Skills
skills/elizaos/eliza/nano-banana-pro/Gen Agent Trust Hub

nano-banana-pro

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill ingests untrusted data from the user or external sources which could contain malicious instructions designed to influence the agent or the image generation model.
  • Ingestion points: The --prompt and --input-image arguments in scripts/generate_image.py accept external content.
  • Boundary markers: Absent; the prompt text is interpolated directly into the API request without delimiters or 'ignore' instructions.
  • Capability inventory: The script performs network operations (Google API), reads local image files (PILImage.open), and writes files to the local file system (image.save).
  • Sanitization: No sanitization or validation of input prompts or image metadata is performed.
  • [External Downloads] (SAFE): The skill depends on reputable packages from trusted sources.
  • Evidence: Uses google-genai (from a Trusted GitHub Organization) and pillow for image processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 06:11 PM