ecommerce-images
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to invoke the baoyu-image-gen tool as its primary functional component for image generation. This represents the intended operation of the workflow within its defined scope.
- [DATA_EXPOSURE]: The skill mentions the use of GOOGLE_API_KEY and OPENAI_API_KEY environment variables for model access. This is consistent with standard configuration practices for AI agents and does not involve the exposure of hardcoded secrets.
- [PROMPT_INJECTION]: The skill evaluates user-supplied style names and parameters to construct image generation prompts. It implements a fallback mechanism to default styles when receiving undefined inputs, which acts as a basic validation layer to ensure consistent behavior.
- Ingestion points: User-provided style names (mainStyle, detailStyle) and image parameters in SKILL.md.
- Boundary markers: None specified for the prompt construction.
- Capability inventory: Call to baoyu-image-gen tool.
- Sanitization: Fallback logic to revert to default styles if undefined inputs are provided.
Audit Metadata