mediaserver-operation

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The script scripts/mp-mediaserver.py uses importlib.import_module() to dynamically load provider classes (e.g., Emby, Plex, Jellyfin).
  • Evidence: The _import_symbol function performs the dynamic import.
  • Mitigation: The module paths are retrieved from a static, hardcoded PROVIDER_CLASSES dictionary. The selection of which module to load is limited to a predefined set of supported providers, preventing arbitrary code loading from untrusted inputs.
  • [DATA_EXPOSURE]: The skill is designed to interact with media server APIs using local credentials stored in the MoviePilot environment.
  • Evidence: The script includes a _jsonable function and a _is_sensitive_key helper that identifies and removes sensitive fields such as password, token, apikey, and sessionid before returning any data to the AI agent.
  • Best Practice: The SKILL.md explicitly instructs the agent never to request or print hostnames, passwords, or API keys, relying on the script to handle authentication internally.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:49 AM
Security Audit — agent-trust-hub — mediaserver-operation