Auto Cryptanalysis

This project implements python module for automatic cryptanalysis of Substitution Permutation Network ciphers by performing extensive linear and differential characteristic search and finding keybits

Structure

The project is structured as follows:

• The cryptanalysis directory contains the main Python module for cryptanalysis.
• The tests directory contains unit tests for the module
• The examples directory contains examples for using the module
• The docs directory contains html documentation autogenerated from code doc-strings

Installation

Pip

The project can be installed directly from pip bash pip install cryptanalysis

Otherwise clone and install is also viable

bash git clone https://github.com/deut-erium/auto-cryptanalysis.git cd auto-cryptanalysis pip install .

Requirements

This project requires Python3.6+ and the following Python packages: - z3-solver - tqdm

Requirements are auto installed as a part of the installation process but

You can also install these packages using pip: bash pip install -r requirements.txt

Usage

```python import random import cryptanalysis

sboxsize = 6 # bits pboxsize = sboxsize * 16 # 16 sboxes numrounds = 4 sbox = list(range(2**sboxsize)) pbox = list(range(pboxsize))

random pbox and sbox

random.shuffle(sbox) random.shuffle(pbox)

randomkey = random.randint(0, (2**pboxsize) - 1)

random spn instance whose key is unknown to us

spn = cryptanalysis.SPN(sbox, pbox, randomkey, numrounds)

dc = cryptanalysis.differentialcryptanalysis.DifferentialCryptanalysis(sbox, pbox, num_rounds+1)

override batch_encrypt with the oracle

maxnumencryptions = 50000 def batch_encrypt(plaintexts): return [spn.encrypt(i) for i in plaintexts]

dc.batchencrypt = batchencrypt differentialcharacteristics = dc.characteristicsearcher.searchexclusivemasks() lastroundkeyblocks = dc.findlastroundkey(differentialcharacteristics, maxnum_encryptions//16)

print("recovered last round key:",lastroundkeyblocks) print("original last round key:",dc.inttolist(spn.round_keys[-1])) ```

Tests

You can run the tests using the following command: bash python -m unittest discover

Documentation

Read the documentation
Autogenerated documentation from code doc-strings can be found under docs

Contributing

Please feel free to submit pull requests or create issues if you find any bugs or have any suggestions for improvements.
List of ideas to implement/TODO is present in CONTRIBUTING.md

License

This project is licensed under the GPL License.