nano-banana-2
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates image generation and editing tasks using the Gemini 3.1 Flash Image Preview model through documented CLI and SDK methods. No malicious code patterns or bypass attempts were detected.
- [EXTERNAL_DOWNLOADS]: The skill references documentation and installation scripts hosted on the vendor's official GitHub organization and primary domain (inference.sh). These references are consistent with the skill's stated purpose of providing access to an external AI service.
- [COMMAND_EXECUTION]: The skill provides examples of using the
infshCLI tool. Usage is restricted in the frontmatter to theinfshcommand, following the principle of least privilege.
Audit Metadata