Skills
Navigate
skills/whytryharder/superpowers/nano-banana-2/Gen Agent Trust Hub

nano-banana-2

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the prompt and images parameters, which are then passed to the belt CLI for processing by an AI model. This creates a potential surface for indirect prompt injection if the input data contains malicious instructions intended to influence the agent's behavior.
  • Ingestion points: prompt and images fields defined in SKILL.md.
  • Boundary markers: Absent; inputs are interpolated into JSON objects for the CLI.
  • Capability inventory: Executes shell commands via the belt CLI (Bash tool).
  • Sanitization: Not specified in the skill instructions.
  • [EXTERNAL_DOWNLOADS]: The skill references external resources including documentation on inference.sh and installation guidelines hosted on raw.githubusercontent.com. It also directs the user to install the belt-sh/cli package via npx and references the inferencesh Python SDK.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 05:50 AM