gemini-image
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied prompts that are passed to generative AI models without sanitization.
- Ingestion points: The image generation prompt is captured from command-line arguments in
scripts/generate_image.jsviaparseCli(). - Boundary markers: Input prompts are interpolated directly into API calls without using delimiters or safety instructions to separate user input from system intent.
- Capability inventory: The skill has the ability to create directories (
mkdir) and write binary files (writeFile) to the local file system. - Sanitization: No validation or filtering is performed on the user-provided prompt before it is sent to the Google GenAI API.
Audit Metadata