image-manipulation-image-magick

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates image processing by providing templates that execute the magick (ImageMagick) binary through shell (Bash) and PowerShell commands as documented in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data which presents a potential injection surface.
  • Ingestion points: Local filesystem paths and image metadata are ingested through commands like Get-ChildItem and magick identify in SKILL.md.
  • Boundary markers: No specific delimiters are used in the command templates.
  • Capability inventory: The skill executes system commands via the magick executable and performs file system traversal.
  • Sanitization: The skill includes proactive security guidance in the 'Guidelines' section of SKILL.md, specifically advising the agent to 'Always quote file paths' to mitigate shell-based command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:35 AM
Security Audit — agent-trust-hub — image-manipulation-image-magick