In this workshop we will present the binary analysis toolkit angr.We will show it's capabilities for reverse engineering and how to use them to improve your reverse engineering workflow. This includes both using angr as a standalone tool, how it's features can be integrated into modern tools like IDA, Binary Ninja or radare2 and how to built your own custom tooling on top of it.
We will introduce core concepts like:
Intermediate Representations
- architecture independent analysis
- formalizing instruction behavior
Symbolic Execution
- Reasoning about how some code would behave depending on all its inputs and not just with one specific input
SMT Solving
- Formalizing your problem
- Use ~~decades of research in automated theorem proving and constraint solving~~ some Python to solve it
And present problems that can be solved using those concepts in addition to introductory exercises:
- First Hands-On experience with angr
- Thinking about programs symbolically instead of concretely
- automatically finding passwords/keys/backdoors
- Breaking anti-reverse-engineering measures and tricks (e.g. opaque predicates, dead code, obfuscation)
Participants should bring:
Some linux environment with angr installed and set up. This can be a linux VM, a docker container like https://github.com/angr/angr-dev or just a virtual environment on your host.
Participants must know or have:
- Basic understanding of concepts of reverse engineering
- Comfortable using the Python programming language