Skip to main content

AWS Lambda Integration

Use KoreShield as a security proxy for LLM requests made from AWS Lambda. Deploy KoreShield separately and route requests through its OpenAI-compatible endpoint.

Use Cases​

  • Serverless APIs that need LLM safety controls
  • Event-driven flows (SQS, API Gateway) with centralized policy enforcement
  • Shared deployments that require request auditing

Prerequisites​

  • A running KoreShield instance
  • Provider API key configured on the KoreShield server

Environment Variables​

Set these in your Lambda configuration:

KORESHIELD_BASE_URL=https://your-koreshield-instance.com
KORESHIELD_API_KEY=your-koreshield-api-key

Example Lambda (Node.js)​

import { createClient } from "koreshield";

const client = createClient({
  baseURL: process.env.KORESHIELD_BASE_URL,
  apiKey: process.env.KORESHIELD_API_KEY
});

export const handler = async (event: any) => {
  const body = typeof event.body === "string" ? JSON.parse(event.body) : event.body;

  const result = await client.createChatCompletion({
    model: "gpt-4o",
    messages: body.messages
  });

  return {
    statusCode: 200,
    headers: { "content-type": "application/json" },
    body: JSON.stringify(result)
  };
};

Operational Tips​

  • Keep provider API keys on the KoreShield server, not inside the Lambda.
  • Use VPC or private networking if your KoreShield endpoint is private.
  • Increase Lambda timeout for streaming responses.
  • Adjust policies in /configuration/.

Troubleshooting​

  • 401 responses: confirm KORESHIELD_API_KEY in Lambda env
  • Timeouts: raise Lambda timeout and enable streaming
  • Network errors: check VPC egress and NAT configuration

Next Steps​