REA

REA is a reverse-engineering tool that connects an AI coding agent to tools for inspecting native binaries, JavaScript and Electron applications, .NET assemblies and websites. It is built for developers who want to understand how a feature in an existing app works, without its source code, and then build a similar version for their own project.

A developer runs a setup command that registers REA as an MCP server with an agent such as Claude Code, Codex, Cursor, Gemini CLI or Grok Build, or uses the same analysis tools directly from the terminal. The agent calls REA to inspect a target and trace relevant code, and REA returns findings together with the reasoning and limitations behind them.

REA can run static JavaScript and .NET inspection without executing an app, capture runtime behavior under the user's own permissions, and perform offline native-binary analysis using an existing Hopper, Ghidra, IDA, Kuna, objdump or Rizin installation. It can also turn a recorded web app session into a reinforcement-learning environment that an agent can practice tasks against.

REA requires Node.js 22.x, 24.x or 26.12.1 or later, plus npm, and is distributed through the official morluto/rea repository, rea.tools and the rea-agents npm package. It is open source under the MIT license. On Linux, several analysis commands require a procps-compatible ps available on REA's PATH.

Use cases

• Reconstructing a game's compiled logic from its binary instructions
• Tracing an Electron app's clipboard bridge through preload and IPC
• Building a reinforcement-learning environment from a recorded web app session