scorecard

Warn

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes dynamic context injection (the !command`` syntax) within SKILL.md to execute shell commands at the moment the skill is loaded by the agent.
  • [COMMAND_EXECUTION]: The injection command node "${CLAUDE_SKILL_DIR}/../../dead-rules-audit.js" --render executes a JavaScript file using Node.js. The path to this script is constructed dynamically using environment variables and points to a location two levels above the skill directory, meaning it executes code not provided within the skill package itself.
  • [COMMAND_EXECUTION]: While the skill includes disable-model-invocation: true to prevent the agent from calling further tools, the initial command execution happens automatically and has the same privileges as the user running the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 16, 2026, 03:49 PM
Security Audit — agent-trust-hub — scorecard