story-cover
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses user-supplied book titles and author names to build image generation prompts.
- Ingestion points: Novel title and author name inputs in
SKILL.md. - Boundary markers: No specific delimiters are used to separate user data from instructions in the prompt.
- Capability inventory: Performs network requests (
curl) and image manipulation (magick,sips). - Sanitization: Employs
jqto safely escape strings for JSON payloads. - [EXTERNAL_DOWNLOADS]: Downloads reference images from external URLs provided by the user.
- Evidence: Uses
curl -fsSLin the image-to-image workflow. - [COMMAND_EXECUTION]: Runs system commands for API communication and image processing.
- Evidence: Invokes
curl,jq,base64,magick, andsipsin the implementation.
Audit Metadata