Data sanitation policy

The Data Sanitation Gateway policy scrubs sensitive information from certain fields in an API request. It provides two approaches for configuration: either by specifying entities to inspect (allowlist-style) or entities to exclude (denylist-style). This policy is ideal for ensuring that sensitive data such as credit card numbers, phone numbers, and email addresses are not passed downstream.

Prerequisites

• Active LLM Gateway instance. Refer to Installation.
• Familiarity with Gateway policy structure and configuration template.

Scenarios

1. Sensitive Data Removal: Remove sensitive data from API requests before they are sent to the provider.

Gateway policy components

Data Sanitation Processor

Gateway policy example

This is an example of a fully configured Data Sanitation Gateway policy that removes credit card and email information from requests to OpenAI.

/etc/lunar-proxy/flows/flow.yaml

name: DataSanitationFlow

filter:
  url: "api.openai.com/*"

processors:
  DataSanitation:
    processor: DataSanitation
    parameters:
      - key: blocklisted_entities
        value: 
          - CreditCard
          - Email

flow:
  request:
    - from:
        stream:
          name: globalStream
          at: start
      to:
        processor:
          name: DataSanitation           
    - from:
        processor:
          name: DataSanitation 
      to:
        stream:
          name: globalStream
          at: end
  response:
    - from:
        stream:
          name: globalStream 
          at: start        
      to:
        stream:
          name: globalStream  
          at: end