Backtest crypto trading strategies
Receive detailed performance metrics including equity curves, Sharpe ratios, and individual trade logs for multiple trading ideas simulated against historical market data.
How it works
The agent takes a user's natural language description of a trading strategy (e.g., 'Buy dips in an uptrend with RSI 20-40') and submits it to the AI Backtester MCP. The server converts the prompt into executable backtests, running them against years of minute-to-daily history for specified assets like BTC, ETH, or SOL. The agent then parses the returned data, which includes risk numbers (Sharpe, Sortino, max drawdown), win rates, and equity curves, presenting a comparative report to the user. Before acting, the agent should ensure the user has provided specific assets and timeframes, and remind them that backtests are simulations, not financial advice.
Prompt
Paste this into an agent that has the servers below connected. Adjust the names and dates to your own.
Test the following strategy: [Insert strategy description here]. Run backtests for BTC and ETH on the 1h timeframe over the last year. Include parameter sweeps for RSI levels between 20 and 40. Return the equity curves, Sharpe ratios, and a list of individual trades for the top performing variations.MCP servers it needs
Source
The vendor page this playbook is drawn from, and the passage that supports it.
Describe your strategies in plain English. Your AI turns each one into a real backtest, runs them against years of market data, and hands back the numbers — dozens at a time, in seconds.AI Backtester — free AI backtesting for crypto trading strategies