A guide to converting addresses between a program file, its virtual memory location, and its runtime location. It explains PIE and ASLR, which can move a program's load address, plus firmware and reverse-engineering tool offsets.
A security-assessment guide for testing AI models through their visible interfaces or available model files. It covers model copying, behavioral fingerprints, training-data membership checks, privacy leakage, and resistance to adversarial examples, which are inputs designed to cause incorrect model behavior.
A guide for examining machine-learning model files such as ONNX, PyTorch, Safetensors, and TFLite. It reconstructs network structure, extracts weights, and checks files for embedded markers or suspicious content.
An entry point for analyzing the security and structure of artificial-intelligence models. It decides whether the input is a model file, an API, both, or a potentially malicious package, then sends the work to the relevant analysis path.
A single entry point for reverse engineering: examining a file, program, firmware image, document, application package, or code to understand how it works. It first identifies the input and task, checks the environment when needed, and selects an analysis path.
An Android security-audit guide for examining how an app stores and uses encryption keys, including Android Keystore, the system for protected key storage. It also covers locating encryption calls and intercepting them with Frida, a runtime inspection tool, without recording key bytes.
A reverse-engineering workflow for Android native libraries: compiled .so files used by apps through JNI, the bridge between Java or Kotlin code and native code. It restores JNI registration and analyzes native behavior, including cases where the Java layer only calls native functions.
A guide to symbolic execution with angr, a tool that explores program paths using unknown input values and a constraint solver. It is aimed at finding inputs that reach or avoid chosen points in a program.
A workflow for examining software that is packed, protected, or deliberately made difficult to understand. It covers identifying the protection, unpacking simple or advanced cases, and undoing some forms of obfuscation.
An authorized research workflow for examining anti-cheat systems such as EAC, BattlEye, and Vanguard, including their drivers, services, and memory checks.
A guide to examining an Android application package (APK) without running it. It covers the app manifest, decompiled Java code, resources, permissions, components, and signs of packing or obfuscation.
A threat-attribution process for assessing whether security incidents, malware samples, infrastructure, or techniques may be linked to the same attacker or campaign. The conclusions are intelligence assessments, not proof suitable for court.
A guide for examining what a suspicious program does after it runs, including its processes, files, registry changes, network activity, persistence, and code injection. It maps observed actions to MITRE ATT&CK, a common catalogue of attacker techniques.
A general reverse-engineering workflow for studying compiled programs by checking their file format, imported libraries, decompiled code, runtime behavior, and memory.
A guide for examining EVM smart contracts when only their compiled bytecode is available. EVM is the software environment used by Ethereum-compatible blockchains, and bytecode is the machine code deployed to a blockchain.
A guide to analyzing software from game consoles and older gaming systems. It explains formats such as Switch program containers, PlayStation executables, Xbox files, and ROMs, which are files containing game code and data.
A tool for examining compiled modern C++ programs when the original source code is unavailable. It helps recover class information, virtual function tables, exception paths, and readable names from a binary file.
A process for investigating program crashes and deciding whether they reveal a security flaw. It covers reproducing the crash, reading memory-error reports, reducing the input, locating the cause, and making a small proof-of-concept (PoC).
A process for turning encrypted data, such as files, traffic, or configuration blobs, back into readable data when the algorithm and key are known. It can also reconstruct the program’s decryption logic as a separate script.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: