claude-code-command-author

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for generating Claude Code custom commands that can execute bash scripts using the ! prefix. It effectively mitigates risk by requiring the use of the allowed-tools configuration to restrict execution to specific, non-destructive commands (e.g., git status).
  • [DATA_EXFILTRATION]: The templates and instructions provided in the skill explicitly prohibit the inclusion of commands that perform external network operations (e.g., curl, wget), preventing the exfiltration of source code or environment variables.
  • [SAFE]: The skill includes a 'Guardrails' section and a 'Quality Checklist' that prevent the generation of commands with destructive capabilities (e.g., rm, sudo, deploy) without explicit user consent, ensuring the generated tools follow the principle of least privilege.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 03:30 PM