Trend Micro Vision One AI Application Security

Trend Micro Vision One AI Application Security’s AI Guard feature uses a configurable policy to identify risks in AI Applications, such as:

Prompt injection attacks
Toxicity, violent, and other harmful content
Sensitive Data

Setup

Create a new Vision One API Key with permissions to Call Detection API
See the AI Guard Integration Guide for details around creating your policy

Colang v1:

# config.yml
rails:
  config:
    trend_micro:
      v1_url: "https://api.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails"  # Trend Micro AI Guard Endpoint
      api_key_env_var: "V1_API_KEY"
      application_name: "my-ai-app"  # Required: Application identifier (max 64 chars, alphanumeric, hyphens, underscores)
      # Optional:
      detailed_response: true  # Set to true for detailed AI Guard results
      # For other regions, use: https://api.{region}.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails
      # where region is: eu, jp, au, in, sg, or mea
  input:
    flows:
      - trend ai guard input
  output:
    flows:
      - trend ai guard output

Colang v2:

# config.yml
colang_version: "2.x"
rails:
  config:
    trend_micro:
      v1_url: "https://api.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails"  # Trend Micro AI Guard Endpoint
      api_key_env_var: "V1_API_KEY"
      application_name: "my-ai-app"  # Required: Application identifier (max 64 chars, alphanumeric, hyphens, underscores)
      # Optional:
      detailed_response: true  # Set to true for detailed AI results
      # For other regions, use: https://api.{region}.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails
      # where region is: eu, jp, au, in, sg, or mea

# rails.co
import guardrails
import nemoguardrails.library.trend_micro
flow input rails $input_text
    trend ai guard $input_text
flow output rails $output_text
    trend ai guard $output_text