amy-expert

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for standard local development tasks using shell commands, including git mv for refactoring and grep for logic identification. These are appropriate for the skill's primary purpose as a developer assistant.
  • [SAFE]: The skill mandates architectural integrity through a 'thin assembly layer' rule, preventing sensitive business or protocol logic from residing in the CLI module. It promotes security best practices by requiring use of the SecureFileIO utility for tightening file permissions on local state directories (~/.amy) to 0600/0700.
  • [SAFE]: The instructions explicitly forbid interactive input (e.g., readLine) and instead require all sensitive data such as keys and passwords to be handled via flags, which facilitates non-interactive and safer automation environments.
  • [SAFE]: The skill defines a dual-output contract that strictly separates human-readable logs on stderr from machine-stable JSON on stdout, preventing potential log-injection confusion during machine consumption of CLI results.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 05:22 PM
Security Audit — agent-trust-hub — amy-expert