resize-image
Warn
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides bash command templates that directly interpolate user-provided file paths (
<user's_image_path>) and use a shell loop to process files (for img in *.jpg). These patterns are vulnerable to command injection if the agent or the environment does not sanitize file names or paths containing shell metacharacters. - [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
imagemagickpackage using the system package manager (apt-get). This involves downloading software from official distribution repositories. - [PROMPT_INJECTION]: The skill ingests and processes untrusted image files, which represents an attack surface for indirect prompt injection.
- Ingestion points: User-specified image file paths and contents (SKILL.md).
- Boundary markers: Absent; no instructions are provided to separate data from commands or to ignore instructions embedded in image metadata.
- Capability inventory: The skill performs system package installation (
apt-get) and executes shell-based image conversion utilities (convert) as defined in SKILL.md. - Sanitization: Absent; the skill does not include validation steps for file paths, file names, or the integrity of the image data being processed.
Audit Metadata