aml-kyc
Skill9 starsChanged 8 months agoMITView on GitHub
Customer due diligence, suspicious activity reporting, PEP screening, sanctions — Anti-Money Laundering and Know Your Customer compliance. Simplified Due Diligence (SDD) — low-risk customers: - Permitted only when ML/TF risk is demonstrably low - Reduced identification requirements but identity must still be established - Examples: regulated financial institutions, listed companies, government entities - Still requires ongoing monitoring (at reduced frequency) Customer Due Diligence (CDD) — standard tier: - Identification and verification of customer identity (natural persons: name, DOB, address, ID…
In this repository52 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- consolidationSkill
- deferred-taxesSkill
- gaap-standardsSkill
- hgb-standardsSkill
- ifrs-standardsSkill
- lease-accountingSkill
- revenue-recognitionSkill
- balanced-scorecardSkill
- deckungsbeitragsrechnungSkill
- investitionsrechnungSkill
- kostenrechnungSkill
- basel-iiiSkill
- mariskSkill
- sox-complianceSkill
- kpi-systemsSkill
- wertschoepfungskettenSkill
- covenant-analysisSkill
- credit-scoring-modelsSkill
- rating-methodologiesSkill
- econometricsSkill
- fiscal-policySkill
- labor-marketsSkill
- monetary-policy-analysisSkill
- trade-economicsSkill
- earnings-modelsSkill
- sector-analysisSkill
- budgetingSkill
- cash-flow-forecastingSkill
- rolling-forecastsSkill
- scenario-planningSkill
- variance-analysisSkill
- solvency-iiSkill
- comparable-analysisSkill
- dcf-valuationSkill
- deal-structuringSkill
- lbo-modelingSkill
- pitch-book-creationSkill
- precedent-transactionsSkill
- corporate-tax-planningSkill
- international-taxSkill
- transfer-pricingSkill
- vat-optimizationSkill
- working-capitalSkill
- cap-table-analysisSkill
- lbo-modeling-detailSkill
- portfolio-management-peSkill
- startup-valuationSkill
- term-sheet-analysisSkill
- asset-allocationSkill
- estate-planningSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Must be completed before establishing business relationship
- Distinctive: line 67
All PEPs require EDD
- Distinctive: line 79
If no natural person identified above threshold: identify persons exercising control through other means
- Distinctive: line 103
No tipping-off: Customer must not be informed about the report
- Distinctive: line 104
Document retention: All records related to the STR for minimum 5 years
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AML/KYC Compliance
Customer due diligence, suspicious activity reporting, PEP screening, sanctions — Anti-Money Laundering and Know Your Customer compliance.
When to Activate
- Customer onboarding and KYC process design
- Risk classification of customers (CDD, EDD, SDD)
- PEP (Politically Exposed Persons) screening procedures
- Beneficial ownership identification and verification
- Suspicious transaction reporting (STR/SAR)
- Sanctions screening implementation
- Ongoing monitoring and periodic review design
- Travel rule compliance for fund transfers
- AML program assessment or audit preparation
Core Concepts
Customer Due Diligence Tiers
Simplified Due Diligence (SDD) — low-risk customers:
- Permitted only when ML/TF risk is demonstrably low
- Reduced identification requirements but identity must still be established
- Examples: regulated financial institutions, listed companies, government entities
- Still requires ongoing monitoring (at reduced frequency)
Customer Due Diligence (CDD) — standard tier:
- Identification and verification of customer identity (natural persons: name, DOB, address, ID document; legal entities: name, registration, registered address, directors)
- Identification of beneficial owners (typically > 25% ownership threshold)
- Understanding the purpose and intended nature of the business relationship
- Ongoing monitoring of transactions
- Must be completed before establishing business relationship
Enhanced Due Diligence (EDD) — high-risk situations:
- Mandatory for: PEPs, high-risk countries (FATF grey/black list), complex ownership structures, correspondent banking, unusual transaction patterns
- Additional measures: source of wealth, source of funds, senior management approval, enhanced ongoing monitoring, more frequent reviews
Customer Risk Classification
Risk factors to assess:
| Category | Higher Risk Indicators |
| Customer type | Cash-intensive business, MSB, trust, shell company, PEP |
| Geography | FATF grey/black list, high corruption (CPI < 40), tax haven |
| Product/service | Private banking, correspondent banking, trade finance, crypto |
| Transaction | Unusual patterns, structuring, rapid movement of funds |
| Channel | Non-face-to-face, third-party introducers |
Risk scoring model:
- Assign risk weights to each factor category
- Score each factor (1 = low, 2 = medium, 3 = high)
- Weighted total determines overall risk tier
- Risk tier determines CDD level and review frequency
PEP Screening
PEP categories:
- Heads of state, senior politicians, senior government officials
- Senior military officers, judiciary members
- Senior executives of state-owned enterprises
- Important political party officials
- Family members of PEPs (spouse, children, parents, siblings)
- Close associates of PEPs (known business partners, beneficial owners of entities jointly held)
PEP handling:
- All PEPs require EDD
- Senior management approval for relationship establishment and continuation
- Establish source of wealth and source of funds
- Enhanced ongoing monitoring
- PEP status does not automatically decline — risk-based approach applies
- De-PEP: After leaving office, typically 12-24 months continued EDD
Beneficial Ownership
Identification requirements:
- Identify all natural persons who ultimately own or control > 25% of the entity (threshold varies by jurisdiction; EU: 25%, some jurisdictions: 10%)
- For complex structures: trace through multiple layers of ownership
- If no natural person identified above threshold: identify persons exercising control through other means
- Last resort: identify senior managing official
Verification:
- Corporate registry extracts, shareholder registers
- Annual returns, constitutional documents
- Trust deeds for trust structures
- Declarations from the customer, corroborated by independent sources
Suspicious Transaction Reporting
Red flags for suspicious activity:
- Transactions inconsistent with customer profile or stated purpose
- Structuring (splitting transactions to avoid reporting thresholds)
- Rapid movement of funds with no apparent business rationale
- Transactions involving high-risk jurisdictions without business justification
- Reluctance to provide information, use of nominees
- Unusual cash transactions, round-amount transfers
Reporting process:
- Front-line staff identifies unusual activity
- Internal report to MLRO (Money Laundering Reporting Officer)
- MLRO assesses and decides whether to file external report
- STR/SAR filed with Financial Intelligence Unit (FIU) — in Germany: Zentralstelle für Finanztransaktionsuntersuchungen (FIU)
- No tipping-off: Customer must not be informed about the report
- Document retention: All records related to the STR for minimum 5 years
Sanctions Screening
- Screen customers, beneficial owners, and counterparties against sanctions lists
- Key lists: UN, EU, OFAC (US), UK HMT, national lists
- Screen at onboarding, ongoing (batch screening), and per-transaction (real-time)
- Fuzzy matching for name variations, transliterations, aliases
- Disposition of hits: true match vs. false positive — document rationale
- Sanctions are absolute prohibitions (unlike AML, which is risk-based)
EU AML Directives (5th and 6th AMLD)
5th AMLD (effective 2020):
- Beneficial ownership registers publicly accessible
- Crypto-asset providers brought into scope
- Prepaid card limits reduced (EUR 150 for anonymous use)
- Enhanced EDD for high-risk third countries
- Central bank account registries
6th AMLD (effective 2021):
- Harmonized list of 22 predicate offences including tax crimes and cybercrime
- Criminal liability for legal persons
- Extended aiding, abetting, inciting, and attempting
- Minimum 4-year imprisonment for ML offences
- Enhanced cooperation between FIUs
Methodology
AML Program Design
- Risk assessment: Enterprise-wide ML/TF risk assessment covering customers, products, channels, geographies
- Policies and procedures: Written AML/CFT policies approved by senior management
- Customer due diligence: CDD/EDD/SDD procedures with clear escalation paths
- Transaction monitoring: Rules-based and/or AI-based detection of suspicious patterns
- Screening: Sanctions, PEP, and adverse media screening at onboarding and ongoing
- Reporting: Internal escalation procedures and external STR/SAR filing
- Record keeping: Minimum 5 years after end of business relationship
- Training: Risk-based training program for all relevant staff
- Independent audit: Regular independent review of AML program effectiveness
- Governance: Designated MLRO with direct board access
Ongoing Monitoring Design
- Transaction monitoring rules: Define scenarios (e.g., cash threshold, rapid movement, structuring patterns)
- Periodic review frequency: High risk — annually; Medium risk — every 2-3 years; Low risk — every 5 years
- Trigger events: Change in customer profile, adverse media, unusual transaction
- Alert management: Triage, investigation, escalation, disposition, documentation
Templates
Customer Risk Assessment
Customer: _______________ Date: ___________ Analyst: ___________
Risk Factor Weight Score (1-3) Weighted Score
Customer type 25% ____ ____
Geography 25% ____ ____
Product/service 20% ____ ____
Transaction profile 15% ____ ____
Channel 15% ____ ____
Total: ____
Risk Tier: [ ] Low (SDD) [ ] Medium (CDD) [ ] High (EDD)
Review frequency: _______________
Approved by: _______________EDD Checklist
Customer: _______________ PEP: [ ] Yes [ ] No
[ ] Source of wealth documented and verified
[ ] Source of funds for the relationship documented
[ ] Senior management approval obtained
[ ] Enhanced transaction monitoring activated
[ ] Adverse media search completed — findings: _______________
[ ] Beneficial ownership verified through independent sources
[ ] Purpose of relationship clearly understood and documented
[ ] Review frequency set to: _______________
[ ] Next review date: _______________Quality Gate
- Enterprise-wide ML/TF risk assessment is current (updated annually)
- CDD tiers (SDD/CDD/EDD) are clearly defined with escalation criteria
- PEP screening covers family members and close associates
- Beneficial ownership is traced to natural persons above threshold
- Sanctions screening covers all relevant lists with fuzzy matching
- Transaction monitoring rules are calibrated (false positive rate managed)
- STR/SAR filing process ensures no tipping-off
- Record retention meets minimum 5-year requirement
- Staff training is risk-based and documented
- Independent audit of AML program performed regularly
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

