LIVE DEMO RUNNING · SINCE NOV 2026
Catalyst trades real US data surprises. Breakout takes one clean breakout a day and stays out on the big news days. AI Observer scores a daily AI forecast in public. XAUUSD only, a stop on every trade, no grid, no martingale.
THE PRODUCTS
Trades gold 5 minutes after a US report that surprises economists, closes 60 minutes later.
One trade a day when gold breaks a level set from yesterday's range. Skips NFP, CPI and Fed days.
Asks an AI every morning if today will be a trend day, then scores it against what really happened.
CATALYST EA · NEWS TRADER
It compares the actual numbers of major US reports (jobs, inflation, retail sales, business surveys) with what economists expected, measured against each report's own history. When the surprise is big enough, it takes one rule-based gold trade.
Weeks without a trade are normal. The backtest made about 10% a year. Research from mid-2017 stayed profitable in every period, at double costs, and without its 5 best trades.
Monthly results in the research log. A public MQL5 signal follows after the demo.
Trade-by-trade research replica of the MT5 test (+$227 vs +$269 in MT5). Simulated, not a promise.
BREAKOUT EA · DAILY BREAKOUT
Gold often picks a direction early and keeps going. Breakout sets a buy level and a sell level from yesterday's range and takes the first one that breaks. Losers are cut at the day's open; winners run until the evening.
Losers are cut early at the open; winners run until the evening. On average a win is about 1.4× a loss, so 48% winners still adds up. Expect losing streaks: 12 in a row happened.
We ran the exact same rules on 15 other markets: forex pairs, US and European indices, big US stocks. None passed, so we don't sell it for them.
Monthly results in the research log. A public MQL5 signal follows after the demo.
Research replica of the MT5 test (+$1,119 vs +$1,024 in MT5). Note the flat and losing stretches: they are normal for this strategy.
AI OBSERVER · RESEARCH TOOL · BETA
Every morning, before the Breakout entry window opens, AI Observer shows an AI model the last 10 daily candles, today's levels and today's US news calendar. The AI answers TREND or QUIET, a direction and a confidence. At 21:00 the tool records what really happened and updates a public scorecard.
Running Nov 2026 – 31 Jan 2027 with rules fixed in advance. We'll publish the scorecard either way.
Panel illustration. Works with your own Anthropic API key.
PROOF, NOT PROMISES
Most bots are tuned until the backtest looks perfect. We did the opposite: rules were written down before each test, chosen on older data (2012–2021), checked once on newer data (2022–2026), with real spreads and again at double costs. Everything that failed was thrown away, even ideas that looked great for ten years.
Only gold stays above 1.15 in both periods. That's why GoldPulse is sold for XAUUSD only.
Updated: November 2026
LAUNCHING EARLY 2027
Catalyst trades the big US news releases. Breakout deliberately stays out on NFP, CPI and Fed days. In our tests their daily results moved almost independently (correlation 0.07). They will be offered separately and as the Gold Suite bundle, with AI Observer included.
We won't sell anything until the live demo matches the backtests. Leave your email and you'll get:
About one email a month. No spam, unsubscribe any time. See our privacy policy.
RESEARCH LOG
Every month we post what the bots did on the demo accounts and how it compares with the backtest, including the losing months.
Catalyst runs on a $1,000 demo account and Breakout on its own $3,000 demo account, both at 0.01 lot. AI Observer records a forecast every trading morning. The rules are frozen for three months: no tweaks, no manual trades.
What we expect: about 2–3 Catalyst trades and about 9 Breakout trades a month, with losing streaks along the way. First results at the start of December.
FAQ
Early 2027 on the MQL5 Market, once the 3-month live demo matches the backtests. If it doesn't, we'll say so here and keep working. Get launch updates to hear first.
No, on purpose. Catalyst and Breakout follow transparent rules. We tested AI as a trade filter and it didn't help, so it isn't in them. AI lives only in AI Observer, which never trades and is being tested in public.
Our tests used $1,000 per 0.01 lot for Catalyst and $3,000 per 0.01 lot for Breakout. Smaller balances mean more risk per trade. Each panel shows the risk at the stop before a trade. Final guidance comes after the live period.
Yes. They use different magic numbers and never touch each other's trades. Put each on its own XAUUSD chart and size for the combined risk.
No. We tested the same rules on 14 other markets and they didn't pass (see the chart in the proof section). GoldPulse is built for XAUUSD only.
We don't claim it. Catalyst trades news releases, which many prop firms restrict. Check your firm's rules first.
Recommended. If MetaTrader is closed at release time or at the breakout, the trade is missed.
Yes. Both bots had losing streaks and drawdowns in testing (up to about 8% and 10%), and live results can be worse. Only trade money you can afford to lose.
We publish the demo results every month, good or bad. Get them by email, plus the launch price first.