The REA Reverse Engineer Anything GitHub repo is github.com/morluto/rea, and it's free, open source and released under the MIT licence.
You install it with one command, which is npx rea-agents setup.
Before you run anything, check that the repo owner in the address bar is morluto, and never download a zip installer from a lookalike repo, because the real project installs through npm.
REA stands for Reverse Engineer Anything, and it's a single MCP server that lets your AI agent inspect an app, a website or a binary and work out how a feature actually works.
I'm Julian Goldie, and I tested it by pointing Claude at my own booking page, which I'll show you further down.
This page is the operator's walkthrough, so you'll get the repo facts, the setup steps, the first prompt, the full list of things REA can look at, my real test and the honest limits.
REA Reverse Engineer Anything GitHub: The Official Repo At A Glance
These are the facts I checked on the repo, the README and the npm registry on 13 October 2026.
| Item | Detail |
|---|---|
| Official repo | The official repo is github.com/morluto/rea. |
| Full name | The project calls itself REA: Reverse Engineer Anything. |
| Tagline | The README describes it as one MCP for reverse engineering across binaries, applications and runtime behaviour. |
| Licence | The code is MIT licensed, so it's free and open source. |
| Language | The project is written in TypeScript. |
| Created | The repo was created on 14 April 2026. |
| Stars | It had over 190,000 stars at the time of writing, and that number moves every day. |
| Forks | It had more than 46,000 forks when I checked. |
| npm package | The package is called rea-agents, and version 6.4.0 was the latest on 13 October 2026. |
| Website | The project site is rea.tools, with worked guides. |
My video showed 181,000 stars when I recorded it, so the repo is still growing quickly.
What REA Is And The Problem It Solves
REA is a toolbox your AI agent can call when you want to understand how a piece of software works.
You see a feature in an app or on a website, you'd love something like it in your own project, and you have no source code to learn from.
Before REA, your agent could only guess from screenshots and whatever you described.
With REA connected, the agent can inspect the target itself, trace the relevant code and come back with findings and the evidence behind them.
The README puts the idea in one line, which is that you see a feature you like and then understand how it works.
It also gives an example prompt, which is "Understand how search works in the Notes app, show me the evidence, and build a similar feature for my project."
How REA Works With Your Agent
Your agent calls REA through MCP to inspect the target and trace the relevant code.
REA sends back findings with their evidence.
Your agent then uses those findings to ask follow-up questions, explain the behaviour, or write and test an implementation.
The README says results include the evidence and the limitations behind each conclusion, which I like, because it stops the agent bluffing.
On privacy, the README says REA analyses targets locally, your agent receives the tool results, and your model provider has its own data policy.
What You Need Before You Install REA
The base requirements are short.
- You need Node.js 22.x at 22.19 or newer, 24.x at 24.11 or newer, or version 26 and above.
- You need npm, which normally comes with Node.js.
- You need an agent that supports local MCP servers, such as Claude Code, Codex, Cursor, Gemini CLI or Grok Build.
- You need a Chrome-family browser if you want REA to look at websites.
- You need Hopper, Ghidra or IDA only if you want deep native binary analysis.
You can check your Node version by typing node --version in your terminal.
How To Set Up REA From GitHub, Step By Step
Step 1: Open the official repo and check the owner
Go to github.com/morluto/rea and confirm the owner is morluto.
Step 2: Run the setup command
Open your terminal and run this command.
npx rea-agents setup
Step 3: Choose your agents
Setup asks which agents you want REA registered with.
It supports Claude Code, Codex, Cursor, Gemini CLI, Grok Build and other agents.
Step 4: Review the proposed changes
Setup shows you the changes it wants to make before it makes them.
Step 5: Approve the changes
Once you approve, setup adds REA's MCP server and the matching workflow instructions, which work like a skill for your agent.
It also backs up your existing configuration, so you can roll back if you don't like the result.
Step 6: Restart your agent
The new MCP server only shows up after a restart.
If your agent isn't on the supported list, the README says any agent that supports local MCP servers can still use REA through manual MCP registration.
🔥 Want my agent setups and the video notes for this build? Inside the AI Profit Boardroom, I share the notes and links from my videos, plus my Agent OS training and four coaching calls a week. You'll be building alongside 3,400+ members who are setting up agents of their own. → Get access here
How To Check You Have The Real REA GitHub Repo
When you search for REA GitHub, some results are copycat repos stuffed with keywords and offering an "official free download" as a Windows zip file.
The real project is not a zip download.
It installs through npm with npx rea-agents setup, and it lives at github.com/morluto/rea.
Here's the quick check I'd run every time.
- The owner in the address bar should be
morluto. - The README should show the
npx rea-agents setupcommand near the top. - The page should link to rea.tools as its website.
- You should close the tab if the page pushes you towards a zip or an installer file.
Your First REA Prompt
A good REA prompt has three parts.
It names the target, it names the feature you want to understand, and it says what you want back.
You should also ask it to show the evidence, the way the README's example prompt does.
In my own test I kept it even simpler than that.
I gave Claude the REA GitHub link and a link to my booking page, and I asked whether it could create the system I had right there.
What REA Can Analyse, And What Each Target Needs
The README has a long table of targets, and the requirements matter, so here are the main ones.
| Target | What REA returns | What you need |
|---|---|---|
| Websites | It returns page structure, scripts, network observations and requested screenshots. | You need a Chrome-family browser. |
| JavaScript and Electron apps | It returns modules, imports, source maps, routes and IPC relationships. | You only need Node.js and npm. |
| Saved network captures | It returns requests, responses and exposed payloads. | You need a HAR file or a Fiddler SAZ file. |
| .NET assemblies | It returns metadata, instructions and declared native dependencies. | You need nothing extra, because it's static only. |
| Native binaries | It returns pseudocode, assembly, strings, symbols, calls and references. | You need Hopper, Ghidra or IDA, which you supply yourself. |
| Android APKs | It returns manifest declarations, classes and decompiled methods. | You need headless JADX and a full JDK. |
| Android devices | It returns devices, packages, processes, logs and screen captures. | You need adb, with an emulator or a real device. |
| Flutter apps | It returns the Dart snapshot hash and build details. | You need nothing extra, because it parses the APK statically. |
| Firmware | It returns regions, extraction results and handoffs to native analysis. | You need Binwalk or Unblob on Linux. |
The README also lists WASM files, EVM bytecode, SQLite databases, Go binaries, packet captures and process behaviour.
Several of those targets are Linux-only or need tools you have to supply yourself, so read the requirements before you promise a client anything.
Setup can optionally install Hopper for you with your approval, while Ghidra and IDA use the copies you already have.
Using REA From The Terminal
The README shows this command for inspecting an extracted JavaScript or Electron app folder.
npx -y rea-agents@latest analyze-javascript-application /absolute/path/to/app --json
The result includes modules, imports, Electron boundaries and their evidence.
If you'll use it regularly, you can install the rea command globally.
npm install --global rea-agents
rea --help
🚀 Ready to put REA Reverse Engineer Anything GitHub to work in your business? Inside the AI Profit Boardroom you get step-by-step video trainings, my installable Agent OS and four coaching calls a week with 3,400+ members building real automations. → Join the AI Profit Boardroom
What Happened When I Pointed Claude At My Own Booking Page
This is the real test from my video, and it's the only test I'm going to claim.
I used Claude inside my Agent OS.
I gave it the REA GitHub link plus a link to one of my own funnels, which is the booking page for my free SEO strategy session.
I gave it no back-end access at all, so it worked from the front end only.
From the front end, Claude found the scheduling rules, including appointments per slot, slot duration, slot interval and calendar type.
It found who on my sales team is on the calendar.
It also worked out the calendar API.
Then it built my own custom scheduler app and plugged it into the workspace in my Agent OS.
Here's what that scheduler can do.
| Feature | What it does |
|---|---|
| Meeting length | You can adjust how long each meeting lasts. |
| Start grid | You can change the grid that slots start on. |
| Office hours | You can set the hours you're open for bookings. |
| Bookable window | You can set how far ahead people can book. |
| Earliest booking | You can set the earliest time someone can book from now. |
| Seats per slot | You can set how many people fit in one slot. |
| Reset and clear | You can reset the settings or clear all bookings. |
| Shareable link | You get a booking link you can send to people. |
| Booking flow | A visitor picks a slot, enters their details, confirms, and you can see who has the slot. |
| Verify versus live | It fetches the latest available appointments from my live calendar, which runs on GoHighLevel. |
Claude also built a page explaining how reverse engineering works, which helped me a lot.
It took a couple of prompts.
I'll keep the caveats in, because they're true.
I think we could improve the design, and I would change the UI.
I'm not a technical person, so if I can get this far, you can too.
If you want the longer write-up, it's in my Reverse Engineer Anything guide.
The Six Steps Of Reverse Engineering, As Claude Explained Them To Me
Claude gave me two analogies that made this click, and I'm crediting Claude for both.
The first is tasting a sauce at a restaurant and working out the recipe.
The second is watching someone's golf swing so you can learn how they do it.
Then it broke the process into six steps.
- You ask the question, which is not just how it works but why it works this way.
- You get the artefact, which is the thing you're going to study.
- You follow the strings, because programs are written by people who name things in English, such as a setting called "allow booking after".
- You decode what you found.
- You watch how it behaves as the settings change.
- You rebuild it and prove the rebuild works, as your own unique version.
Which Agents You Can Teach REA To
I only tested REA with Claude, so that's the only agent I'll vouch for from experience.
In the video I also pasted the GitHub link into Hermes, and I said you could teach the skill to agents like Antigravity and Muse.
The README backs up the general idea, because it says any agent that supports local MCP servers can use it.
If Hermes is your main agent, my guide on how to set up Hermes Agent in one click gets you running first.
The REA Showcases Worth Reading
The project site has three case studies on its showcase page.
They cover a DX-Ball sound-pan calculation, the Electron clipboard bridge in Notion, and a bullet-ring calculation from the PC-98 DOS game TH04.
The Notion case study traces one clipboard feature, and it isn't a clone of Notion.
Limits And Honest Caveats
I did not test native binary analysis, Hopper, Ghidra, IDA, Android or firmware in the video.
Those are documented features, and I'm reporting what the README says about them.
My test was a front-end analysis of my own booking page and nothing more.
The rebuilt scheduler worked, but the design needs work.
The README also says REA changes quickly and ships frequent bug fixes, so expect to update often.
Commercial tools like IDA and Hopper need their own legal licences, and REA doesn't give you one.
Use REA Responsibly
Use REA on things you own or are allowed to examine.
The README says it's for lawful research, education, interoperability, software preservation, and analysis of software you own or are authorised to examine.
The goal is to learn how a feature works and then build your own original version.
Don't copy someone else's code, brand or content.
If you'd like help planning your own SEO and AI systems, you can book a free SEO strategy session, which is the same booking page I ran this test on. If you're already serious about SEO and want to work at a higher level, have a look at the SEO Elite Circle.
How To Keep REA Updated
The README is clear that you should keep your install current.
If you installed the command globally, you run rea update.
If you use npx, you refresh your agent setup with npx rea-agents@latest setup.
If you hit a bug, update first, because a recent release may already have fixed it.
🔥 Want to build tools like this scheduler inside your own Agent OS? The AI Profit Boardroom has my Agent OS training, my agent setups and four coaching calls a week where you can ask questions about your own build. → Join the AI Profit Boardroom
Related Reading
- The Julian Goldie Agent OS GitHub guide explains the workspace I plugged the rebuilt scheduler into.
- How to set up Hermes Agent in one click is the fastest way to get a second agent running beside Claude.
- My Hermes MCP server setup for Codex is useful if MCP is new to you.
Also On Our Network
- 🌐 Seven ways a business can use REA without copying anyone
- 🌐 Every REA feature scored, with my honest verdict
- 🌐 The REA quick start for Agent OS users
- 🌐 Which AI model should drive REA, with real benchmark scores
FAQ: REA Reverse Engineer Anything GitHub
Where is the official REA Reverse Engineer Anything GitHub repo?
The official repo is github.com/morluto/rea.
It's free and MIT licensed, and it installs through npm with npx rea-agents setup.
Check that the owner is morluto, and never download a zip installer from a lookalike repo.
How do I install REA?
With Node.js and npm installed, you run npx rea-agents setup.
You choose your agents, review the proposed changes and approve them.
Setup adds REA's MCP server and matching workflow instructions, backs up your existing configuration, and then you restart your agent.
Which Node.js version does REA need?
The README says REA requires Node.js 22.x at 22.19 or newer, 24.x at 24.11 or newer, or version 26 and above, plus npm.
Which AI agents work with REA?
Setup supports Claude Code, Codex, Cursor, Gemini CLI, Grok Build and other agents.
The README says any agent that supports local MCP servers can use REA through manual MCP registration.
I only tested it with Claude.
Do I need Hopper, Ghidra or IDA to use REA?
You only need one of them for deep native binary analysis.
Static JavaScript and .NET inspection work without a native analysis engine, and website analysis needs a Chrome-family browser.
📺 Video notes + links to the tools 👉
🎥 Learn how I make these videos 👉
🆓 Get a FREE AI Course + Community + 1,000 AI Agents 👉
About Julian
I'm Julian Goldie, an SEO entrepreneur, author and founder of the AI Profit Boardroom, which has 3,400+ members.
I help business owners scale with AI agents, automation and SEO.
- I've built a 7-figure agency, Goldie Agency, with a team of around 50 people.
- I've grown a YouTube channel to 400,000+ subscribers.
- I wrote "Link Building Mastery".
- My Udemy courses have taught over 50,000 students.
→ Get my best AI training inside the AI Profit Boardroom
Check the owner is morluto, run the setup command, and point your agent at something you own, and you'll get real value from the REA Reverse Engineer Anything GitHub repo.
