logos-router-reasoning

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

• Third-party content exposure detected (high risk: 0.70). The skill’s runtime workflow can ingest outsider-authored free text when executing workflows that analyze external documents (e.g., router.create_workflow("analyze_research_paper.pdf") / multi-document analysis), where the PDF’s extracted body text is read and fed into the LLM context.