processing-images
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents methods for processing external image files and inspecting their metadata (e.g., EXIF), which establishes a vulnerability surface for indirect prompt injection.
- Ingestion points: Image data and metadata ingested via tools such as
identify,convert,Pillow, andOpenCV(documented inSKILL.md). - Boundary markers: Absent; no instructions are provided to sanitize or isolate instructions that might be embedded in image metadata.
- Capability inventory: Guidance on shell execution for ImageMagick and ffmpeg and Python scripts for image manipulation is provided in
SKILL.md. - Sanitization: Absent; the content provides functional examples and does not include input validation or sanitization practices.
Audit Metadata