notebooklm

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

• Third-party content exposure detected (high risk: 0.90). This skill programmatically opens and scrapes NotebookLM pages (e.g., https://notebooklm.google.com/...) as described in SKILL.md and implemented in scripts/ask_question.py and the Smart Add flow, ingesting user-uploaded/public notebook content (including linked websites/YouTube/GitHub) and using those responses to drive follow-up queries and synthesis, which clearly allows untrusted third‑party content to influence agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill launches a browser to navigate to NotebookLM at runtime (e.g., https://notebooklm.google.com and notebook URLs like https://notebooklm.google.com/notebook/...) and ingests the NotebookLM responses into the agent flow, so remote content from that URL directly controls prompts/answers the agent uses.