heyeddi-local-first

Pass

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: SAFEPERSISTENCECOMMAND_EXECUTION
Full Analysis
  • [PERSISTENCE]: The skill provides scripts/install_user_hook.py, which automates the setup of IDE-level hooks in ~/.cursor/hooks.json. These hooks persist across IDE restarts to ensure that shell commands and MCP tool calls are always passed through the security classifier.
  • [COMMAND_EXECUTION]: The skill includes a helper module scripts/_skill_cli.py which contains a run_command function utilizing subprocess.run. This is a standard utility pattern in this vendor's skills, and the primary tool classify_command.py itself does not execute external commands.
  • [SAFE]: The core logic in scripts/classify_command.py is a defensive security implementation. It uses regex and tokenization to identify and block potentially destructive operations (e.g., terraform apply, gcloud delete) while allowing safe operations (e.g., terraform plan) after a warning is issued to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 28, 2026, 09:32 PM