notebooklm-exporter

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

• Third-party content exposure detected (high risk: 0.90). This skill clearly ingests untrusted web content: SKILL.md and scripts/source_ops.sh show adding URL/text/file sources and a "research" action with default --from web (source add / source add-research delegating to the notebooklm CLI), and those sources are then used by generation scripts (generate_and_download.sh) so external web content can influence generated artifacts and subsequent actions.