fixing-accessibility

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill initiates the installation of its full version from an external GitHub repository (ibelick/ui-skills) using the npx skills add command. This is a standard procedure for bootstrapping community-contributed skills.
  • [COMMAND_EXECUTION]: Shell commands are provided to perform environment checks (detecting global vs project scope) and to execute the installation command. These are limited to the intended setup process of the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because its primary function is to audit and remediate UI component source code, which is an untrusted external input.
  • Ingestion points: Reads external UI component code (HTML, React, etc.) during accessibility audits.
  • Boundary markers: None specified in the stub file.
  • Capability inventory: The skill is designed to apply code-level fixes, implying file-writing capabilities on the local filesystem.
  • Sanitization: No sanitization or validation of the input code is documented in this installation stub.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 09:34 PM