skills/simota/agent-skills/wield/Gen Agent Trust Hub

wield

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides extensive documentation on macOS security models (TCC/Automation/Accessibility) and instructs the agent to handle permission errors (like -1743) gracefully.
  • [SAFE]: Implements a mandatory 'safety_review' and 'DRY-RUN' workflow for all destructive operations such as file deletion or sending messages.
  • [COMMAND_EXECUTION]: The skill uses 'osascript', 'do shell script', and language-specific subprocess calls (Python/Node.js) to execute automation logic. These are standard tools for the stated purpose of macOS automation.
  • [SAFE]: Explicitly warns against insecure practices like interpolating untrusted input into scripts or recommending the disabling of System Integrity Protection (SIP) or TCC.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 10:17 AM
Security Audit — agent-trust-hub — wield