Input identity
Store a data checksum or version, exact date range, timezone, symbol, aggregation rule, and missing-period policy.
Public engine outputs make candidate strategies inspectable. Treat each result as a calculation under stated data and fill assumptions, not as trading advice or proof of a live edge.
Public output still needs an audit trail
CHECK 14An AI-generated strategy result is not independent evidence merely because it is public. The useful artifact is the full calculation contract: data version, instrument, timeframe, strategy code or template, parameters, costs, fill rules, sample dates, exclusions, and every trial that preceded the published run.
Before treating a result as a candidate, reproduce it from the same inputs, inspect trades around gaps and same-bar collisions, add spread and slippage stress, and reserve untouched dates. Public visibility helps review; it does not correct look-ahead, repeated parameter search, incomplete OHLC execution, or a missing cost model.
Store a data checksum or version, exact date range, timezone, symbol, aggregation rule, and missing-period policy.
Version the strategy template and execution code. A natural-language name such as breakout is not a complete specification.
Publish rejected trials or at least their count. Showing only the strongest run creates selection bias.
Transparency makes an AI backtest inspectable; reproducibility and holdout discipline make it informative.
Free. No auth required. Works with Claude, ChatGPT, or any HTTP client. Up to 30 backtests an hour from one address.
Beginner exploration
Open each answer for a plain-language way to read Inspect public AI backtest requests, test it carefully and decide what to explore next.
This page focuses on “Inspect public AI backtest requests”.Audit public AI backtests through data and engine versions, cost and fill assumptions, search history, reproducibility and untouched holdouts.For “Inspect public AI backtest requests”, a beginner should identify what the AI workflow guide measures, assumes or teaches before acting on its conclusion.Treat this page's account of “Inspect public AI backtest requests” as a learning reference rather than a prediction, signal or promise of future performance.
For “Inspect public AI backtest requests”, give the model explicit data, rules, units, dates and output requirements instead of asking for a vague trading opinion.While exploring “Inspect public AI backtest requests”, require calculations, assumptions and rejected cases to remain inspectable.Keep your “Inspect public AI backtest requests” record honest: verify arithmetic and execution logic outside the language model whenever the result affects a decision.Before leaving “Inspect public AI backtest requests”, keep an untouched holdout because repeated prompting can overfit just like repeated parameter searches.
Turn one idea from “Inspect public AI backtest requests” into a rule with explicit inputs, dates, costs and pass-or-fail conditions.Ask AI to expose missing assumptions in that “Inspect public AI backtest requests” test, not to guess the next market move.Use the FXAbsolute AI Backtesting Lab to inspect calculations connected to “Inspect public AI backtest requests” and the assumptions behind them.Reproduce any important “Inspect public AI backtest requests” result and reserve unseen data before deciding that an apparent pattern is useful.
Continue your exploration of Inspect public AI backtest requests with the beginner AI prompt guide, or inspect public calculations in the AI Backtesting Lab.