Your backtest CSV can hide the exact bugs that make a strategy look profitable when it isnt — inflated win rates, uncontrolled re-entries, single-regime luck. This tool runs 8 automated checks on your trade log and flags the patterns that almost always mean a logic error, not real edge.

WHAT IT CHECKS

• Sample size (enough trades to trust the result?)
• Win rate plausibility (>75% is almost always a bug)
• Directional asymmetry (long/short results too different?)
• Trade frequency (uncontrolled re-entries per session)
• Time coverage (one market regime, or several?)
• Transaction cost realism (suspiciously clean P&L)
• Look-ahead bias heuristic (if your CSV has entry+high/low)
• Walk-forward vs in-sample (if your CSV marks the split)

Each check returns PASS / WARNING / FAIL / SKIPPED with a plain-English explanation — no quant background needed.

FORMAT: works with any CSV with a date + P&L column. Native TradingView List of Trades export is auto-detected and normalized.

HONEST LIMITS: MetaTrader/NinjaTrader exports may need manual column renaming. Checks that need extra columns you dont have will say so — never a fake result.

Runs 100% locally. Python script, no data leaves your machine. Not a substitute for a code audit — a fast sanity check before you ship or sell a strategy.