paper-profile

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely locally within the user's paper directory and does not perform any network operations.
  • [SAFE]: Data persistence is handled via a specialized Python script (profile_io.py) that implements strict validation against a closed vocabulary for critical fields, preventing the introduction of malformed or unexpected data.
  • [SAFE]: Free-text inputs are subject to length constraints (500 characters), which mitigates potential abuse of the configuration file as an indirect injection vector for downstream skills.
  • [SAFE]: The skill relies only on Python standard library modules, avoiding supply chain risks associated with third-party dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 01:03 PM
Security Audit — agent-trust-hub — paper-profile