nano-banana

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external documentation and installation guidelines hosted on GitHub (e.g., https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md). These references are used to provide the user with setup instructions for the required 'belt' CLI tool.
  • [COMMAND_EXECUTION]: The skill uses the belt command-line utility to interact with the inference.sh API for image generation. The allowed-tools configuration appropriately limits execution to the belt command.
  • [INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface where user-provided text prompts are interpolated into shell commands via JSON arguments. While this represents a potential surface for indirect prompt injection, it is the primary intended functionality of the image generation skill and does not pose a direct threat to the execution environment. The agent is responsible for correctly formatting the JSON input.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 01:13 PM
Security Audit — agent-trust-hub — nano-banana