Method · 3 builds · checker

The AI says it understands the app. Ask for proof.

Four rules, three ready-made prompts, and a checker that rejects the agent when it guesses.

Rebuild with evidence: a golden magnifying glass over documents stamped approved
In short

REA with proof it's an open, free toolkit for people who use an AI agent (Claude Code, Codex, or the dsh harness) to study an app and build their own version. REA stands for Reverse Engineer Anything, an open reverse engineering tool: understanding how a program was made by looking at what's already built. The problem: the agent says it understands how the app works and makes up things it hasn't seen. The toolkit includes 4 rules, 3 ready-to-use prompts (agent dashboard, lead reactor, and rebuilt website), and a Python checker that rejects the evidence file when proof is missing. To use it, you need Python 3 and an AI agent in the terminal.

Video

Watch: the method in under 3 minutes

The video walks through the kit's 4 rules and 3 builds (in Portuguese).

Watch on YouTube

⚠️ Safety and license — read first

  • Only software you own or open-source software whose license allows you to study it (MIT, Apache-2.0, BSD, GPL, MPL…). Don’t use this method on another company’s closed-source app.
  • Studying isn’t copying. Your version is written from scratch, with your name, text, and design. Don’t take files, icons, or branding.
  • GPL/AGPL license: if you reuse any code, follow its terms. If in doubt, take only the idea.
  • Run the agent in a separate folder, without passwords, keys, or customer data. The kit only reads and writes files in the work folder; nothing is sent, deleted, or published on its own.

The kit is MIT licensed. Full text of the rules: REGRAS.md (in Portuguese).

What it is

Asking for proof changes what the agent delivers

Without proof, the agent writes a convincing paragraph. With proof, each sentence points to a file and a line you can open.

The kit includes: 4 rules, provas.md, checker, agent dashboard, lead reactor, and rebuilt site
✗ No proof
C1. The app uses a modern architecture
    based on microservices.
    Confidence: high

Sounds good. You have no way to check. Sometimes it’s true; sometimes it’s a guess.

✓ With proof
C1. Each agent is a card in a grid
- Proof: `src/components/AgentGrid.tsx:12-48`
- Excerpt: `repeat(auto-fill, minmax(280px, 1fr))`

Open the file at line 12 and see for yourself. If it’s not there, the agent got it wrong.

The 4 rules

The whole method fits into four rules

Paste them along with any request like "study this app and make mine".

1

Ask for provas.md

Each conclusion with the file and excerpt that proves it, plus the section "O que NÃO deu para provar".

2

Focus on the app’s code

Tell it to ignore node_modules, vendor, and site-packages. Proving it with a library means explaining the library, not the app.

3

Only what’s yours or open source

Your own software or open source software with a license that allows you to study it. Then, YOUR version, with your name and look.

4

Describe the look

Colors with codes, fonts, motion, and density in a design brief. That’s what separates an ugly dashboard from a good screen.

How it works

Proof first, code second

Every request in the kit has two phases. The agent only builds after the checker approves the proof file.

You choose the app (yours or open source)→ Agent studies→ provas.md→ confere_provas.py→ APROVADO?→ Design brief→ Your version
3 ready-to-use builds

Three requests to paste into the agent

In Portuguese, with [brackets] where you swap in your own details. They work in Claude Code, Codex, or the dsh harness. The agents in the builds are simulated: nothing calls AI or an API, and no real messages are sent.

🧩 A · Agent dashboard

Studies how an open source app shows several agents side by side, then builds a dashboard with 6 live agents for a business.

  • Example: dental clinic — lead generation, inbox, SEO, scheduling, support, reports
  • Each card says what the agent does and its status (with a label, not just color)
  • Command box that sends the task to the right agent

prompts/a-painel-de-agentes.md (in Portuguese)

⚡ B · Lead reactor

Studies how an open source app splits a request into tasks for worker agents, then builds the lead workflow.

  • Lead comes in → qualify → reply → schedule → follow up (4 lanes)
  • A lead that doesn’t qualify exits with the reason written down
  • Celebrate on screen when the lead schedules

prompts/b-reator-de-leads.md (in Portuguese)

🌐 C · Rebuilt website

Maps YOUR site (sections, fonts, scripts, buttons) and rebuilds it as a premium version in your own words.

  • Way 1: save the page with curl, without installing anything
  • Way 2: REA looks at the page open in a local Chrome browser
  • Same buttons and destinations; no made-up prices or promises

prompts/c-site-refeito.md (in Portuguese)

The checker

confere_provas.py rejects guesses

A Python file that uses only the standard library and doesn’t need internet access. Reads provas.md and returns a score from 0 to 10, APROVADO or REPROVADO, and a list of what’s missing. Exits with code 0 (approved), 1 (rejected), or 2 (couldn’t read the file).

Rejects when

  • the section "O que NÃO deu para provar" is missing (or empty);
  • a conclusion doesn’t cite a file or excerpt (for example, src/arquivo.ts:40-62);
  • all references point to libraries (node_modules, vendor, site-packages, dist, build, .min.js files…).

Warns (without rejecting) when

  • the evidence doesn’t include the line number;
  • one conclusion cites only a library, but others cite the app;
  • the line "Licença:" is missing from the "Alvo" section;
  • the title still has [brackets] from the template.

The repository's bad example, checked (the checker responds in Portuguese):

python3 confere_provas.py exemplos/provas-ruim.md
NOTA 0/10 — REPROVADO
Conclusões: 4 · com prova no código do app: 0

O que falta (reprova):
  ✗ Falta a seção "## O que NÃO deu para provar". Peça ao agente: "liste o que você acha mas não achou no código".
  ✗ "C1. O app usa uma arquitetura moderna baseada em microsserviços": conclusão sem prova. Falta o arquivo/trecho (ex.: `src/arquivo.ts:40-62`).
  ✗ "C4. Dá para refazer tudo em uma tarde": conclusão sem prova. Falta o arquivo/trecho (ex.: `src/arquivo.ts:40-62`).
  ✗ Todas as referências apontam para bibliotecas (node_modules, vendor, site-packages...). Mande o agente ignorar essas pastas e provar no código do app.

Avisos (não reprova, mas melhore):
  ! "C2. A grade de agentes é feita com uma biblioteca de layout": só cita biblioteca embutida; mostre onde o PRÓPRIO app usa isso.
  ! "C3. As animações deixam a experiência fluida e profissional": só cita biblioteca embutida; mostre onde o PRÓPRIO app usa isso.
  ! Sem a linha "Licença:" em "## Alvo". Confira se você pode estudar esse app.

The good example (exemplos/provas-boa.md) is here NOTA 10/10 — APROVADO. Both use a fictional app to show the format.

Prerequisites

What you need

The checker runs on any computer with Python. REA is optional: it helps the agent see inside apps and websites.

Python 3.8 or newer

For the checker and tests. Check:

python3 --version

An AI agent in the terminal

Claude Code, Codex, or the dsh harness (a coding agent that runs on your computer). Any of them can read files and run commands in the folder.

REA (optional)

REA, the harness, and an open model are installed in another step-by-step guide:

REA without restrictions — installation guide →

Step-by-step guide

From scratch to your version

Commands for Linux, Mac, or WSL. On Windows without WSL, replace python3 with py.

1

Download the kit

Clone the repository into your home folder and check that the checker works. The following steps use this path.

cd ~   # the toolkit goes in your home folder: ~/rea-com-provas
git clone https://github.com/inematds/rea-com-provas.git
cd rea-com-provas
python3 -m unittest discover tests   # should end with OK
2

Choose the target and check the license

An open-source app (download its code) or your own website. Open the license file before doing anything.

git clone https://github.com/[dono]/[app-open-source].git ~/estudo/app
cat ~/estudo/app/LICENSE   # MIT, Apache-2.0, BSD, GPL…? Then you can study it
3

Fill out the design brief

Copy the template and replace the [brackets]: colors with codes, fonts, motion, and density. The template already includes a filled-in example in the INEMA style.

cp brief/MODELO-brief-design.md ~/estudo/brief.md
4

Paste one of the 3 build plans into the agent

Open the file at prompts/, replace the [brackets], and paste the text block into your agent, opened in the study folder. It only does phase 1 (study) and stops.

cd ~/estudo
claude   # or: codex — and paste the text from prompts/a-painel-de-agentes.md
5

Check provas.md

The request itself tells the agent to run the checker. Run it yourself too: this is your "ask for proof" step.

python3 ~/rea-com-provas/confere_provas.py provas.md
python3 ~/rea-com-provas/confere_provas.py provas.md --json   # same result in JSON

REPROVADO? Paste the "O que falta" list back into the agent and ask it to fix the issues. Open 2 or 3 proofs in the editor and check whether the text is really on that line: the checker checks the format; you check the content.

6

Allow the build

Only with APROVADO. Write this in the agent:

pode construir

At the end, it tells you how to open it, what was left out, and what needs manual testing.

Real test

The inema.club website rebuilt by build C

On 11/08/2026, we ran build C (method 1, without REA) in Claude Code, using the subscription, with inema.club as "your website." Both phases ran at once, without intervention. Numbers measured.

23

conclusions

Each one has a line from the saved HTML and a copied excerpt.

10

reviewer score

SCORE 10/10 — APROVADO, with all 23 conclusions proven in the page code.

4/4

proofs checked by hand

We opened 4 cited lines (72, 113, 144, 161): the excerpt was there.

7:52

minutes, both phases

472 s and 42 agent turns. It produced a 94 KB index.html, with no horizontal scrolling at 390 px or 1440 px.

The inema.club website as it looked on 11/08/2026
Before: the live website on 11/08/2026.
Premium version made by build C following the design brief
After: the phase 2 version, following the INEMA style brief (background #0c0c10, yellow #FACC15, Sora).

What the agent itself listed as a limitation: the original website's fonts and colors are in a CSS file that was not saved (only the HTML), and it couldn't open a browser to test the mobile layout (we checked afterward). Files from this run in the repository: exemplos/rodada-real (in Portuguese) (request used, provas.md, saved HTML, and new version).

Reviewer rejects the example without proof
exemplos/provas-ruim.md: NOTA 0/10 — REPROVADO.
Reviewer approves the example with proof
exemplos/provas-boa.md: NOTA 10/10 — APROVADO.
What’s in the repository

Kit files

FileWhat it’s for
REGRAS.mdThe 4 rules and the safety and licensing notice.
prompts/The 3 build plans: a-painel-de-agentes, b-reator-de-leads, c-site-refeito.
provas/MODELO-provas.mdThe format of the proof file the agent must provide.
confere_provas.pyThe checker: score, APROVADO/REPROVADO, and what’s missing.
brief/MODELO-brief-design.mdDesign brief template, with a filled-in example in the INEMA style.
exemplos/provas-boa.md (approved), provas-ruim.md (rejected), and the real round.
tests/Checker tests, no internet required.