use-subagents

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

• Third-party content exposure detected (high risk: 0.78). The workflow delegates to subagents (e.g., the “research” role explicitly says to “fetch and verify primary sources” from external sites), which can ingest outsider-authored free-form web page text into the agent’s LLM context at runtime via runtime retrieval/web-fetch tools.