worked, so I didn't investigate further.)
· Significantly more verbose.  (I expected to have to spend more time
  debugging.)
· Does not mind more than one file embedded.  (The version bundled
  with sharutils stops after the first file.)
· Does not throw an error if zero files are embedded.
· Does not support base64, just old-style uuencoded.
· Somewhat more forgiving with bad data.  (Invalid characters decode
  to long strings of null bytes instead of throwing an error.)
· Security is "best effort" because I am the only user because I just
  wrote it today and I don't foresee this program getting too
  much use from people that are not me.

The old, standard-ish uudecode.c probably works with ape/pcc, and if
you want a "real" uudecode, that's probably what you should do.  This
version is just a stunt-hack, it is only in a git repo because I told
a couple of friends on fedi that I thought it could be done, so I'm
showing them.  The repo is only named "uu" because maybe I port
uuencode to awk at some point.  As things stand, I just wanted to look
at a file included in an old issue of Phrack, noticed there was no
uudecode on Plan 9, and thought it looked easy to implement, it was,
so I was able to view the files.