There is no way to do this today internally to hashcat itself.
Instead, external tools like combinator3 or combinatorX (available in hashcat-utils) can be used, piping the results directly into hashcat:
combinator3 list1.txt list2.txt list3.txt | hashcat -a 0 [...]
But this is much slower than -a 1
. One way to improve performance is to pipe in the first two lists combined, and then append the third list with rules that append the third word:
combinator list1.txt list2.txt | hashcat -a 0 -r list3-append.rule [...]
... but then the length of your third wordlist is constrained by the number of rules that your setup can load into memory (usually a few million).