Skip to main content

Overview

Rogue’s red teaming engine tests AI agents against a comprehensive catalog of 87+ vulnerability types organized into 13 categories. Each vulnerability includes predefined attack techniques that are most effective for exploitation.

Vulnerability Categories

Tests for generation of harmful, inappropriate, or dangerous content.
Tests for exposure of personally identifiable information.
Tests for code injection and technical exploitation.
Tests for discriminatory or biased responses.
Tests for prompt manipulation and system compromise.
Tests for authorization and permission bypass.
Tests for business rule violations and misalignment.
Tests for IP violations and theft.
Tests for factual accuracy and reliability.
Tests for regulatory compliance violations.
Tests for critical and dangerous content.
Tests for AI agent architecture vulnerabilities.
Tests for resource exhaustion and denial of service.

Default Attack Mappings

Each vulnerability has default attacks that are most effective:

Accessing the Catalog

Vulnerability Definition Structure