nano-banana

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external installation documentation from the inference-sh GitHub organization and suggests the installation of the belt-sh/cli skill.
  • [COMMAND_EXECUTION]: The skill uses the belt CLI to perform image generation tasks. The Bash tool access is restricted to the belt binary, following the principle of least privilege.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied prompts which are interpolated into CLI command arguments.
  • Ingestion points: User-provided prompt text in the SKILL.md examples.
  • Boundary markers: Prompts are contained within a JSON string assigned to the --input argument.
  • Capability inventory: Execution of the belt command via the Bash tool.
  • Sanitization: No specific sanitization or escaping logic is described for the prompt inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 05:50 AM