image-prompting
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of documentation and prompt engineering guidelines with no executable scripts or external code references.
- [PROMPT_INJECTION]: No instructional overrides, jailbreak attempts, or safety bypass patterns were detected. The skill promotes structured, constructive interaction with image generation models.
- [DATA_EXFILTRATION]: No network operations or commands for exfiltrating data are present. The file path references (~/nanobanana-images/) are localized and used for defining operational scope within the user's home directory.
- [COMMAND_EXECUTION]: The skill does not contain shell commands, subprocess calls, or dynamic execution patterns. All mentioned tools (edit_image, continue_editing) are conceptual references to external agent capabilities.
Audit Metadata