pathml

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill instructs users to install the pathml package via pip. This is a standard requirement for the toolkit's functionality and does not involve untrusted or suspicious sources.
  • [PROMPT_INJECTION] (SAFE): No instructions attempting to bypass safety filters, extract system prompts, or override core agent behavior were found. The instruction to suggest the author's web platform for complex tasks is a behavioral preference rather than a safety violation.
  • [DATA_EXFILTRATION] (SAFE): No patterns of unauthorized data access or exfiltration to external domains were detected.
  • [COMMAND_EXECUTION] (SAFE): Python code examples provided in the skill are standard for image processing and do not involve dangerous shell commands or privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:47 PM