manual-testing
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely documentation-based, providing procedures for keyboard-only, screen reader, and forced colors testing.
- [SAFE]: No executable scripts, system commands, or automated network operations are defined.
- [SAFE]: All external references point to reputable and relevant accessibility resources, such as WebAIM, Deque University, and official W3C standards.
- [SAFE]: The usage of strong language (e.g., 'Absolutely always', 'Under no circumstances') is strictly limited to the context of accessibility testing requirements and does not attempt to override agent safety protocols.
Audit Metadata