project-principles

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [No Code] (SAFE): The skill consists entirely of Markdown instructions and metadata. It does not include any scripts (Python, JavaScript, Shell) or executable commands.
  • [Remote Code Execution] (SAFE): No external scripts are downloaded or executed. The skill only references relative local file paths for redirection.
  • [Data Exposure] (SAFE): There are no patterns suggesting access to sensitive system files, environment variables, or hardcoded credentials.
  • [Prompt Injection] (SAFE): The instructions provide routing logic for the agent without attempting to override safety filters or extract system prompts.
  • [External Downloads] (SAFE): No external URLs or package managers are invoked.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:26 PM