web-ready
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates image processing by invoking several well-known command-line utilities.
- Evidence: Procedures in
SKILL.mdcallexiftool,vipsthumbnail,avifenc,cwebp,cjpeg, andoxipng. - Mitigation: Command templates utilize double-quoting for variables (e.g.,
"<staged>","<input>"), which helps prevent command injection if filenames contain shell metacharacters. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface by processing untrusted data from a user-provided directory.
- Ingestion points: User-provided image files and directory paths in
SKILL.md. - Boundary markers: None explicitly defined for binary data; shell quoting is used for filename boundaries.
- Capability inventory: The skill is capable of executing shell commands and performing file system operations (read/write/delete) within the specified input and output directories.
- Sanitization: Uses double quotes around variable expansions in shell scripts to mitigate filename-based injection attacks.
Audit Metadata