Skip to main content
The AML Compliance Pack provides 11 production-ready rules based on FinCEN regulations, covering currency transaction reporting (CTR), structuring detection, suspicious activity reporting (SAR), and fraud monitoring. Each rule includes historical fines, breach examples, and regulatory references to provide context for risk assessment.

Rule Categories

Rules are organized into five categories:
Reporting requirements for large cash transactions.
  • CTR Threshold — Transactions exceeding $10,000
  • CTR Aggregation — Multiple transactions aggregating to $10,000+ in one day
Detection of patterns intended to evade reporting thresholds.
  • Structuring Pattern Detection — 3+ transactions between 8K8K-10K in 24 hours
  • Sub-Threshold Velocity — 5+ transactions between 8K8K-10K in 24 hours
  • SAR Velocity — Total transaction volume exceeding $25K in 24 hours
Detection and reporting of potentially illegal activity.
  • SAR Threshold — Transactions ≥ $5K with high-risk types
  • High Value Transfer — Wire/transfer transactions > $50,000
Monitoring for unusual account activity and dormant accounts.
  • Dormant Account Reactivation — High-value transactions from dormant accounts
  • Balance Mismatch — Transaction amounts not matching balance changes
  • Round Amount Pattern — 3+ round-dollar transactions in 30 days
Identifying potential fraudulent transactions and patterns.
  • Fraud Indicator — Cash-out/transfers to previously empty accounts

Detailed Rule Reference

Currency Transaction Reporting

Severity: CRITICAL
Threshold: $10,000
Type: Single Transaction
What it detects:Flags any single transaction via CASH_OUT, TRANSFER, or WIRE exceeding $10,000.Conditions:
{
  "AND": [
    { "field": "type", "operator": "IN", "value": ["CASH_OUT", "TRANSFER", "WIRE"] },
    { "field": "amount", "operator": ">=", "value": 10000 }
  ]
}
Policy Excerpt:
Transactions exceeding $10,000 via CASH_OUT, TRANSFER, or WIRE must be reported to FinCEN.
Regulatory Reference: 31 CFR 1010.311 (Section 1 - Currency Transaction Reporting)Historical Context:
  • Average Fine: $1.4B (HSBC 2012 fine - failure to monitor transactions)
  • Breach Example: HSBC allowed $881M in money laundering by drug cartels
Severity: CRITICAL
Threshold: $10,000
Type: Aggregation (24-hour window)
What it detects:Flags when multiple transactions by the same person aggregate to more than $10,000 in a single business day.Conditions:
{ "field": "amount", "operator": "aggregate_sum", "value": 10000 }
Policy Excerpt:
Multiple transactions by or on behalf of the same person during any one business day shall be aggregated for CTR purposes.
Regulatory Reference: 31 CFR 1010.313 (Section 1 - CTR Aggregation)Historical Context:
  • Average Fine: $700M (FinCEN fine - systemic aggregation failure)
  • Breach Example: Bank failed to aggregate multiple $9K cash deposits made same day

Structuring and Velocity

Severity: CRITICAL
Threshold: 3 transactions
Type: Structuring (24-hour window)
What it detects:Detects structuring/smurfing — 3 or more transactions between 8,000and8,000 and 10,000 within 24 hours.Conditions:
{ "field": "amount", "operator": "BETWEEN", "value": [8000, 10000] }
Policy Excerpt:
No person shall structure, or assist in structuring, any transaction for the purpose of evading the CTR reporting requirement.
Regulatory Reference: 31 USC 5324 (Section 2 - Structuring Detection)Historical Context:
  • Average Fine: $200k+ (Individual fines for structuring)
  • Breach Example: Customer made three $9,500 deposits over 3 days to avoid CTR
Severity: HIGH
Threshold: 5 transactions
Type: Velocity (24-hour window)
What it detects:Flags any customer account with 5 or more transactions in a rolling 24-hour period where individual amounts are between 8,000and8,000 and 10,000.Conditions:
{ "field": "amount", "operator": "BETWEEN", "value": [8000, 10000] }
Policy Excerpt:
Flag any customer account with 5 or more transactions in a rolling 24-hour period where individual transaction amounts are between 8,000and8,000 and 10,000.
Regulatory Reference: 31 USC 5318(g) (Section 2 - Sub-Threshold Velocity)Historical Context:
  • Average Fine: $2M (Institutional fine for pattern detection failure)
  • Breach Example: Gambling site ignored frequent $9K transfers between users
Severity: HIGH
Threshold: $25,000
Type: Aggregation (24-hour window)
What it detects:Flags any individual account with transaction volume exceeding $25,000 within any 24-hour period.Conditions:
{ "field": "amount", "operator": "aggregate_sum", "value": 25000 }
Policy Excerpt:
Flag any individual account with transaction volume exceeding $25,000 within any 24-hour period.
Regulatory Reference: 31 CFR 1020.320 (Section 3 - SAR Velocity)Historical Context:
  • Average Fine: $613M (U.S. Bancorp 2018 - capped transaction monitoring)
  • Breach Example: Bank set monitoring thresholds too high, missing high-velocity smurfing

Suspicious Activity Reporting

Severity: HIGH
Threshold: $5,000
Type: Single Transaction
What it detects:Flags transactions ≥ $5,000 with high-risk types (TRANSFER, WIRE, CASH_OUT) for SAR review.Conditions:
{
  "AND": [
    { "field": "type", "operator": "IN", "value": ["TRANSFER", "WIRE", "CASH_OUT"] },
    { "field": "amount", "operator": ">=", "value": 5000 }
  ]
}
Policy Excerpt:
The Institution shall file a SAR for any transaction totaling $5,000 or more where it suspects illegal activity.
Regulatory Reference: 31 CFR 1020.320 (Section 3 - Suspicious Activity Reporting)Historical Context:
  • Average Fine: $450M (Capital One 2021 fine - SAR program failures)
  • Breach Example: Bank failed to file SARs on millions of dollars in suspicious activity
Severity: HIGH
Threshold: $50,000
Type: Single Transaction
What it detects:Flags any WIRE or TRANSFER transaction exceeding $50,000.Conditions:
{
  "AND": [
    { "field": "type", "operator": "IN", "value": ["TRANSFER", "WIRE"] },
    { "field": "amount", "operator": ">", "value": 50000 }
  ]
}
Policy Excerpt:
Flag any WIRE or TRANSFER transaction exceeding $50,000.
Regulatory Reference: Travel Rule (Section 5 - High Value Transfer Monitoring)Historical Context:
  • Average Fine: $100M+ (Penalty for failing to verify origin of funds)
  • Breach Example: High-value wire transfers to shell companies in offshore jurisdictions

Account Behavior Monitoring

Severity: MEDIUM
Threshold: $5,000
Type: Dormant Reactivation (90-day window)
What it detects:Flags any account inactive for 90+ days when that account conducts a transaction exceeding $5,000 within the first 30 days of reactivation.Conditions:
{ "field": "amount", "operator": ">", "value": 5000 }
Policy Excerpt:
Flag any account inactive for 90+ days when that account conducts a transaction exceeding $5,000 within the first 30 days of reactivation.
Regulatory Reference: FFIEC Guidance (Section 4 - Account Behavior Monitoring)Historical Context:
  • Average Fine: $15M (Average fine for identity theft/money laundering via dormant accounts)
  • Breach Example: Identity thieves hijacked dormant accounts to funnel fraudulent funds
Severity: MEDIUM
Type: Single Transaction
What it detects:Flags any transaction where the balance change does not match the transaction amount.Conditions:
{ "field": "balance", "operator": "mismatch", "value": 0.01 }
Policy Excerpt:
Flag any transaction where the balance change does not match the transaction amount.
Regulatory Reference: GAAP / FinCEN Controls (Section 4 - Balance Verification)Historical Context:
  • Average Fine: $5M (Fine for poor internal controls)
  • Breach Example: Technical error allowed double-spending, used as exploit by fraudsters
Severity: MEDIUM
Threshold: 3 transactions
Type: Round Amount (720-hour / 30-day window)
What it detects:Flags any series of 3 or more round-dollar transactions (amounts divisible by $1,000) within 30 days.Conditions:
{ "field": "amount", "operator": "round_check", "value": 1000 }
Policy Excerpt:
Flag any series of 3 or more round-dollar transactions within 30 days.
Regulatory Reference: FATF Guidance (Section 4 - Transaction Pattern Monitoring)Historical Context:
  • Average Fine: $1M (Indicator of sophisticated money laundering)
  • Breach Example: Iterative 10,000,10,000, 20,000, and $5,000 transfers used to mask origin

Fraud Detection

Severity: HIGH
Type: Single Transaction
What it detects:Flags cash-out or transfer transactions to accounts with zero prior balance that result in a credit.Conditions:
{
  "AND": [
    { "field": "type", "operator": "IN", "value": ["CASH_OUT", "TRANSFER"] },
    { "field": "oldbalanceDest", "operator": "==", "value": 0 },
    { "field": "newbalanceDest", "operator": ">", "value": 0 }
  ]
}
Policy Excerpt:
Transactions to accounts with zero prior balance that result in a credit may indicate fraudulent activity.
Regulatory Reference: FFIEC IT Handbook (Section 5 - Fraud Detection)Historical Context:
  • Average Fine: $3.4B (Combined annual impact of transaction fraud)
  • Breach Example: Attacker emptied multiple newly-created accounts via rapid transfers

Rule Summary Table

Rule IDNameSeverityThresholdTime WindowCategory
CTR_THRESHOLDCurrency Transaction Report ThresholdCRITICAL$10,000CTR Reporting
CTR_AGGREGATIONCTR AggregationCRITICAL$10,00024 hoursCTR Reporting
STRUCTURING_PATTERNStructuring Pattern DetectionCRITICAL3 txns24 hoursStructuring
SUB_THRESHOLD_VELOCITYSub-Threshold VelocityHIGH5 txns24 hoursStructuring
SAR_THRESHOLDSuspicious Activity Report ThresholdHIGH$5,000SAR
SAR_VELOCITYSAR VelocityHIGH$25,00024 hoursStructuring
DORMANT_ACCOUNT_REACTIVATIONDormant Account ReactivationMEDIUM$5,00090 daysBehavioral
BALANCE_MISMATCHBalance MismatchMEDIUMBehavioral
ROUND_AMOUNT_PATTERNRound Amount PatternMEDIUM3 txns30 daysBehavioral
FRAUD_INDICATORFraud IndicatorHIGHFraud
HIGH_VALUE_TRANSFERHigh Value TransferHIGH$50,000SAR

Using the AML Pack

  1. Select AML Framework — Choose “AML / FinCEN” when creating a new audit
  2. Review Rules — All 11 rules are loaded and active by default
  3. Toggle Rules — Disable any rules not applicable to your use case
  4. Upload Data — Upload your transaction CSV (must include fields like amount, type, account)
  5. Confirm Mapping — Approve the AI-suggested column mappings
  6. Run Scan — Execute the compliance scan
  7. Review Violations — Violations include policy excerpts, historical fines, and evidence
All AML rules use deterministic logic — no AI models are used during enforcement. Confidence scores are based on rule quality, signal specificity, statistical anomalies, and Bayesian precision from user feedback.

Next Steps

GDPR Compliance Pack

Explore the 15 GDPR rules for data protection compliance

SOC2 Compliance Pack

Explore the 9 SOC2 rules for trust service criteria

Custom PDF Upload

Upload your own regulatory documents

Rule Engine

Learn how rules are evaluated

Build docs developers (and LLMs) love