skill-author

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes local PowerShell scripts (scaffold.ps1, sync-metadata.ps1, and package.ps1) to automate file and directory management. These scripts are used for legitimate project maintenance tasks and do not execute untrusted remote code.
  • [DATA_EXFILTRATION]: No network operations or sensitive data access patterns were found. The scripts perform read and write operations strictly within the context of the skill's own directory structure.
  • [PROMPT_INJECTION]: The documentation and instructions focus on workflow steps and do not contain any patterns intended to manipulate or bypass the AI agent's core safety instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 08:36 PM
Security Audit — agent-trust-hub — skill-author