About · ProEA Lab
A quant lab on the trader's side.
We build real trading tools — MT5 systems and TradingView engines — and ship them as source you own and extend with your own AI. Then we publish what we learn, including what lost.
ProEA Lab is a small team of systematic traders and builders. We got tired of the two things this industry runs on: black-box EAs you can't open, and indicators that flash a 90% win rate with zero proof. So we build the opposite — tools you can read line by line, score and audit yourself, and extend with your own AI — and we hold every one of them to a single rule: no claim we can't show you.
What we build
One lab, a growing line of tools — each shipped as full source with an AI-agent kit, so you don't just buy it, you own it and it grows with you.
MetaTrader 5 · Expert Advisor
MTR FullSystemEA
A complete MT5 trading system, shipped as full source — 16,923 lines across 21 files, a ~29-month modelled backtest with the drawdowns and flat months left in, run in a Rust tick simulator with explicit costs. One-time $49.
Inspect MTR →TradingView · Pine v6 · Built for AI
Advanced Smart-Money Engine
Scores every order block & FVG 0–100, re-weights by market regime, learns from how zones actually resolve, then shows its real Resolved % on your chart — a descriptive stat, never a fabricated win rate. Non-repainting, full source + an AI-agent kit so you own and extend it.
TradingView · Pine v6…plus a growing Pine v6 line — an all-in-one engine, a multi-timeframe confluence matrix, and a prop-firm challenge toolkit. Same rule across every one: full source, built for AI, and no number we can't prove.
We publish everything — wins and losses
Most shops show you one clean equity curve and quietly bury everything that didn't work. We do the opposite. We publish the development logs — the experiments that held up, the ones that lost money, the strategies we built and then killed, the dead ends that taught us something. MTR's ~29-month modelled backtest ships with its drawdowns and flat months intact. Our Smart-Money engine shows its real Resolved % — how often its zones actually held on your chart, a descriptive stat rather than a number we invented. If an idea lost, we'll tell you it lost. That's the only kind of track record worth trusting.
What we'll never do
- ✕Promise you a return, a win rate, or a monthly percentage.
- ✕Quote an accuracy number we can't prove on your own chart.
- ✕Call anything a "holy grail" or a "set-and-forget money machine."
- ✕Hide costs, drawdowns, or the experiments that lost.
- ✕Lock you into a black box you can't open, audit, or extend.
If our tools can't survive your suspicion, you shouldn't buy them — and we'd rather tell you that than sell you a screenshot.
What we believe
We're on your side of the screen.
We build for traders, not at them — the tools we ship are the ones we'd want to trade ourselves.
You own the engine; your AI extends it.
Every product is full source with an AI-agent kit. Buy once, then grow it forever with Claude, ChatGPT, or Gemini.
A backtest is a hypothesis, not a promise.
It can disqualify bad ideas. It cannot predict your future.
Adaptive scoring, not prediction.
Our tools rank, score, and learn — they never pretend to see the future.
Show the ugly months — and the losing experiments.
A track record with no scars is young, filtered, or lying by omission.
Honesty isn't the marketing — it's the moat.
It is also the only thing that survives a live account.
We do the research in public
This blog is where we write down what we learn building and testing real systems — free, and technical. The point is simple: make traders harder to fool, including by us. The more you understand backtests, execution costs, and real risk, the sharper every decision you make becomes, whatever you end up trading.
Verify us — don't trust us
Everything here has a public place you can check it. Start with the receipts, not our word: the source manifest, the ~29-month modelled backtest, the public changelog, and the product itself at proea.app. Our TradingView engines carry their own proof on the chart — the live Resolved %, which you can verify on Bar Replay before you trust a thing.
Trading carries risk, and most retail traders lose money. Our tools are educational decision-support, not financial advice or a signal service. Past performance is not future results; a backtest is broker-specific and the Resolved % is a descriptive statistic, not a guarantee. We say so every time numbers appear — here too.