gate-a11y

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external accessibility audit reports (可访问性审计报告) to guide the agent in fixing WCAG and ARIA issues. This creates a vulnerability surface where malicious instructions could be embedded within a report to influence agent actions during the code optimization process.
  • Ingestion points: Input field for accessibility audit reports defined in the skill frontmatter.
  • Boundary markers: Absent; there are no instructions provided to the agent to treat input as data only or to ignore embedded commands.
  • Capability inventory: The skill body instructs the agent to modify code, specifically HTML structure, ARIA attributes, and JavaScript focus management logic.
  • Sanitization: Absent; no validation or filtering steps are defined for the incoming audit data.
  • [NO_CODE]: The skill consists entirely of natural language instructions in Markdown format and does not include any executable scripts or complex shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:25 AM
Security Audit — agent-trust-hub — gate-a11y