skills/garrytan/gstack/guard/Gen Agent Trust Hub

guard

Warn

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The setup instructions for the skill contain a shell command FREEZE_DIR=$(cd "<user-provided-path>" 2>/dev/null && pwd) that directly incorporates user input from an AskUserQuestion tool. This pattern is susceptible to command injection if a user provides a path containing shell metacharacters such as backticks, semicolons, or pipe operators.
  • [COMMAND_EXECUTION]: The skill uses PreToolUse hooks to execute scripts located in sibling directories (../careful/bin/check-careful.sh and ../freeze/bin/check-freeze.sh). This creates a dependency on executable files located outside the skill's own package structure.
  • [COMMAND_EXECUTION]: The skill performs automated local logging by executing shell commands to append JSON-formatted usage data to ~/.gstack/analytics/skill-usage.jsonl.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 22, 2026, 08:08 AM