I Built an AI Trading Bot for Gold Using ONLY Prompts — No Coding Required
I didn’t write a single line of code for this bot — I designed the strategy, then let GitHub Copilot build the entire thing from a detailed prompt blueprint.
This bot trades Gold (XAU/USD) using a weighted scoring system built from six classic indicators, then layers a machine learning model on top to predict short-term price direction. It connects live to MetaTrader 5 and includes a real-time Streamlit dashboard to track balance and model accuracy.
In this video, I walk through the entire build — the strategy logic, the exact prompts I used, the ML model, the broker connection, and the dashboard — so you can see that the real skill in AI-assisted coding isn’t syntax, it’s how clearly you can describe what you want.
⚠️ This is an educational project built and tested on a demo account. Nothing here is financial advice, and past performance doesn’t predict future results.

Python • GitHub Copilot • scikit-learn • MetaTrader 5 • Streamlit • VS Code
Indicators covered: EMA (7, 13), RSI (7), MACD (5, 13), ATR (14), Bollinger Bands (20), ADX (14)
If you want the full prompt blueprint used in this video, it’s linked below. And if you found this useful, subscribe — next up I’m rebuilding this same bot with Claude Code to compare how it handles the same blueprint.
#algotrading #githubcopilot #machinelearning #python #tradingbot #gold #xauusd #aicoding