nano-banana-2
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires installing the
belt-sh/cliskill and references an external installation guide athttps://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md. - [COMMAND_EXECUTION]: The skill uses the
beltcommand via the Bash tool to perform login and run image generation apps. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data (prompts and image URLs) to interact with the inference service. 1. Ingestion points: The
promptandimagesparameters in thebelt app runcommand. 2. Boundary markers: The input is structured as a JSON string within the command line arguments. 3. Capability inventory: The skill is limited to running thebeltcommand using the Bash tool. 4. Sanitization: No explicit input sanitization is implemented in the documented workflows.
Audit Metadata