Skip to main content

​
SDK Version

Current Version: 0.6.4

​
Configuration Methods

Configure once and use everywhere: 
import { continum } from '@continum/sdk';

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  comply: ['GDPR', 'SOC2'],
  alerts: {
    slack: process.env.SLACK_WEBHOOK_URL,
    pagerduty: process.env.PAGERDUTY_KEY
  }
});

// Now use protect() without passing config every time
const response = await continum.protect(
  () => openai.chat.completions.create({...})
);

​
Per-Call Configuration

Pass configuration to each protect() call: 
import { protect } from '@continum/sdk';

const response = await protect(
  () => openai.chat.completions.create({...}),
  {
    apiKey: process.env.CONTINUM_API_KEY!
  },
  {
    preset: 'customer-support',
    comply: ['GDPR', 'SOC2']
  }
);

​
Configuration Options

interface ContinumConfig {
  // Required
  apiKey: string;
  
  // Automatic configuration
  preset?: Preset;
  comply?: ComplianceFramework[];
  
  // Manual sandbox configuration
  sandbox?: {
    types?: SandboxType[];
    frameworks?: ComplianceFramework[];
    customRules?: string[];
    region?: string;
    blockOn?: RiskLevel;
  };
  
  // Custom rules
  customRules?: string[];
  
  // Environment
  environment?: 'production' | 'staging' | 'development';
  region?: string;
  
  // Blocking mode
  blockOn?: RiskLevel;
  
  // Alerts
  alerts?: {
    slack?: string;
    pagerduty?: string;
    discord?: string;
    webhook?: string;
  };
  
  // Violation handlers
  onViolation?: Partial<Record<ViolationCode, ViolationHandler>>;
  onRiskLevel?: Partial<Record<RiskLevel, ViolationHandler>>;
  
  // Error handling
  onError?: (error: Error) => void;
  
  // Custom endpoint
  baseUrl?: string;
}

​
Required Configuration

​
apiKey

Your Continum API key: 
{
  apiKey: process.env.CONTINUM_API_KEY!
}
Get your API key from the dashboard.

​
Automatic Configuration

​
preset

Automatically configure detection types for your use case: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support'
}
Available presets:
  • customer-support - PII, content policy, bias
  • legal-ai - PII, legal compliance, hallucination, bias
  • fintech-ai - PII, financial compliance, security, bias
  • healthcare-ai - PII, content policy, bias, hallucination
  • coding-assistant - Security, supply chain, prompt injection
  • agent - Agent safety, prompt injection, data exfiltration
  • content-generation - Content policy, bias, hallucination
  • internal-tool - PII, security audit
  • data-pipeline - PII, data exfiltration, security
  • education-ai - Content policy, bias, hallucination
Learn more about presets →

​
comply

Specify compliance frameworks: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  comply: ['GDPR', 'SOC2', 'HIPAA']
}
Supported frameworks:
  • GDPR, CCPA, HIPAA, SOC2, ISO_27001
  • EU_AI_ACT, FINRA, PCI_DSS, NIST_AI_RMF
  • FCA, PDPA, PIPEDA, UK_DPA_2018
Learn more about compliance →

​
Manual Sandbox Configuration

For advanced use cases, configure sandboxes directly: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  sandbox: {
    types: ['PII_DETECTION', 'SECURITY_AUDIT', 'PROMPT_INJECTION'],
    frameworks: ['GDPR', 'SOC2'],
    customRules: ['CUSTOM_RULE_1'],
    region: 'eu-west-1',
    blockOn: 'HIGH'
  }
}
Available detection types:
  • PII_DETECTION, BIAS_DETECTION, SECURITY_AUDIT
  • PROMPT_INJECTION, DATA_EXFILTRATION, AGENT_SAFETY
  • HALLUCINATION_DETECTION, CONTENT_POLICY
  • FINANCIAL_COMPLIANCE, LEGAL_COMPLIANCE
  • MULTI_TURN_ATTACK, SUPPLY_CHAIN_INTEGRITY
  • FULL_SPECTRUM, CUSTOM

​
Environment Configuration

​
environment

Specify the environment: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  environment: 'production'  // 'production' | 'staging' | 'development'
}

​
region

Specify the region for data residency: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  region: 'eu-west-1'  // EU data residency
}

​
Blocking Mode

Wait for audit results before returning response: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  blockOn: 'HIGH'  // Block if risk level is HIGH or CRITICAL
}
Risk levels:
  • LOW - Block all violations
  • MEDIUM - Block MEDIUM, HIGH, CRITICAL
  • HIGH - Block HIGH, CRITICAL
  • CRITICAL - Block only CRITICAL
Learn more about blocking mode →

​
Alerts

Configure real-time alerts: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  alerts: {
    slack: process.env.SLACK_WEBHOOK_URL,      // HIGH/CRITICAL
    pagerduty: process.env.PAGERDUTY_KEY,      // CRITICAL only
    discord: process.env.DISCORD_WEBHOOK_URL,  // MEDIUM/LOW
    webhook: process.env.CUSTOM_WEBHOOK_URL    // All violations
  }
}
Learn more about alerts →

​
Violation Handlers

React to specific violations in code: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  onViolation: {
    PII_LEAK: async (signal) => {
      await sendAlert('PII detected', signal);
    },
    PROMPT_INJECTION: async (signal) => {
      await logSecurityIncident(signal);
    }
  },
  onRiskLevel: {
    CRITICAL: async (signal) => {
      await notifySecurityTeam(signal);
    }
  }
}
Learn more about violation handlers →

​
Error Handling

Handle audit errors: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  onError: (error) => {
    console.error('Audit error:', error);
    // Never throws — audit failures don't break your app
  }
}

​
Custom Endpoint

For self-hosted deployments: 
{
  apiKey: process.env.CONTINUM_API_KEY!,
  baseUrl: 'https://your-continum-instance.com'
}

​
Per-Call Overrides

Override global configuration for specific calls: 
continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  blockOn: 'HIGH'
});

// Override for this call
const response = await continum.protect(
  () => openai.chat.completions.create({...}),
  {
    blockOn: 'CRITICAL',  // Only block CRITICAL for this call
    userId: 'user_123',
    sessionId: 'session_456',
    metadata: { feature: 'chat' }
  }
);

​
Environment-Specific Configuration

​
Development

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  environment: 'development'
});

​
Staging

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  environment: 'staging',
  alerts: {
    slack: process.env.SLACK_WEBHOOK_URL
  }
});

​
Production

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  comply: ['GDPR', 'SOC2'],
  environment: 'production',
  blockOn: 'HIGH',
  alerts: {
    slack: process.env.SLACK_WEBHOOK_URL,
    pagerduty: process.env.PAGERDUTY_KEY
  }
});

​
Configuration Best Practices

​
1. Use Environment Variables

Never hardcode API keys: 
// ❌ Bad
{ apiKey: 'co_abc123...' }

// ✅ Good
{ apiKey: process.env.CONTINUM_API_KEY! }

​
2. Start with Presets

Use presets instead of manual configuration: 
// ✅ Good
{ preset: 'customer-support', comply: ['GDPR'] }

// ❌ Avoid (unless you have specific requirements)
{ sandbox: { types: ['PII_DETECTION', 'BIAS_DETECTION', ...] } }

​
3. Validate Configuration

Check configuration at startup: 
if (!process.env.CONTINUM_API_KEY) {
  throw new Error('CONTINUM_API_KEY is required');
}

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY,
  preset: 'customer-support'
});

​
4. Use Global Configuration

Configure once, use everywhere: 
// continum.ts
import { continum } from '@continum/sdk';

continum.configure({
  apiKey: process.env.CONTINUM_API_KEY!,
  preset: 'customer-support',
  comply: ['GDPR', 'SOC2']
});

export { continum };

// app.ts
import { continum } from './continum';

const response = await continum.protect(
  () => openai.chat.completions.create({...})
);

​
Next Steps

Presets

Explore presets and compliance frameworks

Alerts

Set up real-time alerts

Blocking Mode

Block requests based on violations

Configuration

Advanced SDK configuration