notebooklm
Warn
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to clone a codebase from a third-party GitHub repository (
PleasePrompto/notebooklm-skill). Therun.pywrapper additionally performs automated dependency installation and Chromium setup at runtime. - [COMMAND_EXECUTION]: The skill relies extensively on executing local Python scripts through a wrapper (
scripts/run.py). These scripts perform complex operations including browser automation, session management, and file system operations. - [DATA_EXFILTRATION]: While not necessarily exfiltrating data to an attacker, the skill specifically manages and stores sensitive Google session cookies and authentication state in
~/.claude/skills/notebooklm/data/. Access to authenticated browser sessions for a Google account represents a significant security surface for potential data exposure.
Audit Metadata