nano-banana-builder

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references standard development dependencies from official registries, including Google's AI SDK, Vercel Blob storage, and AWS S3 clients for storage operations.- [COMMAND_EXECUTION]: Includes standard documentation for environment setup, instructing users to run typical package manager commands (npm install) to install project dependencies.- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for processing user-supplied text and image data for generation.
  • Ingestion points: User prompts and base64-encoded image data in generateImage and editImage functions in advanced-patterns.md.
  • Boundary markers: Not explicitly implemented in the example code snippets, which is common for educational templates.
  • Capability inventory: Capabilities are limited to remote storage (Vercel Blob/S3) and AI model invocation; no arbitrary shell execution or local file system writes are performed.
  • Sanitization: Standard input validation is assumed to be handled by the implementation environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 11:27 PM
Security Audit — agent-trust-hub — nano-banana-builder