Analyze anything.
Install nothing.

Reverso is a native-first reverse engineering toolkit. Triage unknown binaries, inventories of JavaScript bundles and offline website mirrors, from one CLI or one MCP server. Everything runs locally, with zero runtime dependencies and no Ghidra, Hopper, IDA or license server.

  • Zero runtime dependencies
  • Fully offline
  • MIT licensed
$ reverso analyze ./target.elf
{
  "file": "target.elf",
  "sizeBytes": 48104,
  "entropy": 6.913,
  "format": { "format": "elf64",
    "confidence": "high" },
  "hashes": { "sha256": "b4f9…c21a" },
  "hardening": [
    { "name": "PIE", "status": "pass" },
    { "name": "NX stack", "status": "pass" }
  ]
}

What it does

Install

From source, no global package needed beyond Node.js.

git clone https://github.com/nekonaru/reverso.git
cd reverso
npm ci
npm run build
npm link
reverso doctor

Requires Node.js ^22.23.4, ^24.22.1 or >=26.12.1. Analysis is static; nothing you analyze is executed. Use only on targets you are authorized to analyze.

Use it from your agent

reverso mcp speaks Model Context Protocol over stdio (protocol 2025-06-18). Add it to your client:

{
  "mcpServers": {
    "reverso": {
      "command": "reverso",
      "args": ["mcp"]
    }
  }
}

FAQ

Does Reverso execute the binary it analyzes?

No. All analysis is static. Nothing is loaded, run or emulated.

Does it need an internet connection?

No for analysis. Only the test suite binds to loopback during development, and reverso site fetches the site you explicitly ask it to mirror.

Can it decompile to pseudocode?

No. Reverso triages, inventories and scores. Feed the file to a dedicated decompiler for interactive work; use Reverso for triage, extraction and reporting.

Is the source code open?

Yes, under MIT. github.com/nekonaru/reverso