AI Native Landscape

rea

Tracked

An MCP server and agent skill for reverse engineering anything — native binaries, JavaScript and Electron apps, .NET assemblies, and websites — with analysis running locally and every conclusion backed by evidence.

AuthormorlutoOpen Sourced2026-04-14Last CommitUnknown

Overview

REA (Reverse Engineer Anything) connects an AI agent to tools for inspecting native binaries, JavaScript and Electron apps, .NET assemblies, and websites. Point it at a feature in an app you do not have source for, and the agent can explain how the feature works, show the evidence behind each conclusion, and build a similar version for your own project. The same tools are usable directly from the terminal.

Key Features

  • One MCP for reverse engineering across binaries, applications, and runtime behavior
  • Inspect native binaries through an existing Hopper, Ghidra, IDA, or Kuna installation
  • Static analysis of JavaScript and Electron apps, including ASAR packages, with no native analysis engine required
  • Every result includes the evidence and the limitations behind each conclusion
  • Setup registers both the MCP server and matching workflow instructions with the agent
  • Also usable as plain CLI commands from the terminal

Use Cases

  • Understand how a feature in a closed-source app works and rebuild it in your own project
  • Audit Electron or JavaScript applications without their source code
  • Investigate native binaries alongside disassembly workflow in Hopper, Ghidra, or IDA

Technical Details

  • Distributed as the rea-agents npm package; npx rea-agents setup registers REA's MCP server and workflow instructions with Claude Code, Codex, Cursor, Gemini CLI, Grok Build, and other agents, backing up existing configuration before changes
  • Native binary analysis bridges into an existing Hopper, Ghidra, IDA, or Kuna installation; setup can optionally install Hopper with approval, while static JavaScript/Electron analysis needs no native engine
  • MCP contracts advertise compact input schemas by default to reduce context overhead; REA_MCP_INPUT_SCHEMA_PROFILE and REA_MCP_OUTPUT_SCHEMA_PROFILE switch between compact and full schema profiles while keeping canonical server-side validation
  • Analysis runs entirely locally; on Linux, process-ownership checks for tool launches (Ghidra, JADX, firmware tools) require a procps-compatible ps on REA's PATH