nano-banana-openrouter

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

• Third-party content exposure detected (high risk: 0.80). The skill sends user prompts to the third‑party OpenRouter API (base_url "https://openrouter.ai/api/v1") and directly reads and prints the returned model message (response.choices[0].message.content) as part of its workflow, thereby ingesting untrusted third‑party content that could carry implicit instructions.