paw-pa-research

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow can ingest outsider-authored free text from public web content: in live-browser-research.md / cursor-browser-research.md, the LLM navigates to and extracts text from URLs it discovers/browses, then appends that extracted “examples”/benchmarks/intel into the findings JSON that is later rendered using render_dossier.py (LLM-to-report context includes that web-sourced prose).