REA (Reverse Engineer Anything)

MIT-licensed MCP/CLI toolkit that lets coding agents reverse-engineer apps and native binaries locally, which hit #1 on GitHub Trending in October 2026.

REA · Open-source momentum unrelated to any model · open-source

Published in Feed

Overview

REA is an MIT-licensed TypeScript toolkit, published to npm as rea-agents, that connects coding agents to decompilation and static-analysis tools (Ghidra, IDA Pro, JADX, Binwalk, EVMole, pwntools, LLDB, and more) over MCP, letting an agent investigate app behavior down to native binaries entirely locally. First published under the REA name in July 2026, the project grew quietly until version 5.0.0 (October 7, 2026) triggered a breakout: stars grew from roughly 5K on October 5 to over 116K by October 12, 2026, briefly topping GitHub's global trending page.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Key features

  • 68 built-in MCP tools covering decompilation, static analysis, and binary/protocol tracing
  • Works with any MCP-compatible agent (Claude, Codex, and custom agents)
  • Analyzes native binaries, Electron apps, Android APKs (via JADX), firmware (Binwalk/Unblob), and EVM bytecode (EVMole)
  • Produces explanations with evidence, confidence scores, and recreation steps rather than raw decompiled output
  • Installs globally via npm (rea-agents) with a `rea setup` command

Use cases

  • Reverse-engineering competitor app features or undocumented protocols for legitimate interoperability research
  • Security research and malware/firmware analysis assisted by an AI agent
  • Recreating a feature's behavior in your own stack from binary evidence

Pricing

Free and open source under the MIT license.

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
12 Jul 2026
Record updated
13 Oct 2026
Record last checked
13 Oct 2026

Explore the Feed

Other entries filed under open-source momentum unrelated to any model; inclusion does not imply equivalent features.

Browse the Feed