notebooklm-hardened

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly supports adding arbitrary public URLs as notebook sources (see "source add https://example.com" in references/cli-commands.md and examples in SKILL.md), which the CLI will ingest into notebooks and be read/queried (e.g., via ask commands), so untrusted third‑party web content can directly influence agent outputs and subsequent actions.