advance-directive-vs-polst

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

• Third-party content exposure detected (high risk: 0.70). The skill's "State Terminology" step explicitly directs the agent to "Verify via the National POLST program directory before asserting any state's form name," which requires fetching and interpreting a public third-party website that can materially change the skill's state-specific outputs.