gemini-image-simple
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Communicates with Google's official Gemini API (generativelanguage.googleapis.com) to generate images. This is a well-known service and the interaction is intrinsic to the skill's purpose.
- [DATA_EXFILTRATION]: Accesses local image files provided by the user via the --input argument for the purpose of image editing. Data is only sent to the trusted API endpoint.
- [PROMPT_INJECTION]: The skill processes user-provided prompts and image files as input for the Gemini model.
- Ingestion points: The prompt CLI argument and file contents from the --input path in scripts/generate.py.
- Boundary markers: None used.
- Capability inventory: Network access (urllib.request) and local file writing in scripts/generate.py.
- Sanitization: No sanitization or filtering of the prompt text is performed.
Audit Metadata