autokaker

Automated vulnerability discovery and refactor (autopatcher)

  _____
 /     \  AI
|  o o  | ___
| \___/ |/   \
|__   __|     |
   | | | 01001|
 __|_|_|_____/
|           |
|  HACK.exe |
|___________|

Install

Execute:

pip install -r requirements.txt

API Requirements

Autokaker and Autopatcher can use two different LLM APIs:

1. Neuroengine.ai free API using LLama3 and other open models
2. OpenAI API

To use the OpenAI api, the api key must be in the file api-key.txt, and select the OpenAI model name from the combo box.

The Neuroengine API do not need any key.

Usage

autok.py [-h] [--patch] [--make MAKE] patha

AutoKaker v1.0: Simple LLM bughunter/autopatcher

positional arguments:
  path         The path to a file or directory

options:
  -h, --help   show this help message and exit
  --patch      Activate patch mode
  --make MAKE  Feedback command

Auto vuln-discovery example

Analize a single source file:

python autok.py source.c

Auto patching an existing project

Patching an example 'zlib' project at directory './zlib':

python autok.py --patch ./zlib

Patching an example 'zlib' project and testing the patch using 'make':

cd zlib
python ../autok.py --patch . --make "make"

Patching an example 'zlib' project, testing executing 'make' and compression/decompression test 'example64':

cd zlib
python ../autok.py --patch . --make "make&&./example64"

Current LLM bug-hunting leaderboard (From Crashbench V1)