Live Lab

Generate real synthetic fraud data — live.

This page runs the exact same generator that powers the CLI, compiled to run in your browser. Tune the controls, hit Generate, and you get genuine deterministic data you can download, test rules against, and explore. Nothing is sent to a server.

Generation controls

These mirror the CLI flags one-to-one.

Use case preset

Users {{ number(form.users) }}

Capped at 2,500 here so the browser stays snappy.

Fraud rate {{ Math.round(form.fraudRate * 100) }}%

Country Currency

Txns min Txns max

Seed

Same seed + options = byte-identical data, every time.

Fraud patterns

⚙️

Your dataset will appear here

Pick a use case or set the controls, then hit Generate dataset. Everything below — KPIs, charts, the rules tester, and the transaction explorer — fills in with live data.

Rules tester

Write a rule pack, score it instantly.

Edit the JSON, click Run, and see matched transactions plus precision / recall / F1 against the generated is_suspicious labels — the same engine as rules-test.

Available fields & operators

Operators: eq, ne, gt, gte, lt, lte, in, not_in, contains, exists. Useful fields include risk_score, amount, payment_rail, recommended_action, is_suspicious, country_mismatch, corridor, reason_code_count, user_kyc_status, user_account_age_days, user_device_count, user_failed_login_attempts_24h, user_beneficiary_count_24h, user_ip_country_mismatch.

Run the rule pack to see results.

Precision

Recall

F1 score

Matched {{ number(rulesResult.matched_transactions) }}

Actually suspicious {{ number(rulesResult.actual_suspicious) }}

True positives {{ number(rulesResult.true_positive) }}

False positives {{ number(rulesResult.false_positive) }}

False negatives {{ number(rulesResult.false_negative) }}

True negatives {{ number(rulesResult.true_negative) }}

Per-rule matches

{{ id }}: {{ number(n) }} No rules matched any transaction.

Transaction explorer

Search, filter, and inspect.

{{ number(filteredTx.length) }} of {{ number(dataset.transactions.length) }} transactions match

Transaction	Amount	Risk	Action	Pattern
{{ t.transaction_id }}	{{ t.currency }} {{ number(Math.round(t.amount)) }}	{{ t.risk_score }}	{{ t.recommended_action }}	{{ t.fraud_pattern }}
No transactions match these filters.

Showing {{ rangeText }} of {{ number(filteredTx.length) }}

Rows Page {{ page }} / {{ totalPages }}

Select a transaction to see why it was flagged.

User {{ selectedTx.user_id }} sent {{ selectedTx.currency }} {{ number(Math.round(selectedTx.amount)) }} via {{ selectedTx.payment_rail }} on corridor {{ selectedTx.ip_country }}-{{ selectedTx.beneficiary_country }}.

Risk score

{{ selectedTx.risk_score }}/100

Suspicious

Pattern

Status

Reason codes

{{ r }}
No reason codes — clean transaction.

Generate real synthetic fraud data — live.

Your dataset will appear here

Risk-score distribution

Fraud pattern mix

Write a rule pack, score it instantly.

Search, filter, and inspect.

{{ selectedTx.transaction_id }}