mage-character-muse

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the magick (ImageMagick) command-line tool to perform image cropping when local commands can be executed. This is a standard image manipulation task used to ensure compatibility with the required 4:5 aspect ratio.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the official Mage connector at https://mcp.mage.space/mcp. This is a vendor-owned resource used for legitimate skill functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied character descriptions. It mitigates risk by explicitly instructing the agent to respect content filters and not attempt to bypass safety guidelines.
  • Ingestion points: User descriptions in SKILL.md (Stage 2: Refine).
  • Boundary markers: None explicitly defined for user descriptions, but presets use a specific prefix to maintain context.
  • Capability inventory: generate (API call), magick (subprocess call for cropping).
  • Sanitization: Includes an explicit instruction to never attempt to bypass content filters or model-level blocks.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 07:22 AM