REA, short for Reverse Engineer Anything, is an agent-oriented toolkit for investigating software without access to its original source code. It connects coding agents to reverse-engineering tools such as Hopper and Ghidra through CLI and MCP interfaces. The system can inspect native binaries, JavaScript and Electron apps, managed assemblies, packages, and selected web applications. Agents can explore functions, strings, symbols, references, call graphs, assembly, and decompiled pseudocode. Analysis is designed to run locally and retain evidence behind each conclusion. REA can then help an agent translate what it learned into a compatible implementation for another project. It emphasizes explainable investigation rather than claiming to recover original source code.
Features
- Agent-oriented reverse engineering
- Hopper and Ghidra integration
- CLI and MCP interfaces
- Native, JavaScript, Electron, and managed-code analysis
- Local evidence-backed investigations
- Feature understanding and compatible reimplementation workflows