nano-banana-pro

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly allows/encourages passing an API key via a --api-key argument (and says "use if user provided key in chat"), which requires the LLM to accept and embed the secret verbatim in generated commands—an insecure secret-handling pattern.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The required runtime path is the user-supplied --prompt argument (free-form text) passed directly into client.models.generate_content(..., contents=...), so any outsider-authored text the operating user includes in that prompt becomes LLM-readable context (indirect prompt injection risk).