marketing-images
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill mentions several standard CLI tools for image processing and network verification, including
cwebp,mogrify,jpegoptim,curl,grep,sips, andffprobe. These are used appropriately for their intended purposes (optimization, listing images, and verifying dimensions). - [DATA_EXPOSURE]: The skill explicitly instructs users to store API keys in environment variables and warns against pasting them into chat windows, demonstrating a strong security posture regarding credential management.
- [REMOTE_CODE_EXECUTION]: While the skill references external resources and repositories (e.g., GitHub links to other skills), these are treated as documentation sources rather than executable code. No automated remote script execution or piped downloads were found.
- [PROMPT_INJECTION]: The skill includes numerous prompts for image generation models, but these are focused on marketing aesthetics (lighting, composition, style) and do not attempt to bypass safety filters or override system instructions.
- [DYNAMIC_CONTEXT_INJECTION]: The skill does not utilize the dynamic
!command`` syntax for execution at load time. References to shell commands are provided as instructional examples for the user to run manually. - [EXTERNAL_DOWNLOADS]: The skill links to official documentation from trusted sources (Google Developers, OpenAI Developers, LinkedIn Help, and UK/US government copyright sites) for reference purposes. These are well-known, authoritative sources.
Audit Metadata