FinTechTask Prompt

Compliance Prompt for FinCEN ITIN Account Risk Screening

This task prompt configures Luna, the deepidv compliance agent, to screen your live application queues against the June 2026 Joint Advisory on non-work authorized financial risks. It sets background monitoring triggers for unverified ITIN accounts and nominee structures showing repetitive sub-threshold cross-border activity, and stamps every flag with the required SAR key term FINANCIALINTEGRITY-2026-A002. Built for fintech BSA officers and AML analysts who need advisory directives translated into running detection rules.

How to use this prompt

1
Paste the prompt into Luna inside the deepidv dashboard so it can act on your actual application queues; in Claude, ChatGPT, or Gemini it works as a planning exercise against exported queue data instead.
2
Customize the advisory reference, the SAR key term, and your sub-threshold amount cutoffs to match your institution's monitoring thresholds and the latest FinCEN guidance.
3
Expect a set of active monitoring triggers plus a flagged-account list covering unverified ITIN profiles and nominee structures, each entry tagged in the audit trail.
4
Route every flagged account into your SAR review workflow and confirm the key term carries through to the narrative before filing.
5
Re-run the prompt after each FinCEN advisory update so trigger logic stays aligned with current directives.

The prompt

Luna, evaluate our user application queues against the tracking directives outlined in the June 2026 Joint Advisory on Non-Work Authorized financial risks. Configure active background monitoring triggers to flag unverified ITIN accounts or nominee structures displaying repetitive sub-threshold cross-border financial activity, adding the required key term 'FINANCIALINTEGRITY-2026-A002' to the audit trail.

Test it in Claude or another LLM

This prompt is built for the Luna agent inside deepidv, where it sets live background monitoring triggers to flag unverified ITIN accounts and nominee structures with repetitive sub-threshold cross-border activity under the June 2026 Joint Advisory, stamping each flag with the key term FINANCIALINTEGRITY-2026-A002. Here is how to dry-run the same workflow in any LLM with synthetic queue data before you wire it to live data.

1
Paste the full prompt into Claude, ChatGPT, or Gemini and replace the direct address 'Luna,' with a role instruction like 'Act as an AML compliance analyst configuring background monitoring triggers under a FinCEN joint advisory.'
2
Paste the synthetic application-queue block below right after the prompt so the model has fake accounts to reason over instead of waiting for live data.
3
Add a line telling the model to keep the audit key term FINANCIALINTEGRITY-2026-A002 on every flagged entry so you can confirm the tagging behavior carries through.
4
Good output is a short set of named monitoring triggers (unverified ITIN, nominee structure, repetitive sub-threshold cross-border) plus a flagged-account table where each row cites the triggering pattern and carries the key term in an audit-trail field. If the model flags an account with full work-authorized verification and no structuring pattern, the rule logic is too loose.
5
Once the output shape is right, run the prompt live in the deepidv dashboard where Luna executes it against your real application queues and writes the tagged flags into the SAR review workflow.

Synthetic sample data to paste alongside the prompt

Fake test data, safe to share with any LLM. Swap in your own once the output looks right.

ITIN APPLICATION QUEUE (synthetic, do not use real data):
ACCT ACME-TEST-001 | Jane Testcase | ITIN ending 0000 | work-auth: none on file | 6 cross-border wires last 30d, each $2,900 to same SG counterparty
ACCT ACME-TEST-002 | John Sample | ITIN ending 1111 | nominee listed: Sample Holdings LLC (formed 14d ago) | 9 inbound transfers, each $1,400
ACCT ACME-TEST-003 | Maria Placeholder | SSN verified, work-auth on file | 2 domestic ACH transfers, $5,200 total
ADVISORY: June 2026 Joint Advisory on Non-Work Authorized financial risks | required key term: FINANCIALINTEGRITY-2026-A002

Pairs with on deepidv

Sources & further reading

FAQ

How do I screen ITIN accounts for FinCEN advisory compliance with AI?

Give an AI agent the advisory's tracking directives and your application queue data, then have it set triggers for unverified ITIN profiles and nominee structures with repetitive sub-threshold cross-border activity. In the deepidv dashboard, Luna runs these as live background monitoring rules rather than one-off checks, so new applications are screened as they arrive.

What is sub-threshold cross-border activity in AML monitoring?

It refers to repeated international transfers kept just under reporting thresholds, a classic structuring pattern used to avoid currency transaction reports. FinCEN advisories ask institutions to flag this behavior and, where a joint advisory specifies one, include the designated key term in SAR filings so cases can be aggregated across institutions.

Run it with live verification data

These prompts work in any LLM. Inside the deepidv dashboard, Luna, Arbiter, and Arc run them against your real sessions, screening lists, and audit trails.

Book a Demo