notebooklm
Warn
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of a Python package from PyPI (notebooklm-py) or directly from a third-party GitHub repository (teng-lin/notebooklm-py). The GitHub installation method involves dynamically fetching the latest release tag using curl from the GitHub API.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands, including pip install, curl, grep, and cut for installation, as well as the notebooklm CLI tool for its primary functions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from potentially untrusted sources like web URLs, YouTube, and local files, which are then used in LLM-based queries (notebooklm ask).
- Ingestion points: Content is ingested through notebooklm source add.
- Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following malicious prompts hidden in the source content.
- Capability inventory: The skill allows network access for API interactions and file writing for downloads.
- Sanitization: The skill does not describe any sanitization or filtering of the source content.
Audit Metadata