typespec-create-agent

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly requires adding a WebSearch capability "for web content" (and asks "What knowledge sources should it access?"), so the generated agent is expected to fetch and read open/public web content which could contain untrusted, user-generated instructions and influence agent actions.