readme-expert

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly performs network fetches of external resources (e.g., "WebFetch: url="<external_url>"" in Layer 4 link integrity checks and registry queries like "npm view my-package" / PyPI checks in application/script-executor.md), which are open/public third‑party sources whose untrusted content the agent is expected to read and interpret as part of validation.