ux-experience-audit

Warn

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill runs a local PowerShell script (ux-audit.ps1) using the -ExecutionPolicy Bypass argument. This bypasses system-level protections intended to prevent the execution of untrusted scripts on the host machine.
  • [COMMAND_EXECUTION]: The audit uses the rg command to find sensitive configuration keys such as apiKey, provider, and baseURL. This process reads potentially sensitive environmental data or hardcoded credentials into the agent's context, which may then be visible in session history.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the analysis of project source files and documentation.
  • Ingestion points: Source code in packages/ and documentation in docs/ are read and interpreted by the agent during the audit.
  • Boundary markers: There are no instructions to the agent to treat audited file content as untrusted or to ignore instructions embedded within those files.
  • Capability inventory: The skill possesses the ability to execute shell commands (rg) and PowerShell scripts.
  • Sanitization: The agent does not perform any sanitization or escaping of the data retrieved from the codebase before processing it.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 9, 2026, 06:44 AM