multi-agent-patterns

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md and the implementation reference (references/frameworks.md) explicitly define a "researcher" agent and create_agent calls that use tools like search, browse, and read (e.g., SKILL.md "Researcher (web search, document retrieval)" and references/frameworks.md researcher_node/create_agent with perform_research and tools [search, browse, read]), meaning the workflow fetches and interprets untrusted public web content whose results drive coordinator decisions and downstream actions.