How to check a gold EA before you buy: 10 checks

Before you buy a gold EA, check what you can verify yourself: a plain description of the method, labelled figures, a run on your own broker’s ticks, the licence terms and the seller’s written answers.

• A review is somebody’s opinion of software you cannot inspect. A checklist is a set of things you can verify before money moves.
• Nothing here predicts whether an Expert Advisor makes money — a check that claimed to would be the problem this page exists to catch.
• We answer all ten for our own Products at the end, including the three answers that go against us.

If you are working out how to check a gold EA before you buy, you know the shape of the problem: the claims are loud, the evidence is a screenshot, and the only way to find out is to pay. Here are ten things you can establish first. None of them labels anything a gold EA scam; together they tell you how much of what you have been told you can check.

Why a checklist, not a review

Asking “is this gold EA legit” usually returns someone’s verdict on a binary they never opened. A verdict cannot be re-run; a check can. Passing every item below tells you a seller is legible, not that the strategy will make money. No item predicts results, and a list that implied one did would be committing the fault it was written to spot. MQL5’s Market rules draw the same line from the other side: a Product there must not “guarantee, promise or hint at profits through the Product name, logo, screenshots or description”, must not “contain words and expressions in superlative degree relating to Product features or the underlying concept”, and must not “present trading strategy backtest results as real trading results, whether expressly or implicitly” (MQL5 Market rules). Those are forex EA red flags a marketplace already refuses to host. Off it, nobody enforces them for you.

Four forex EA red flags you can check before you pay anything

1. Is the strategy described in plain language, or only by outcome? A page that says what the software earns but never what it does is describing a result, not a method. You should be able to say in one sentence when it enters, how many positions it holds and what closes them.

2. Is every published figure labelled? A number means nothing without what produced it, over what window, at what deposit, on what data. It is the commonest item on any how to spot forex EA scams checklist, and the easiest to apply.

3. Can you run it yourself before paying, without handing over an email? A download you must trade an address for says the seller expects to sell by follow-up rather than by evidence.

4. Is the licence model stated? What you get, on how many accounts, what happens when your account changes, and what happens if the seller disappears — software that needs a licence server stops working when that server does. What an account-locked build is covers both shapes.

Three things to test yourself before you buy a gold EA

5. Run it on your own broker’s ticks, over a window the seller did not choose — theirs is the one that flatters the rules, and gold EA curve fitting explains why. How to backtest a gold EA in MT5 has the procedure, how to read a backtest report what the numbers mean.

6. Run it at the deposit you actually have, not theirs. Where size does not follow the balance, the deposit alone moves the drawdown percentage: our own Helix Trail runs, the same trades and the same dollar net in the Strategy Tester, read 19.26 % maximum equity drawdown on $100,000 and 53.78 % on $25,000.

7. Check what it needs permission to do. Under Tools → Options → Expert Advisors in MetaTrader 5, two settings matter: Allow DLL imports (potentially dangerous, enable only for trusted applications) and Allow WebRequest for listed URL (MetaTrader 5 Help). Both are off by default. If a seller asks you to switch either on, ask what the code does with it first.

Three things to ask the seller in writing

8. Is there a published track record, and where does it come from? A broker-sourced statement and a screenshot are not the same object. Verifying a gold EA track record sets out what a verified badge proves and what it leaves open.

9. How do you pay, and what is the refund position? The payment method decides whether a chargeback exists at all. Get the refund wording in writing before you send anything.

10. Who answers when it breaks? Make a regulator’s register part of the habit. The FCA publishes warnings in exactly these words — “This firm is not authorised by us and is targeting people in the UK” (FCA) — and the CFTC’s advisory on AI trading schemes adds the underlying point: no technology predicts market moves, and promises of high or certain returns are the classic sign of fraud (CFTC).

How XauStack answers its own ten

In our favour, and checkable: each Product’s page describes its method in words rather than numbers (1). Every figure carries the label Strategy Tester, the window 2026-01-01 to 2026-09-07, the deposit, the broker feed and the two departures — no commission charged, fills modelled, against live costs of roughly $7 per lot round-turn our net does not deduct (2). The free Backtest Edition downloads with no email and no form, and runs only inside the Strategy Tester (3). The licence is perpetual, covers up to 2 accounts, is bound at checkout and carries no key and no server call; moving to a different MT5 login is a new licence at the current price, not a free re-issue (4).

Now the three that go the other way.

On check 8, we publish no live track record at all, by decision. Our own FAQ says why: “a link like that looks like proof without being one… The only proof that transfers is the one you generate yourself.” Weight that against us; it is a real absence.

On check 9, payment is crypto only, and there are no refunds once a build has been delivered, because it is compiled for your own account numbers. The single exception is a payment that never produced a working build.

And on our own check 2: the published runs read History Quality 63 %, because IC Markets served real ticks only from 2026-04-21, so January to April were synthesised by the tester from M1 bars. Read those four months as an approximation.

Before you act on any of this

Every figure above comes from the MetaTrader 5 Strategy Tester, never from live trading, and no result is promised anywhere on this site. Trading gold on leverage carries a high risk of losing money rapidly: read the risk disclosure, then run check 5 on the strategies yourself.