muapi-photo-pack-generator
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/generate-pack.shinvokes shell scripts located at../../../../core/media/upload.shand../../../../core/media/generate-image.shto handle media processing and image generation within the host environment. - [EXTERNAL_DOWNLOADS]: When the
--viewflag is enabled,scripts/generate-pack.shusescurlto download generated images from the service's content delivery network to a localmedia_outputsdirectory. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection by interpolating user-controlled data into instructions for the image generation model.
- Ingestion points: The
--likenesscommand-line argument and the content of reference images (provided via--imageor--image-url) inscripts/generate-pack.sh. - Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when constructing the final prompt.
- Capability inventory: The skill possesses the ability to execute shell scripts and perform network operations via
curl. - Sanitization: The user-supplied likeness string is used directly in the
SUBJECTvariable without escaping or validation before being included in the model prompt.
Audit Metadata