An agent skill that turns a short JSON spec into animated, explorable diagrams exported as HTML, SVG, PNG, GIF or MP4.
morluto/rea
An MCP server that lets AI agents reverse engineer native binaries, JavaScript/Electron apps, .NET assemblies, websites and more, with evidence for each finding.
- Stars
- 305,411
- Forks
- 77,308
- Open issues
- 1,219
- Last commit
- 21 hours ago
What is it?
REA is a reverse engineering toolkit exposed as an MCP server, so agents such as Claude Code, Codex and Cursor can inspect a target and trace its behavior. The typical use is to examine a feature in an application without its source code, explain how it works, and build a comparable version in your own project. The same workflows are also available as CLI commands.
The supported targets range from native binaries (pseudocode, assembly, strings, symbols and references, via Hopper, Ghidra or IDA) to JavaScript and Electron apps, .NET assemblies, Android APKs and devices, WASM, EVM bytecode, SQLite databases, packet captures, websites and firmware. Static JavaScript and .NET inspection reads files without running them and needs no native analysis engine. Runtime capture runs the selected target under the user's own permissions.
Results are returned with the evidence and limitations behind each conclusion. Analysis runs locally, and setup registers the MCP server with supported agents and installs matching workflow instructions, with backups of existing configuration.
Key features
- 01Exposes reverse engineering tools to agents through MCP and the same workflows via a CLI.
- 02Analyzes native binaries using Hopper, Ghidra or IDA, returning pseudocode, assembly and references.
- 03Statically inspects JavaScript/Electron apps and .NET assemblies without running them.
- 04Covers many formats, including Android APKs, WASM, EVM bytecode, SQLite, packet captures and firmware.
- 05Returns each finding with its supporting evidence and known limitations.
Use cases
- Understanding how a feature in a closed-source desktop or Electron app works, then reimplementing it.
- Inspecting a native binary's logic with an agent to explain its behavior and verify a rewrite.
- Auditing an Electron app's IPC and module boundaries without its source code.
- Examining an Android APK's manifest and decompiled methods during security or interoperability research.
Topics
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Agent Skills that make coding agents design distinctive frontends with art direction, real photos and in-browser visual checks instead of generic AI landing pages.
An agent skill that turns ideas, plans, or codebases into interactive HTML diagrams, working with Claude Code, Codex, Cursor, and OpenCode.
A macOS menu bar app and local gateway that lets Claude Code, Codex, Gemini CLI and other agents use any model or subscription you have.
A Claude Code and Codex agent skill that builds cinematic product promo videos in Remotion from your own screenshots and copy.
A Claude Code skill that turns a website URL into a 60 fps product launch video, built from HTML animations rendered with Playwright and ffmpeg.