nano-banana

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides links to installation guides and documentation hosted on github.com/inference-sh and the inference.sh domain. These resources are for the setup of the legitimate belt CLI tool used by the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill involves ingesting user-provided prompts and external image URLs to be processed by an AI model. This ingestion point represents a surface where untrusted data enters the agent's context, although the functionality is restricted to image generation parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — nano-banana