Zscaler AI Guard icon

Zscaler AI Guard

Scan AI prompts and responses for security threats using Zscaler AI Guard

Actions3

Prompt Scan

AI-generated

Summary

Scans a single text prompt for security threats using Zscaler AI Guard API.

Inputs

  • content (required) — The text content to scan for security threats.
  • aiModel — AI model identifier for metadata (default: n8n-integration).
  • applicationName — Application name for audit trails (default: n8n-workflow).
  • environment — Environment identifier for attribution and tracking (optional).
  • maxRetries — Maximum number of retry attempts for failed requests (default: 3, capped at 6).
  • policyIdOverride — Override the default policy ID from credentials. Leave empty to use credential policy or auto-resolution.
  • timeout — Request timeout in milliseconds (default: 30000).
  • transactionId — Custom transaction ID for tracking. If empty, one will be generated.
  • userId — User identifier for audit trails (default: n8n-user).

Output shape

A single item containing the scan result, detectors triggered, and blocking status.

The response includes severity, detectors triggered, and blockingDetectors. If content exceeds 5MB, an ApplicationError is thrown. On failure with 'Continue on Error' enabled, the output item includes an error message and sets action to BLOCK.

Examples

Example 1: Scan a user prompt for malware.

Content set to user input, maxRetries set to 2.

Example 2: Perform a blocked scan for policy violations.

Policy ID override set to force a specific policy check.

Links

Discussion