New story - Test

US30 — yes, and it fits perfectly US30 is NOT for grids (too trendy), but ideal for Opening Range Breakout — which suits your "notify → approve → execute" idea because it happens at a fixed time daily (7:00 PM IST). The proven recipe: mark the 9:30–9:45 NY range, enter on a candle close beyond it, stop at the opposite side of the range, take profit at 1.5x the range, with filters: price above VWAP for longs, and skip the day if the opening range is unusually wide — that signals a choppy session. Realistic stats: 40–60% win rate, but with 2:1 and 3:1 reward-to-risk, even a low win rate is profitable over time. TradingView + 2 My proposed system for you (two engines) Engine A — "Harvester" (fully auto): ATR-spaced capped grid on AUDNZD + EURGBP, ADX regime filter, news filter, hard equity stop at 10%, tiny base lots, weekly profit withdrawal via n8n reminder. Engine B — "Hunter" (semi-auto, your approval idea): US30 + Gold NY-open ORB. n8n workflow watches the 9:30 range, sends you a WhatsApp via Evolution API with chart context, you reply "GO" → trade executes with anti-martingale pyramiding on winners. Tech note: the MetaTrader5 Python package is Windows-only, so your Linux VPS can't run it natively. Cleanest open-source path: write the EA in MQL5 with WebRequest() calling your n8n webhooks for the approval loop — but keep all safety logic (equity stop, max lots) inside the EA itself, never dependent on n8n, because network can fail mid-trade. Quick honesty note: I'm not a financial advisor, and even the "capped" versions can lose the full defined risk — this is research to build and test from, starting on demo. Want me to design Engine A's full logic (exact entry rules, lot table, filters) or Engine B's n8n + MQL5 architecture first?