Digital circuits and Python: low-level meets high-level in the solution to this oddball of a challenge.


We’re given some text being said by a “RoboDuck”:

Nak Nanak naknak Nananak naknaknak Nak? Nak Nanak nak. Nananak naknaknak Nak? Naknak naknak nak. Nanak naknak Naknaknaknak naknaknak Nak? Nak. Nanak Nak? Nak? naknaknak Nak? Nak Nak? Naknak Nak? nak. Nanak naknak Naknaknaknak naknak naknak naknaknak Nanak naknaknak Nanak naknaknak Nanak

We’re also given this malicious circuit that was injected into the RoboDuck’s voicebox: Malicious circuit

Coding up the circuit

First things first, let’s code up this circuit in the voicebox. We make a Python function that takes an 8-tuple of bools (A-H on the left of the circuit) and returns another 8-tuple of bools (A-H on the right of the circuit).

def voicebox(input):
  (a,b,c,d,e,f,g,h) = input
  output = (
    (not g) ^ h, # A
    f or g, # B
    f, # C
    d and (e if d else f), # D
    # d and (f if d else e), # D
    not d, # E
    not c, # F
    b and c, # G
    not (a or b) # H
  return output

Reversing the circuit

We’re given what the RoboDuck says out loud, but what we really want is what it’s trying to say (the flag). We need to reverse the voicebox circuit. We could try to reverse this circuit by hand, but there might be some outputs produced by multiple inputs, and other outputs that can’t be produced at all! On the other hand, there are only 256 possible inputs (2^8), so we can just bruteforce it instead, with this slightly hacky bit of Python code:

def voiceboxReverse(output):
  bits = [False,True]
  inputs = []
  for a in bits:
    for b in bits:
      for c in bits:
        for d in bits:
          for e in bits:
            for f in bits:
              for g in bits:
                for h in bits:
                  input = (a,b,c,d,e,f,g,h)
                  if voicebox(input) == output:
  return inputs

We’re done! (Nope)

So, we have the output from the voicebox, and we have a function to reverse the voicebox’s effect. All we need to do is look up each output character in an ASCII table, and we should be able to figure out what’s being inputted to the voicebox, right?

print(voiceboxReverse((False,True,False,False,True,True,True,False))) # 'N' = 4E in ASCII
# []


Google saves the day

What does the modern man do when confronted with a problem like this? Google it, of course! A quick search for “naknak duckspeak” brings up info about a duckspeak cipher. Each word in the duck’s speech corresponds to a single hexadecimal digit: “Nak” means 0, “Nanak” means 1, “naknak” means C, and so on for the rest of the words. We wrote up another quick Python file to decode duckspeak to bytes:

encoded = "Nak Nanak naknak [... and so on ...]"

mapping = {
    "Nak":  0,
    "Nanak":  1,
    "Nananak":  2,
    "Nanananak":  3,
    "Nak?":  4,
    "nak?":  5,
    "Naknak":  6,
    "Naknaknak":  7,
    "Nak.":  8,
    "Naknak.":  9,
    "Naknaknaknak": 10,
    "nanak": 11,
    "naknak": 12,
    "nak!": 13,
    "nak.": 14,
    "naknaknak": 15,

decoded = ""
for word in encoded.split(" "):
    decoded += str(hex(mapping[word]))[2:]
decoded = bytes.fromhex(decoded)


We get b'\x01\xc2\xf4\x01\xe2\xf4l\xe1\xca\xf4\x81D\xf4\x04d\xe1\xca\xcc\xf1\xf1\xf1' as our output, which we can now feed into voiceboxReverse (after some ASCII conversion) to get the possible inputs:

['s', 'ó']
['\\', '_', '\x9c', '\x9f', 'Ü', 'ß']
['t', 'w', 'ô', '÷']
['\\', '_', '\x9c', '\x9f', 'Ü', 'ß']
['E', 'F', 'M', 'N', '\x85', '\x86', '\x8d', '\x8e', 'Å', 'Æ', 'Í', 'Î']
['4', '7']
['c', 'k', 'ã', 'ë']
['\\', '_', '\x9c', '\x9f', 'Ü', 'ß']
['R', '\x92', 'Ò']
['\\', '_', '\x9c', '\x9f', 'Ü', 'ß']
['Q', '\x91', 'Ñ']
['U', 'V', '\x95', '\x96', 'Õ', 'Ö']
['4', '7']
['c', 'k', 'ã', 'ë']
['C', 'K', '\x83', '\x8b', 'Ã', 'Ë']
['<', '?']
['<', '?']
['<', '?']

Now all we need to do is pick one from each list to make a valid-looking flag. After staring for a bit, we find 1s_1t_N4k_0R_QU4cK???, which, when wrapped with UDCTF{}, gives us the flag!