Zero-dependency TypeScript SDK for the Truthlocks Anti-Fraud Identity Firewall. Evaluate risk signals, block/challenge suspicious actors, manage fraud cases, and export SIEM data -- all with full type safety and native fetch.

Installation

npm install @truthlocks/sdk-antifraud

yarn add @truthlocks/sdk-antifraud

pnpm add @truthlocks/sdk-antifraud

Quick Start

import { TruthlockRisk } from "@truthlocks/sdk-antifraud";

const client = new TruthlockRisk({
  baseUrl: "https://api.truthlocks.com",
  apiKey: process.env.TRUTHLOCKS_API_KEY!,
});

// Evaluate a risk signal
const decision = await client.evaluate({
  signal_type: "login",
  subject_id: "user-123",
  subject_type: "user",
  risk_score: 72,
  metadata: { ip: "203.0.113.42", country: "US" },
});

console.log(decision.action); // "allow" | "challenge" | "block"

Features

| Feature | Description | | ---------------------- | --------------------------------------------------- | | Risk Evaluation | Real-time risk scoring with configurable policies | | Signal Ingestion | Ingest raw risk signals from any source | | Block/Challenge | Instantly block or challenge suspicious subjects | | Quarantine | Check and manage quarantined entities | | Case Management | Create, assign, and track fraud investigation cases | | Evidence Packs | Generate evidence packs for case review | | Dashboard Metrics | Retrieve fraud dashboard KPIs and trends | | SIEM Export | Export signals in JSON, CSV, CEF, or LEEF format | | Policy Management | CRUD operations on risk policies and configs | | Deepfake Detection | Scan content for AI-generated manipulation |

API Reference

Client Initialization

const client = new TruthlockRisk({
  baseUrl: "https://api.truthlocks.com", // required
  apiKey: "tlk_live_...", // required
  timeout: 30000, // optional (ms)
});

Risk Evaluation

// Evaluate risk
const result = await client.evaluate({
  signal_type: "transaction",
  subject_id: "user-456",
  subject_type: "user",
  risk_score: 85,
});

// Ingest a raw signal
await client.ingestSignal({
  signal_source: "device-fingerprint",
  signal_type: "new_device",
  subject_id: "user-456",
  risk_score: 40,
});

Block & Challenge

// Block a suspicious subject
await client.blockOrChallenge({
  subject_id: "user-789",
  subject_type: "user",
  action: "block",
  reason: "Multiple failed payment attempts",
  duration_seconds: 3600,
});

// Check quarantine status
const status = await client.getQuarantineStatus("user-789", "user");

Case Management

// Create a fraud case
const fraudCase = await client.createCase({
  subject_id: "user-789",
  subject_type: "user",
  severity: "high",
  title: "Suspected account takeover",
});

// Generate evidence pack
const evidence = await client.generateEvidencePack(fraudCase.id);

Dashboard & SIEM

// Get dashboard metrics
const metrics = await client.getDashboardMetrics();

// Get trend data
const trends = await client.getDashboardTrends("7d");

// Export to SIEM
const siem = await client.createSIEMExport({
  format: "cef",
  since: "2026-01-01T00:00:00Z",
});

Account Takeover (ATO)

// Evaluate ATO risk
const ato = await client.evaluateATO({
  subject_id: "user-123",
  signal_type: "password_reset",
  risk_score: 90,
});

// Get ATO profile
const profile = await client.getATOProfile("user-123", "user");

Error Handling

import {
  TruthlockRiskError,
  AuthenticationError,
  RateLimitError,
} from "@truthlocks/sdk-antifraud";

try {
  await client.evaluate(request);
} catch (err) {
  if (err instanceof AuthenticationError) {
    // Invalid or expired API key
  } else if (err instanceof RateLimitError) {
    // Back off and retry
  } else if (err instanceof TruthlockRiskError) {
    console.error(`API error ${err.status}: ${err.message}`);
  }
}

Requirements

• Node.js >= 18.0.0 (uses native fetch)
• TypeScript >= 5.0 (optional, full type definitions included)

Documentation

• SDK Guide
• API Reference
• Integration Examples

License

MIT -- see LICENSE for details.