duplicate-id-aria
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely informational, consisting of markdown documentation for accessibility compliance. No scripts or command-line instructions are included.
- [SAFE]: All external URLs point to legitimate accessibility resources and educational sites such as Deque University and Microsoft Accessibility Insights.
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to review external HTML code (untrusted data). 1. Ingestion points: Rendered HTML and interactive components mentioned in SKILL.md. 2. Boundary markers: None provided in instructions. 3. Capability inventory: No dangerous tools or commands (file writes, network, shell execution) are used. 4. Sanitization: No sanitization of external content is defined. The lack of dangerous capabilities mitigates the risk of this surface.
Audit Metadata