image
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references documentation for trusted and well-known services including Google Gemini, OpenAI, Black Forest Labs, and Vercel. These references are used for legitimate service integration and documentation.
- [COMMAND_EXECUTION]: Instructions contain benign examples of command-line utilities for image conversion and optimization, specifically cwebp, ImageMagick, and jpegoptim.
- [INDIRECT_PROMPT_INJECTION]: The skill uses local marketing context files to personalize output. 1. Ingestion points: .agents/product-marketing.md (SKILL.md). 2. Boundary markers: None. 3. Capability inventory: Examples of command line processing (SKILL.md). 4. Sanitization: None. This is a standard pattern for context-aware agents and represents an acceptable surface.
Audit Metadata