viceroy

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md and references/fastly-compute-config.md) instructs configuring [local_server.backends] with arbitrary external URLs (e.g., https://api.example.com) and references/fastly-compute-test.md shows the runtime/tests send requests and interpret backend responses, so untrusted third‑party content can be fetched and materially influence behavior.