image-manipulation-image-magick
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for processing images by reading filenames from the local file system and interpolating them into shell commands.
- Ingestion points: Filenames and paths retrieved from the system (referenced in
SKILL.md). - Boundary markers: The skill guidelines section recommends quoting paths, but specific code snippets in Examples 1, 2, and 3 demonstrate command construction without quotes around path placeholders (e.g.,
path/to/image.jpg). - Capability inventory: The skill enables shell command execution through the
magickutility and local file system access. - Sanitization: No explicit sanitization or validation logic is provided to handle filenames that might contain shell metacharacters or escape sequences.
- [COMMAND_EXECUTION]: The skill's primary functionality relies on executing external shell commands using the ImageMagick
magickexecutable across multiple platforms (Windows, Linux, macOS). - The skill uses the
&operator in PowerShell and direct command calls in Bash to invoke themagicktool with arguments derived from potentially untrusted file system paths.
Audit Metadata