Setup a Trading Bot with Volume CVD Divergence & other confluences - Help welc

[ALCFX1]

Hello friends !!


I'm currently using AI to help me build the trading bot of my dreams, which tbh originates from a private community i'm in, they have a super good EA but i hate losing control and not having it for me.


Basically it's a trading bot created by some veterans of a big broker and it's invite only for big players in the brokers, which i'm not but my friend is so a few of us are getting access to it this way - BUT this is not the subject of this thread.


Is there any ways to reverse engineer the actual strategies and functions of the trading bot without having any files provided but solely by watching behavior ?


From what i know it's based on standard deviations, CVD divergences and TDI for entry refinement.For entries it's also basing itself on EMA's cross on the lower TF's but i have no details.


I know a few of you are smart, so could you tell me if it's mission impossible to reverse engineer it without the files or more precise informations ?


I hate losing control haha


anyways thanks for your time & help colleagues!!

 

[Enivid]

Possible, but very very difficult. Maybe an AI can help you with this.

 

[ALCFX1]

Possible, but very very difficult. Maybe an AI can help you with this.

I'm trying with AI to watch behavior but it's not smart enough for now..so difficult

 

[GemiMQLDev]

Hi ALCFX1,

Without the source files or exact rule documentation, you probably cannot clone the EA reliably just by watching it. You can sometimes infer the broad idea, but the hidden parts are usually the filters: session rules, spread/news filters, confirmation order, risk model, partial close logic, recovery rules, and when the bot decides not to trade. Those are hard to reconstruct from screenshots alone.

The safer route is to build an independent version from observed behavior:

1. collect trade history with entry time, symbol, direction, SL/TP, close reason, and screenshots at entry;
2. mark the visible signals at that moment: standard deviation zone, CVD divergence, TDI state, EMA cross/timeframe, session, spread;
3. turn that into a written rule spec;
4. build a small MT4/MT5 prototype that logs why it would or would not enter;
5. compare the prototype decisions against the observed bot behavior and adjust the rules.

That will not copy the private EA, but it can produce your own controlled EA based on the same style of trading logic. I would not start with full automation immediately. First build an alert/backtest prototype, then add order execution only after the entry and exit rules are stable.

If you can provide 20-50 example trades with screenshots and the exact platform you want, I can help scope a clean MT4/MT5 prototype. I do not decompile or bypass protected EAs, but I can build an independent version from clear rules and testable observations.