Scan Response
Scan an LLM's output for PII leaks and sensitive data before returning it to the user.
POST
/v1/scan/responseScan an AI-generated response for leaked PII, education records, and other sensitive information.
Request Body
| Parameter | Type | Required | Description |
|---|---|---|---|
response | string | Yes | The LLM response text to scan |
Response
200 OK
{
"clean": false,
"signals": [
{
"domain": "pii",
"type": "email",
"confidence": 0.95,
"weight": 30
}
],
"sanitized_response": "The user's email is [REDACTED]..."
}Examples
cURL
curl -X POST https://indigiarmor.com/v1/scan/response \
-H "Authorization: Bearer ia_sk_..." \
-H "Content-Type: application/json" \
-d '{"response": "The user\'s email is john@test.com and their SSN is 123-45-6789."}'SDK
const llmOutput = await callYourLLM(prompt);
const check = await armor.scanResponse(llmOutput);
if (check.clean) {
return llmOutput; // Safe to show
} else {
return check.sanitized_response; // Use sanitized version
}