Test any AI trading prompt against 5 years of XAUUSD H4 data. Get Sharpe, drawdown, and equity curve in minutes — not weeks of demo trading.
Follow Telegram ChannelYou wrote a prompt. Is it good? You won't know until you've burned 2 weeks on demo.
Tweaking 5 prompts means 5 weeks of waiting — and most of them won't even be worth shipping.
There's no way to A/B test a trading prompt today. You're flying blind on the most important variable.
Three steps from a prompt file to a production-ready edge — backed by the same agent pipeline we ship in the EA.
[PRESET]
name: "my-edge"
entry: rsi < 30
[/PRESET]
Replaying 10,000 bars…
Export .txt → drop into MT5
Prompt Lab runs the exact same Bull / Bear / Judge / Decision pipeline that ships inside our Expert Advisor.
The numbers you see in the lab are the numbers you'll get in production. No simulator gap. No "well, the live LLM behaves differently." Same model calls. Same arbitration. Same execution rules.
Anonymous quotes from closed-beta testers running the same pipeline live.
Saved me 6 weeks of demo testing. I shipped two prompts in a weekend that I'd have killed on a live account otherwise.
The A/B compare is the killer feature. I finally have a number telling me which prompt is actually better — not a vibe.
Same Sharpe in the lab as on my forward-test demo. That's what convinced me — no simulator gap.
Be first in when the lab opens. First 100 followers get a gift.
Follow Telegram Channel