Knowledge Graph Builder

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly lists "Unstructured: Documents, web content, text files" as data sources in Phase 3 and includes pipelines (EntityExtractionPipeline, RelationshipExtractor, extract_with_llm) that ingest and interpret that content, meaning the agent would consume untrusted/public third‑party web content as part of its workflow.