Overview
Gold House MT5 by Chen Jia Qi is a breakout EA for MetaTrader 5 that trades only gold (XAUUSD). It costs $749 on the MQL5 Market, with a free demo and 10 activations. It was released on 12 February 2026, the current version is 3.0 of 16 June 2026 (read on 19 November 2026). It has 61 reviews and 446 comments (as of 19 Nov 2026). It is not available at AlgoTradingSpace.
Gold Snap comes from the same vendor, described as the sister product that takes profits faster.
What the Vendor Sells
Per the product page (read 19 Nov 2026):
- Strategy: five independent sub-strategies that trade breakouts of swing highs and lows, with a multi-layer direction filter. Every trade has a stop-loss, plus trailing stop and partial close. Stop, target and trailing scale with the gold price. “Mandatory stop loss on every trade — no grid, no martingale, clear risk boundaries.” What each of the five sub-strategies does, the page does not describe.
- Modes: “Profit Priority” and “BE Priority” (stop to break-even early), since version 2.0 “Conservative” and since version 3.0 “Adaptive”. Version 2.1 added risk presets L1 to L5 that size lots by equity.
- Account: ECN account with a gold spread up to 25 points, VPS recommended. Start from $100 with 0.01 lot, recommended from $1,000, “optimal” from $2,000. Recommended parameters and the manual come by private message after purchase.
- Origin: “This EA comes from our team's internal live trading account.” Those accounts are not public; only what is on the linked signals can be checked.
The product page warns itself: “Gold does not always experience intense volatility — consolidation periods of 1-3 months are normal.”
Marketing, quoted as read on 19 Nov 2026: “The price will increase by $50 after every 10 purchases. Final planned price: $1,999.” How many purchases there have been and when the price last rose, we cannot check. A price ladder should play no part in a buying decision.
Design and Risk
Breakouts with a fixed stop on every trade limit the single loss. Four points still drive the risk.
1. Five sub-strategies, one instrument. All trade gold. The developer writes himself: “Of course, one disadvantage is that several positions can sometimes lose at the same time.” (comment of 24 Jun 2026, read 19 Nov 2026). In “Adaptive” mode the sub-strategies A, B, C and E cannot be switched off one by one. How to cap total risk is covered in Lot size and risk per trade.
2. Per the developer, the default mode carries more risk. Since version 3.0, “Adaptive” is the default. Per the developer it uses a higher default lot size and is “a medium-to-high risk setting” (comment of 17 Jul 2026, read 19 Nov 2026). He advises against going below 0.02 lot and below $1,000 of capital. Anyone who starts the EA on defaults is therefore not running the variant shown on the two public signals. The product page names no recommended risk per trade in percent; the presets L1 to L5 set it.
3. False breakouts. Breakout systems lose several trades in a row in sideways markets. The comments from July 2026 describe exactly that: streaks of false breakouts, on single days several losing trades at once.
4. Leverage. The two signals run at leverage 1:100 and 1:500 with low margin usage. In the EU retail clients may trade gold at most at 1:20, major pairs at 1:30. With small lots that is enough; with “Adaptive” on a small account margin can get tight.
The average holding time on both signals is in the range of minutes, and the vendor requires a tight gold spread and ECN execution. Like Gold Snap, we therefore also list Gold House as a scalper.
The Vendor's Signals
The product page links two public MQL5 signals, both at the same broker, each with $2,000 starting capital and no deposits or withdrawals. Neither had subscribers (subscription $999 a month).
- “Gold House” (2359124, Profit Priority mode) has been running since 12 February 2026, the day of the market launch, at leverage 1:100.
- “Gold House BE priority mode” (2372604) has been listed as a signal since 7 May 2026; the account has traded since April. Leverage 1:500.
What we read on 19 November 2026, without figures, as in all our reviews:
- Both accounts are in profit. On the older signal a large part of the gain came in the first two months; since then up and down months alternate, and October 2026 was clearly negative. On the BE signal most of the gain came from May.
- The largest decline is moderate on both; the worst value is on the older signal by balance. By equity it was less, but that does not count as an advantage for us. Under our methodology the worst value falls in the “moderate” band of the table in the calculation.
- MQL5 has repeatedly set and removed the warning about too-frequent deals for copying on both signals. On 19 Nov 2026 it was active on the BE signal (last set on 12 Nov 2026) and had been removed on “Gold House” since 16 Nov 2026.
- On both signals MQL5 points out that most of the gain came on very few days.
The third signal. On 30 Jul 2026 the developer wrote that the reference signal for version 3.0 deliberately runs a higher lot size: “It should therefore be viewed as a high-risk example rather than as the expected performance under our recommended risk setting.” (comment of 30 Jul 2026, read 19 Nov 2026) Buyers reported that this signal fell sharply within a few weeks in July. It can no longer be found on the vendor's profile, and an announced new signal with the recommended risk is not linked on the product page (as of 19 Nov 2026). For the rating this means: the two visible signals do not show the mode that is set by default.
Real money or demo: Per MQL5 rules only real-money accounts are allowed for signals, MQL5 shows the “Real Account” icon on both, and the accounts run on the broker's live server. Which lot size, preset and version run there is open. We do not run an account with Gold House in our own EA park.
What Buyers Report
On the MQL5 Market Gold House stands at 4.39 out of 5 from 61 reviews, plus 446 comments (as of 19 Nov 2026). Buyer reviews do not count toward our grade.
The first months are dominated by requests for set files and praise for quick replies. From July 2026 the tone shifts: several buyers report losing streaks and having switched the EA off. Why backtests and live accounts often diverge for gold EAs is covered in Gold EAs: backtest vs live. One wrote: “The tough thing about this EA is that the backtest results are so inconsistent with the live results.” (comment of 21 Jul 2026, read 19 Nov 2026). Others reported at the same time that they were roughly flat, and a review from August 2026 reports a profit after several months despite a deep decline from the peak.
One buyer questioned the robustness after his own test with time-shifted price data. The developer replied publicly (1 Aug 2026): the method breaks time-based filters and is not a valid backtest; the “Conservative” mode with hardly any extra filters was kept on purpose as a baseline. We cannot verify either side. A comment of 29 Oct 2026 notes that there has been no update since version 3.0 in June.
How the Provisional Grade Is Derived
The grade follows the same thresholds as for every EA (calculation):
- Risk control: The worst decline on the two signals falls in the “moderate” band. We deduct because several sub-strategies can lose at once, the default mode uses a higher lot size, and its reference signal, per buyer reports, fell sharply and is no longer public.
- Verifiability: Two public accounts, the older one for more than half a year; vendor accounts are capped with us. We lower further: both at the same broker, one signal no longer public, and several buyers report different results live.
- Transparency: Public signals count. Deduction because settings and the manual only come by private message after purchase, the sub-strategies are not described and the reference signal for the default mode is no longer public.
- Value for money: $749 falls in the $500 to $1,000 band. If the price rises as announced, we will reassess.
- Usability: We have not tested the installation ourselves; the uniform value applies.
- Broker independence: evidenced at one broker, deduction for the scalper design with a spread limit.
As with every EA, an AI from a different model family also rates the EA blind; where it deviates strongly on a criterion, the mean applies there until a human has checked. For Gold House that is the case for transparency and broker independence.
Who It Is (Not) For
Suitable is the design for traders looking for a gold EA with a stop on every trade who are willing to pick the mode themselves. Anyone checking it should test the demo in all modes, compare “Adaptive” with “Conservative” and watch how many positions are open at once. The two public signals show “Profit Priority” and “BE Priority”, not the default.
Not suitable if another gold breakout EA is already running, such as Gold Snap from the same vendor or The Gold Reaper: all sit in the same instrument, which is the same bet several times (see EA portfolio management). Nor for accounts with a wide gold spread, or for anyone unwilling to sit through a sideways phase of several months.
Verdict
Gold House MT5 is, per the vendor, a gold breakout EA with five sub-strategies and a stop on every trade, without grid or martingale, for $749. In its favour are two public signals with a moderate decline, the older one since the market launch, though not in the default mode, and detailed replies from the developer in the comments, including to criticism. Against it stand a single broker, a default mode with higher risk whose reference signal is no longer public, a gain that came largely from the first two months, and reports from several buyers whose live results did not match the backtest. Use it only after your own test and only with money you can afford to lose. We earn nothing from the EA itself; we link the vendor and MQL5 only as sources, never as partner links. Links marked “Ad” on this page (VPS, ATS-listed alternatives) are partner links. Past results are no guarantee of future ones, and trading leveraged products can lead to the loss of the capital invested.
