A completed checkout.
Is it the right order?
Simulate a shopping journey, introduce a fault, and follow the evidence from intent to checkout.
Not a real checkout. Runs in your browser.
“Buy one medium Sand item. Apply SAVE10 and use standard shipping.”
Everyday Carry · M / Sand
$80.00 · In stock · $5.00 standard shipping
SAVE10 takes 10% off the subtotal
Choose a clean run or introduce one checkout fault.
Jump to results ↓No account or personal details required
Signals from the sample
These figures come from a seeded dataset, not from the interactive runs above or any real merchant.
Loading the dataset…
Follow the data
A small, reproducible pipeline makes the dashboard and its limitations easier to inspect.
Generate
Seeded synthetic journeys produce versioned events and a fixed sample.
JSONLValidate
Typed records are deduplicated; malformed events are quarantined.
SQLiteAggregate
SQL builds scenario, funnel, quality, and daily summaries.
SQL → JSONFollow a run through the pipeline
Each event records its identity, stage, expected and observed state, and outcome. Amounts and latency provide context for the checks.
Download 40 sample eventsevent_id · run_id · scenario_id · schema_versionCONTEXTtimestamp_utc · merchant_id · product_id · variant_idBEHAVIORagent_action · stage · expected_state · observed_stateRESULTassertion_id · passed · category · currency · latency_msWhat this does not simulate
This is a rule-driven sandbox aligned to public commerce concepts. It is not a live Shopify store, a certified integration, or an LLM-powered agent. No customer data, checkout, or payment is involved.
About the numbers
The dashboard uses generated events. Interactive runs stay in this browser tab and do not change the dataset.
Run a second scenario to compare its outcome with the previous run.