Skip to main content

Personal information

Configuration​

The following configuration has to be placed in your LLM provider entity in the Guardrail Validation section.

"guardrails": [
  {
    "enabled": true,
    "before": true,
    "after": true,
    "id": "pif",
    "config": {
      "provider": "provider_28ccfcf1-85b0-4e8d-b505-6793d07c0ee3",
      "pif_items": [
        "EMAIL_ADDRESS",
        "PHONE_NUMBER",
        "LOCATION_ADDRESS",
        "NAME",
        "IP_ADDRESS",
        "CREDIT_CARD",
        "SSN"
      ]
    }
  }
]

Field Explanations​

  • enabled: true → The guardrail is active.
  • before: true → The guardrail applies before processing data.
  • after: true → The guardrail applies after LLM response.
  • id: "pif" → The identifier for this guardrail

Config Section​

  • provider: The LLM provider responsible for enforcing the guardrail.

  • pif_items: A list of sensitive data types to be filtered or protected Available pif_items :

    • EMAIL_ADDRESS
    • PHONE_NUMBER
    • LOCATION_ADDRESS
    • NAME
    • IP_ADDRESS
    • CREDIT_CARD
    • SS

Prompt example​

$ curl --request POST \
  --url 'http://guardrail-pif.oto.tools:8080/v1/chat/completions' \
  --header 'authorization: Bearer $BEARER_TOKEN' \
  --header 'content-type: application/json' \
  --data '{
  "messages": [
    {
       "role": "user",
       "content": "Could you tell me if the user email bob@example.com is present in the 'users' table please ?"
    }
  ]
}'

Response example​

{
  "id": "chatcmpl-5iTwZroMP6B3qKEWqfHLbvcVsQhucOHE",
  "object": "chat.completion",
  "created": 1743177176,
  "model": "gpt-4o-mini",
  "system_fingerprint": "fp-1HS2LIBA0cK0N8rBVi30YVUbayejNDAo",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "This message has been blocked by the 'personal-information' guardrail !"
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0,
    "completion_tokens_details": {
      "reasoning_tokens": 0
    }
  }
}