Unified Guard

One request. Multiple security checks. Orchestrate prompt injection, PII, output safety, and agent action checks.

Endpoint

POST /api/v1/guard

Request

{
  "text": "User input or AI output to analyze",
  "checks": ["prompt_injection", "pii_secrets"],
  "redact": false
}

Response

{
  "request_id": "req_abc123",
  "api": "unified-guard",
  "decision": "review",
  "checks": [
    {
      "check": "prompt_injection",
      "verdict": "allow",
      "risk": "low",
      "findings": [],
      "reasons": ["No injection patterns detected"]
    },
    {
      "check": "pii_secrets",
      "verdict": "review",
      "risk": "medium",
      "findings": [{ "category": "email", "reason": "Detected email (pii)" }],
      "reasons": ["Detected email (pii)"]
    }
  ],
  "usage_units": 2,
  "processing_time_ms": 45,
  "detector_version": "1.0.0"
}

Decision priority

Overall decision uses: block > review > allow

Usage

Each check consumes 1 usage unit. A request with 3 checks = 3 usage units.

Limitations

Maximum 4 checks per request
Text input limited to 32,000 characters
Verdicts are risk signals, not guarantees