multi-agent-thinker

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly describes agents performing "web search" and "document retrieval" and discusses "searching multiple independent sources" and a "Researcher (web search, document retrieval)" role, which means the agent is expected to fetch and read untrusted public web content (open sites, documents) as part of its workflow.