What it does

REA (Reverse Engineer Anything) is an MIT-licensed toolkit that connects an AI agent to local analysis tools through the Model Context Protocol. The idea is to point an agent at a feature in an app, have it inspect the app without source code, explain how the feature works with evidence, and then write a version for your own project. Setup is one command, npx rea-agents setup, which registers REA with agents such as Claude Code, Codex, Cursor, Gemini CLI and Grok Build, with backups of existing configuration. The same tools are available from the terminal through the rea-agents npm package. It needs Node.js 22.19 or later, 24.11 or later, or 26 and up. Targets include native binaries, JavaScript and Electron apps, .NET assemblies, websites, Android APKs and devices, firmware, saved network captures and packet files. Native analysis uses an existing Hopper, Ghidra or IDA installation, while static JavaScript and .NET inspection need no extra engine. Analysis runs locally, and each result lists the evidence and limits behind it. The project says REA is meant for lawful research, interoperability, preservation and software you are authorized to examine.

Capabilities

MCP server that works with Claude Code, Codex, Cursor, Gemini CLI and Grok Build
Native binary analysis through Hopper, Ghidra or IDA
Inspection of JavaScript and Electron apps, .NET assemblies and websites
Android APK, firmware and network capture analysis
Findings returned with evidence and stated limitations
Same workflows available from the command line

When to use it

Understanding how a feature in a compiled or packaged app works without its source
Recovering behavior from an Electron app or .NET assembly for interoperability work
Giving a coding agent evidence-backed analysis of a binary or capture
Scripting repeatable analysis from the terminal