README.md

Length Extension Attack on CBC-MAC

Prerequisites:

1. CBC-MAC

Using the attack described here, the attacker can generate a valid Authentication Tag of message M1 || M2 given MAC(M1), without having the knowledge of key k. Note that M2 can selected purely by choice of the attacker.

Case scenario:

• Code running on the server that allows the user to give a string as an input
• Checks if the string is equal to M1 || M2, if not then it
  • Calculates CBC-MAC of the input string
  • Returns the Authentication Tag of the message
    Otherwise,
  • Returns a null string

As an attacker, we want to get the authentication tag of M1 || M2, without sending M1 || M2 as the input. Let us see how we can do it.

The vulnerability and exploit

There are two cases that we have to take into consideration while implementing the exploit: case when the IV is a null string and the other when IV is not a null string.

• Case 1: When IV is a null string: We can get the authentication tag of M1 || M2 by the following steps:
  • Get the authentication tag of message M1 = MAC(M1)
  • XOR it with M2, and send the result as input to the code running on the server.
  • The output will be MAC(M1 || M2)
• Case 2: When IV is not null: It's value will most probably be given, we can get the authetication tag of M1 || M2 by the following steps:
  • Get the authentication tag of message M1 = MAC(M1)
  • XOR it with (M2 xor IV) and send the result as input to the code running on the server
  • The output will be MAC(M1 || M2)

Example

Check out the example script here.