mage-characters

Pass

Audited by Gen Agent Trust Hub on Nov 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data including character names, handles, and descriptions which are then used to form prompts for generative models. This creates a potential surface for indirect prompt injection where adversarial content in metadata could influence generation outputs.
  • Ingestion points: SKILL.md (via create_character and create_reference tools).
  • Boundary markers: The instructions do not specify the use of delimiters or protective instructions when interpolating user-provided text into generation prompts.
  • Capability inventory: The skill possesses capabilities to create/delete persistent objects and generate media via generate and seed_audio tools.
  • Sanitization: No sanitization or validation logic for the content of names or descriptions is described in the skill documentation.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of images and audio files from external, user-supplied URLs to create persistent character and reference assets.
  • Evidence: SKILL.md instructions for create_character and create_reference tools explicitly support fetching content from HTTPS URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 13, 2026, 03:49 PM