policy: Finance "AML Transaction Screening" Gate

[akarlaraytu]

📘 Context

A banking AI assistant needs a deterministic AML (Anti-Money Laundering) gate
to decide whether a wire transfer can be AUTHORIZED, ESCALATED, or BLOCKED before
execution.

This policy should act as a pre-execution screening layer for cross-border and
high-value transactions, enforcing FATF and FinCEN-aligned thresholds without
relying on model judgment.

🎯 The Goal

Write an AML transaction screening policy for a banking AI assistant.

Variables you can use:

• transaction_amount (Float): USD value of the transaction.
• transaction_type (Enum): "DOMESTIC", "CROSS_BORDER", "CRYPTO_OFFRAMP".
• sender_kyc_verified (Boolean): Whether sender has completed KYC.
• receiver_kyc_verified (Boolean): Whether receiver has completed KYC.
• sender_risk_score (Integer): Risk rating 0-100.
• receiver_country_sanctioned (Boolean): Whether receiver country is on sanctions list.
• pep_match (Boolean): Whether sender or receiver is a Politically Exposed Person.
• daily_volume_usd (Float): Sender's cumulative transaction volume that day.
• compliance_officer_approved (Boolean): Whether human compliance reviewer signed off.

Requirements:

• BLOCK if sender_kyc_verified is False.
• BLOCK if receiver_country_sanctioned is True.
• BLOCK if transaction_amount > 10000 and transaction_type == "CROSS_BORDER" and compliance_officer_approved is False.
• ESCALATE if pep_match is True.
• ESCALATE if transaction_amount > 50000.
• ESCALATE if sender_risk_score > 75.
• BLOCK if daily_volume_usd > 100000 and compliance_officer_approved is False.

📚 Resources

• Start here: docs/getting-started.md
• Check syntax: quickstart/01_hello_world.csl
• See how examples are structured: examples/community/

✅ Definition of Done

Submit a PR with: examples/community/finance_aml_screening.csl