API Integration

KoreShield provides a simple OpenAI-compatible API that makes integration seamless. No code changes required for most applications.

OpenAI-Compatible API

KoreShield acts as a transparent proxy with the same API as OpenAI:

import openai

# Instead of direct OpenAI calls
client = openai.OpenAI(api_key="sk-your-key")

# Route through KoreShield for security
client = openai.OpenAI(
    api_key="sk-your-key",  # Your actual OpenAI key
    base_url="http://localhost:8000/v1"  # KoreShield proxy
)

# All your existing code works unchanged
response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "Hello, world!"}]
)

REST API

For direct API access, KoreShield provides a comprehensive REST API:

# Health check
curl http://localhost:8000/health

# Get security statistics
curl http://localhost:8000/api/v1/stats

# List security events
curl http://localhost:8000/api/v1/events?limit=10

SDKs and Libraries

Python SDK

from koreshield import KoreShield

# Initialize client
ks = KoreShield(
    api_key="sk-your-key",
    base_url="http://localhost:8000/v1"
)

# Secure API calls
response = ks.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "Secure prompt"}]
)

JavaScript/TypeScript SDK

import { KoreShield } from 'koreshield-sdk';

const ks = new KoreShield({
  apiKey: 'sk-your-key',
  baseURL: 'http://localhost:8000/v1'
});

const response = await ks.chat.completions.create({
  model: 'gpt-4',
  messages: [{ role: 'user', content: 'Secure prompt' }]
});

Webhook Integration

Receive real-time security events via webhooks:

webhooks:
  security_events:
    url: "https://your-app.com/webhooks/security"
    secret: "your-webhook-secret"
    events: ["prompt_injection", "policy_violation"]

  alerts:
    url: "https://slack-webhook-url"
    events: ["high_severity"]

CI/CD Integration

Integrate security scanning into your deployment pipeline:

# .github/workflows/security.yml
name: Security Scan
on: [push, pull_request]

jobs:
  security-scan:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - name: Run KoreShield Security Tests
        run: |
          pip install koreshield
          koreshield test-prompts --config security-config.yaml

Enterprise Integration

Single Sign-On (SSO)

auth:
  sso:
    provider: "okta"  # okta, auth0, azure_ad
    client_id: "your-client-id"
    client_secret: "your-client-secret"
    domain: "your-domain.com"

API Gateway Integration

Works with popular API gateways:

Kong Gateway: Custom plugins available
AWS API Gateway: Lambda integration
Azure API Management: Policy integration
NGINX: Reverse proxy configuration

Monitoring Integration

monitoring:
  datadog:
    api_key: "your-datadog-key"
    tags: ["env:production", "service:koreshield"]

  prometheus:
    enabled: true
    port: 9090
    path: "/metrics"

Migration Guide

From Direct API Calls

Install KoreShield
Configure your LLM providers
Change base_url to point to KoreShield
Test with a subset of traffic
Gradually migrate all requests

From Other Security Tools

KoreShield can complement existing security tools:

Before: App → Security Tool → LLM
After: App → KoreShield → Security Tool → LLM

Or replace them entirely for LLM-specific protection.

KoreShield stores your request history so you can analyze past traffic patterns. The retention period depends on your subscription plan:

Free Plan: 3 days retention
Ray Plan: 14 days retention
Beam Plan: 30 days retention
Pulse Plan: 90 days retention

API Integration

On this page