Example reports

Six real reports, each from a verification run on synthetic data where the truth is known. Open one to see what a verdict looks like: the one-sentence reason, the category cards, the charts, the evidence and the checks. Every report is one self-contained HTML file: no scripts, no network, and it prints to PDF from the browser.

Report	What it shows	Verdict
A leak	A strategy that reads the next close (shift(-1)). The report names the line and the bars where the signal changes when the future is removed	REJECT
No edge	A moving-average crossover on a random walk: it loses money after costs and does not survive the number of variants tried	REJECT
An honest strategy	Momentum on data with a planted edge: beats its 200 time-shifted copies, survives costs, and shows that one bar of execution delay removes the profit	PASS WITH WARNINGS
A parameter search	A grid over lookback with walk-forward validation and a parameter heat map	PASS WITH WARNINGS
A universe	Cross-sectional momentum on several symbols, with the survivorship check	REJECT
Bad ticks	Honest code whose profit is one-bar price spikes: the data, not the strategy, makes the money	REJECT

How to read a report

• The line under the title says why the verdict is what it is.
• The four cards count the checks per family: integrity, look-ahead, economics, statistics.
• Equity is on a log scale next to buy and hold, with the drawdown underneath.
• The timing test shows the net return of 200 time-shifted copies of the same positions. A real signal beats most of them; a strategy that only holds a rising market does not.
• Sensitivity to costs shows how fast the profit disappears as trading costs grow, and where the modeled cost sits.
• Evidence lists the bars where a signal changed when the future was removed, and the source lines the linter flagged.

The reports are regenerated with uv run python scripts/make_report_gallery.py. Make your own:

monte-neo verify --ohlcv prices.csv --strategy my_strategy.py --html report.html