UNPKG

claude-flow

Version:

Ruflo - Enterprise AI agent orchestration for Claude Code. Deploy 60+ specialized agents in coordinated swarms with self-learning, fault-tolerant consensus, vector memory, and MCP integration

344 lines (275 loc) • 11.3 kB
# @claude-flow/plugin-financial-risk [![npm version](https://img.shields.io/npm/v/@claude-flow/plugin-financial-risk.svg)](https://www.npmjs.com/package/@claude-flow/plugin-financial-risk) [![npm downloads](https://img.shields.io/npm/dm/@claude-flow/plugin-financial-risk.svg)](https://www.npmjs.com/package/@claude-flow/plugin-financial-risk) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) A high-performance financial risk analysis plugin combining sparse inference for efficient market signal processing with graph neural networks for transaction network analysis. The plugin enables real-time anomaly detection, portfolio risk scoring, and automated compliance reporting while maintaining the explainability required by financial regulators (SEC, FINRA, Basel III). ## Features - **Portfolio Risk Analysis**: Calculate VaR, CVaR, Sharpe ratio, Sortino ratio, and max drawdown - **Anomaly Detection**: Detect fraud, AML violations, and market manipulation in transactions - **Market Regime Classification**: Identify current market regime through historical pattern matching - **Compliance Checking**: Automated verification against Basel III, MiFID II, Dodd-Frank, AML, and KYC - **Stress Testing**: Run historical and hypothetical stress scenarios on portfolios ## Installation ### npm ```bash npm install @claude-flow/plugin-financial-risk ``` ### CLI ```bash npx claude-flow plugins install --name @claude-flow/plugin-financial-risk ``` ## Quick Start ```typescript import { FinancialRiskPlugin } from '@claude-flow/plugin-financial-risk'; // Initialize the plugin const finance = new FinancialRiskPlugin({ marketDataPath: './data/market', modelPath: './data/models', auditLogPath: './logs/audit' }); // Analyze portfolio risk const risk = await finance.portfolioRisk({ holdings: [ { symbol: 'AAPL', quantity: 100, assetClass: 'equity' }, { symbol: 'GOOGL', quantity: 50, assetClass: 'equity' }, { symbol: 'TLT', quantity: 200, assetClass: 'bond' } ], riskMetrics: ['var', 'cvar', 'sharpe'], confidenceLevel: 0.95, horizon: '1d' }); // Detect anomalies in transactions const anomalies = await finance.anomalyDetect({ transactions: [ { id: 'tx-001', amount: 50000, timestamp: '2024-01-15T10:30:00Z', parties: ['ACC-123', 'ACC-456'] }, { id: 'tx-002', amount: 1000000, timestamp: '2024-01-15T10:31:00Z', parties: ['ACC-123', 'ACC-789'] } ], sensitivity: 0.8, context: 'fraud' }); // Check regulatory compliance const compliance = await finance.complianceCheck({ entity: 'FUND-001', regulations: ['basel3', 'mifid2'], scope: 'capital', asOfDate: '2024-01-15' }); ``` ## MCP Tools ### 1. `finance/portfolio-risk` Calculate comprehensive portfolio risk metrics. ```typescript const result = await mcp.invoke('finance/portfolio-risk', { holdings: [ { symbol: 'SPY', quantity: 1000, assetClass: 'equity' }, { symbol: 'BND', quantity: 500, assetClass: 'bond' }, { symbol: 'GLD', quantity: 200, assetClass: 'commodity' } ], riskMetrics: ['var', 'cvar', 'sharpe', 'max_drawdown'], confidenceLevel: 0.99, horizon: '1w' }); // Returns: // { // var: { value: 0.023, confidenceLevel: 0.99 }, // cvar: { value: 0.031 }, // sharpe: { value: 1.45 }, // maxDrawdown: { value: 0.082, period: '2023-10-01 to 2023-10-15' } // } ``` ### 2. `finance/anomaly-detect` Detect anomalies in financial transactions and market data. ```typescript const result = await mcp.invoke('finance/anomaly-detect', { transactions: [ { id: 'tx-12345', amount: 250000, timestamp: '2024-01-15T14:30:00Z', parties: ['CORP-A', 'CORP-B'], metadata: { type: 'wire', currency: 'USD' } } ], sensitivity: 0.9, context: 'aml' }); // Returns: // { // anomalies: [{ // transactionId: 'tx-12345', // score: 0.87, // reasons: ['unusual_amount', 'first_time_counterparty', 'velocity_spike'], // recommendation: 'review' // }] // } ``` ### 3. `finance/market-regime` Identify current market regime through pattern matching. ```typescript const result = await mcp.invoke('finance/market-regime', { marketData: { prices: [100, 102, 99, 101, 103, 105, 104], volumes: [1000000, 1200000, 900000, 1100000, 1300000, 1500000, 1400000], volatility: [0.15, 0.16, 0.18, 0.17, 0.14, 0.13, 0.14] }, lookbackPeriod: 252, regimeTypes: ['bull', 'bear', 'high_vol', 'crisis'] }); // Returns: // { // currentRegime: 'bull', // confidence: 0.82, // similarHistoricalPeriods: ['2017-Q3', '2019-Q4'], // transitionProbabilities: { bull: 0.75, bear: 0.15, high_vol: 0.10 } // } ``` ### 4. `finance/compliance-check` Automated regulatory compliance verification. ```typescript const result = await mcp.invoke('finance/compliance-check', { entity: 'BANK-001', regulations: ['basel3', 'dodd_frank'], scope: 'capital', asOfDate: '2024-01-15' }); // Returns: // { // compliant: true, // metrics: { // tier1Capital: { value: 0.125, requirement: 0.06, status: 'pass' }, // leverageRatio: { value: 0.05, requirement: 0.03, status: 'pass' } // }, // warnings: ['tier1 buffer approaching minimum'] // } ``` ### 5. `finance/stress-test` Run stress testing scenarios on portfolios. ```typescript const result = await mcp.invoke('finance/stress-test', { portfolio: { holdings: [ { symbol: 'SPY', quantity: 1000 }, { symbol: 'QQQ', quantity: 500 } ] }, scenarios: [ { name: '2008 Financial Crisis', type: 'historical', shocks: {} }, { name: 'Interest Rate +300bp', type: 'hypothetical', shocks: { rateShock: 0.03 } } ], metrics: ['pnl', 'var_change', 'margin_call_risk'] }); ``` ## Configuration Options ```typescript interface FinancialRiskConfig { // Data paths marketDataPath: string; // Path to market data cache modelPath: string; // Path to trained models // Compliance auditLogPath: string; // Path for SOX-compliant audit logs regulatoryReporting: boolean; // Enable auto-generated reports // Performance maxMemoryMB: number; // WASM memory limit (default: 1024) maxCpuTimeSeconds: number; // Operation timeout (default: 60) // Security encryptionEnabled: boolean; // AES-256 encryption (default: true) hsmKeyId?: string; // HSM key identifier for production // Rate limiting rateLimits: { requestsPerMinute: number; maxConcurrent: number; }; } ``` ## Security Considerations ### Regulatory Compliance This plugin is designed for financial regulatory compliance: | Requirement | Implementation | |-------------|----------------| | **PCI-DSS Compliance** | No storage of PAN/CVV in plugin memory | | **SOX Compliance** | Immutable audit logs for all risk calculations | | **Data Encryption** | AES-256 for data at rest, TLS 1.3 in transit | | **Key Management** | HSM or secure enclave for cryptographic keys | | **Segregation of Duties** | Separate roles for trading, risk, and compliance | ### Role-Based Access Control ```typescript const roles = { TRADER: ['portfolio-risk', 'market-regime'], RISK_MANAGER: ['portfolio-risk', 'anomaly-detect', 'stress-test', 'market-regime'], COMPLIANCE_OFFICER: ['compliance-check', 'anomaly-detect'], AUDITOR: ['compliance-check'], // Read-only with full audit access QUANT: ['portfolio-risk', 'market-regime', 'stress-test'] }; ``` ### Audit Logging (SOX/MiFID II) All calculations are logged with: - Timestamp with microsecond precision - User ID and role - Transaction IDs affected - Portfolio state hash - Model version used - Input/output hashes for reproducibility Logs are retained for 7 years per MiFID II requirements and available for regulatory inspection within 72 hours. ### WASM Security Constraints | Constraint | Value | Rationale | |------------|-------|-----------| | Memory Limit | 1GB max | Handle large transaction datasets | | CPU Time Limit | 60 seconds | Allow complex risk calculations | | No Network Access | Enforced | Prevent data exfiltration | | No File System Write | Enforced | Analysis-only mode | | Sandboxed Data Access | Enforced | No direct database queries | ### Input Validation All inputs are validated using Zod schemas: - Stock symbols: `/^[A-Z0-9.]{1,10}$/` - Position quantities: between -1 billion and 1 billion - Transaction amounts: between -1 trillion and 1 trillion - Batch limits: maximum 100,000 transactions per request - Timestamps: ISO 8601 format, within reasonable date range - Entity identifiers: Alphanumeric with limited special characters ### Rate Limiting ```typescript const rateLimits = { 'portfolio-risk': { requestsPerMinute: 60, maxConcurrent: 5 }, 'anomaly-detect': { requestsPerMinute: 100, maxConcurrent: 10 }, 'stress-test': { requestsPerMinute: 10, maxConcurrent: 2 }, 'market-regime': { requestsPerMinute: 120, maxConcurrent: 10 }, 'compliance-check': { requestsPerMinute: 30, maxConcurrent: 3 } }; ``` ## Performance | Metric | Target | |--------|--------| | Portfolio VaR calculation | <100ms for 10K positions | | Transaction anomaly scoring | <5ms per transaction | | Market regime classification | <50ms for 1-year history | | Compliance check | <1s for full entity scan | ## Dependencies - `micro-hnsw-wasm`: Fast similarity search for historical pattern matching - `ruvector-sparse-inference-wasm`: Efficient processing of sparse financial features - `ruvector-gnn-wasm`: Transaction network analysis for fraud detection - `ruvector-economy-wasm`: Token economics and market microstructure modeling - `ruvector-learning-wasm`: Reinforcement learning for adaptive risk thresholds ## Related Plugins | Plugin | Description | Use Case | |--------|-------------|----------| | [@claude-flow/plugin-legal-contracts](../legal-contracts) | Contract analysis | Financial agreements, derivatives documentation | | [@claude-flow/plugin-healthcare-clinical](../healthcare-clinical) | Clinical decision support | Healthcare portfolio analysis | | [@claude-flow/plugin-perf-optimizer](../perf-optimizer) | Performance optimization | High-frequency trading latency optimization | ## License MIT License Copyright (c) 2026 Claude Flow Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.