google-zx-scripting

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

• Third-party content exposure detected (high risk: 0.80). The skill's reference workflows (e.g., "Download Files in Parallel" and "AI Script Patterns" in references/processing-recipes.md) explicitly fetch and curl arbitrary URLs (from urls.txt) and send fetched content to LLM APIs, meaning untrusted third‑party web content is ingested and can directly influence subsequent actions.