picture
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Bash wrapper (
imagen.sh) to facilitate the execution of the Python generation script and ensure the correct environment is used. - [CREDENTIALS_UNSAFE]: The skill implements secure credential handling by looking for API keys in environment variables or the macOS Keychain, preventing the need for hardcoded secrets.
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection via user-supplied image descriptions.
- Ingestion points: The prompt parameter in generate.py accepts arbitrary user text.
- Boundary markers: No specific delimiters or instruction-ignoring warnings are used when passing the prompt to the API.
- Capability inventory: The skill can write files to the local ./images/ directory and perform network requests to the Google GenAI API.
- Sanitization: Input text is passed directly to the model; safety is managed by the model provider's safety filters.
Audit Metadata