fal-nano-banana-2-image-gen

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

• Third-party content exposure detected (high risk: 0.70). The skill's scripts (scripts/generate.mjs) explicitly call external fal.run / queue.fal.run endpoints, follow status_url/response_url values from those responses, and download image URLs returned by the service (see fetchTextWithTimeout, runFalQueue, and downloadToFile), so it ingests and acts on untrusted third‑party API responses/URLs provided at runtime.