script to generate hexadecimal "words". warning: this makes your identifiers more vulnerable to discover.

hexawords.sh

#!/bin/bash -e
sources='/usr/share/dict/american-english /usr/share/dict/brazilian'

function usage()
{
  echo "Script to find words that can be spelled in hexa.
Notice that this can make identifiers more vulnerable to discovery.

By default, this uses English and Brazilian wordlists; edit the script
header if this is undesirable.  Works better if uni2ascii is installed.
tl;dr: apt-get install wamerican wbrazilian uni2ascii

Usage: $0 <OPTIONS>
Options:
  -l: enable l33t
  -s: match size in regex format (default: 4)
  -u: include words in the wordlist that use uppercase (often proper names)

Examples:
  $0 -s 4,8 -> find hexawords between 4 and 8 digits
  $0 -l -s 8, -> find hexawords 8 digits or longer, with l33t
"
  exit
}

temp=$(tempfile)
function cleanup()
{
  rm -f $temp 
}
trap cleanup EXIT

unset leet
unset uppercase
size=4
while getopts "ls:uh" option; do
  case $option in
    l) leet=y;;
    s) size=$OPTARG;;
    u) uppercase=y;;
    h) usage;;
    *) usage;;
  esac
done

for f in $sources; do
  if ! [ -f "$f" ]; then
    echo "Can't find file "$f"." >&2
    echo "Do you need to install a wordlist package (wamerican, wbrazilian...)?" >&2
    exit 1
  elif file -i "$f" | grep -q -i iso-8859-1; then
    iconv -f iso-8859-1 -t utf-8 < "$f" >> $temp
  else
    cat "$f" >> $temp
  fi
done

chars="0-9a-f"
if [ $leet ]; then
  chars="${chars}oisl"
fi

if [ $uppercase ]; then
  grepopts='-i'
fi

if which uni2ascii >/dev/null; then
  filter='uni2ascii -B -q'
else
  echo "Warning: missing uni2ascii(1), will ignore non-ascii characters." >&2
  filter='cat'
fi

$filter < $temp | egrep $grepopts "^[$chars]{$size}$" | tr oisl 0157 | tr '[:upper:]' '[:lower:]' | sort -u