About risk-assessment
A compliance skill for classifying customer risk during onboarding and review. It applies a weighted four-factor model and leaves final classification approval to an analyst.
Key features
- 01Weighted geographic, customer, product, and channel factors
- 02Deterministic score calculation
- 03Four risk bands with due diligence actions
- 04Automatic escalation triggers
- 05Analyst review and sign-off
Use cases
- 01Classifying customer risk at onboarding
- 02Reassessing customers during periodic refresh
- 03Escalating PEP, sanctions, or unexplained funds concerns
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ vyayasan/ kyc-analyst --skill risk-assessmentAfter installing, type/risk-assessmentin Claude Code to confirm it loaded.
Details
From the project READMERisk Assessment Skill - v1.0.0
Deterministic heuristic risk scoring using a four-factor weighted model aligned with Basel III, EBA Guidelines, FATF standards, and FinCEN guidance.
Overview
Risk assessment is performed during onboarding and refresh to classify customers into risk bands that determine the level of due diligence required and ongoing monitoring frequency.
Four-Factor Risk Model
Factor 1: Geographic Risk (30% Weight)
| Indicator | Score Range |
|---|---|
| UK/EU/US/developed economy | 5-15 |
| Mid-tier economies | 20-40 |
| FATF grey-list jurisdictions | 50-70 |
| High-risk/sanctioned jurisdictions | 80-100 |
| Multiple jurisdictions with offshore | +10-20 |
Assessed for: Customer residence, nationality, transaction destinations, wealth origin, business operations.
Factor 2: Customer Risk (35% Weight)
| Indicator | Score Range |
|---|---|
| Salaried employee, standard profile | 5-15 |
| Self-employed, professional | 20-35 |
| Business owner, HNWI | 40-60 |
| PEP (any tier) | 60-80 |
| Complex structures, unclear ownership | 70-90 |
| Adverse media findings | +20-40 |
Assessed for: Occupation, PEP status, source of wealth clarity, business complexity, directorships, adverse media.
Factor 3: Product Risk (25% Weight)
| Indicator | Score Range |
|---|---|
| Personal current/savings account | 5-15 |
| Investment/wealth management | 20-40 |
| Trade finance, correspondent banking | 50-70 |
| Virtual assets, high-value transactions | 60-80 |
| Complex structured products | 70-90 |
Assessed for: Account type, expected activity, transaction volume, product complexity.
Factor 4: Channel Risk (10% Weight)
| Indicator | Score Range |
|---|---|
| Face-to-face, standard onboarding | 5-10 |
| Remote/digital onboarding with verification | 15-25 |
| Intermediary-introduced | 30-50 |
| Anonymous/pseudonymous channels | 70-90 |
Assessed for: Onboarding method, documentation quality, intermediary involvement.
Risk Score Calculation
Total Score = (Geographic x 0.30) + (Customer x 0.35) + (Product x 0.25) + (Channel x 0.10)
Risk Bands and Required Actions
| Band | Score | Due Diligence | Approval Level | Monitoring |
|---|---|---|---|---|
| LOW | 0-20 | Standard CDD | Analyst | 3-5 year review |
| MEDIUM | 21-60 | Enhanced monitoring | Senior Analyst | 1-2 year review |
| HIGH | 61-80 | Full EDD | Manager + MLRO | 6-12 month review |
| CRITICAL | 81-100 | Immediate escalation | Senior Management | Quarterly or reject |
Determinism Guarantee
The risk score is deterministic:
- Same inputs always produce the same score
- No subjective judgment in calculation
- Mathematical formula applied consistently
- Ensures fairness and reproducibility
- Auditable by regulators
HITL Checkpoint
While the score is calculated automatically, the analyst must:
- Review the calculated score
- Confirm the input factors are accurate
- Sign off on the risk classification
- Approve or challenge the recommended action
The human always has final say on risk classification.
Escalation Triggers
Automatic escalation regardless of score:
- PEP status identified
- Sanctions match (true or uncertain)
- Material misrepresentation detected
- Source of wealth/funds unexplained
- Multiple high-risk indicators combined
Version: 1.0.0 Author: Vyayasan