multi-agent-patterns

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

• Third-party content exposure detected (high risk: 0.90). The skill includes explicit research agents that invoke tools like "search", "browse", and "read" (see references/frameworks.md — LangGraph Swarms create_agent and the AutoGen researcher descriptions), meaning the agent is expected to fetch and interpret open web/public third‑party content (untrusted/user‑generated) as part of its workflow.