AI Agent
Common Mistakes
Avoid these common pitfalls when using the AI Agent.
Even experienced users make these mistakes when working with the AI Agent. Learning to avoid them will save you time and produce better strategies.
Top Mistakes
- Being too vague — "Make me money" gives the AI nothing to work with. Specify indicators, timeframes, and risk levels
- Over-complicating prompts — Adding 10 conditions to a single strategy usually leads to overfitting. Start simple
- Ignoring risk management — Always include stop loss and take profit in your prompt. The AI may set loose defaults otherwise
- Not specifying a timeframe — The same strategy can perform very differently on 1-minute vs daily charts
- Expecting perfection — The AI creates a good starting point, but you should always review and refine the output
- Skipping backtesting — Never deploy an AI-generated strategy without thorough backtesting first
How to Fix Them
- 1Start with a simple, clear prompt covering entry, exit, risk, and context
- 2Review the generated strategy before backtesting
- 3Run multiple backtests with different parameters
- 4Iterate — refine the prompt based on backtest results
- 5Only add complexity when simple versions prove profitable
Complexity is not the same as quality. The best strategies are often surprisingly simple.