NYC

image-enhancer

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
  • PROMPT_INJECTION (HIGH): The skill instructions define an agent behavior that processes untrusted external data (images) and performs file system operations. This creates a significant surface for Indirect Prompt Injection.
  • Ingestion points: Processes user-provided images and folder contents (SKILL.md).
  • Boundary markers: Absent. No delimiters or instructions to ignore embedded content are present.
  • Capability inventory: Explicitly mentions file creation and modification ('Saved as', 'Original preserved as') and batch processing.
  • Sanitization: Absent. No validation or filtering of image content or metadata is specified.
  • NO_CODE (SAFE): No executable scripts, binaries, or configuration files for package managers were found in the skill directory.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 03:13 AM