notebooklm-import

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests NotebookLM outputs derived from public third‑party sources (e.g., workflows/import.md uses "notebooklm source list --json" and import_sources.py calls "notebooklm source guide {id}", workflows/ask.md and scripts/extract_passages.py read Q&A JSONs containing third‑party cited_text and resolve_citations.py processes answer text/references), so untrusted user-generated or web content is read and directly influences how files are created and links/resolution logic is applied.