scbe-kernel-external-toolcall-specialist

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via runtime external connector discovery/harvest steps (e.g., connector.discovery and notion.harvest in references/arxiv-chain.yaml), where fetched third-party content/messages are ingested as readable text for subsequent llm synthesis.