About

This website collects my attempts at reverse engineering binaries – mostly crackmes and keygens – and the notes I took in the process in an effort to learn more about a discipline I was interested in many years ago but gradually abandoned. Posts are listed in order of increasing complexity (kind of) as I find more difficult challenges to solve. To follow these posts with an RSS reader, use RSS feed icon this feed.

The tools and techniques presented here are for educational purpose only. No commercial software is reversed, attacked, or disassembled in any way.

Please exercise caution when running unknown binaries, especially if they were made by people who are interested in bypassing software protections.

⚠️ Important: challenge websites

Some of the binaries presented here come from challenge websites, which sometimes have leaderboards to track progress. Please do not cheat. People work hard to set up these challenge websites, often for free. Please follow the rules and use these resources responsibly for your own learning, not to undermine the spirit of these competitions. By cheating you won't get anywhere near the top of these leaderboards anyway, and you will also have entirely missed the point of these challenges, only reaching a mediocre score. The goal is to practice and learn, not to score meaningless points.

Docker image

I usually run unknown binaries in a Docker container with no network access, based on an image with a few different tools pre-installed.

The image is built from a short Dockerfile and includes the GDB dashboard tool, which is packaged as a standalone .gdbinit file and provides a great user interface for GDB. Download the GDB dashboard .gdbinit file from their GitHub project and the custom settings to save as gdbinit-re.local here.

FROM debian:11
RUN apt-get update
RUN apt-get install -y libc6-i386 lib32stdc++6 binutils less man manpages-dev strace gdb \
    gdbserver gcc ltrace curl wget unzip xxd python3 vim-nox procps python3-pip \
    silversearcher-ag upx-ucl

RUN pip3 install pygments

# Dashboard
COPY .gdbinit /root/.gdbinit
# Custom settings
RUN mkdir /root/.gdbinit.d
COPY gdbinit-re.local /root/.gdbinit.d/

# Add non-root user
RUN mkdir -p /home/re && useradd -N -d /home/re -u 1001 -s /bin/bash re && chown re /home/re

CMD ["/bin/bash"]

To build the image and tag it as re-tools:

$ docker build -t re-tools .

I run it with SYS_PTRACE enabled and networking disabled, and with a local directory mounted to /re:

$ docker run --rm --network none --cap-add=SYS_PTRACE --privileged \
  --security-opt seccomp:unconfined -ti -v ~/reverse-engineering:/re re-tools

In some cases I have also found it useful to disable address space layout randomization (ASLR) inside the container (see also this page explaining the feature):

# echo 0 > /proc/sys/kernel/randomize_va_space

Additional tools

The Dockerfile above is only a starting point, and I would recommend you make it your own and add the tools you need. Here are a few that I have found useful:

  • checksec to list the security features of a binary (e.g. ASLR, NX, PIE, RELRO, etc.)
  • GEF (GDB Enhanced Features), focused on exploitation and reverse engineering
  • pwntools, a CTF framework and exploit development library in Python

Reverse engineering software

Outside of the containerized environment, I use the following software for reverse engineering. Some need a license, but if you are looking for a free option I would highly recommend Ghidra.

  • Ghidra (Free software, multi-platform)
  • Hopper Disassembler (Commercial with free trial, macOS and Linux)
  • Binary Ninja (Commercial with free trial, multi-platform)
  • Radare2 (Free software, multi-platform)
  • Hex Fiend (Free software, macOS only)
  • angr (Free software, Python-based binary analysis platform)

Reference documents


This work is © Nicolas Favre-Felix, licensed under a Creative Commons icon Creative Commons Attribution-ShareAlike 4.0 International License.

You can reach me on Twitter, Mastodon, or by email at n.favrefelix@gmail.com.