nano-banana

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool via Bash to interface with Google Gemini image generation models. The execution environment is restricted to the belt command only through the allowed-tools configuration in the YAML frontmatter, which follows the principle of least privilege.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the belt CLI from the vendor's official GitHub repository (inference-sh) and adding related functional skills via npx skills add. These references target established platform resources and are documented as part of the standard setup process.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a wrapper for image generation, ingesting user-provided prompts via JSON input in the --input flag. This is a standard functional requirement for image generation; the use of structured JSON for tool parameters reduces the risk of instruction confusion during processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 10:40 PM
Security Audit — agent-trust-hub — nano-banana