The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a collection of Bash scripts (notebook_ops.sh, source_ops.sh, artifact_ops.sh, etc.) that act as wrappers for the notebooklm CLI tool. These scripts allow the agent to manage NotebookLM notebooks, sources, and artifacts. All command arguments are handled using Bash arrays (e.g., "${ARGS[@]}"), which effectively prevents shell injection within the wrapper logic by ensuring variables are not re-interpreted as shell commands.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the download of generated artifacts (such as audio, video, and documents) from the NotebookLM platform to the local filesystem using download_artifact.sh. It also provides instructions for installing the notebooklm-py package from PyPI and the Playwright browser dependencies.\n- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection (Category 8) because it processes untrusted data from external URLs, files, and research queries through the source_ops.sh and generate_and_download.sh scripts.\n
Ingestion points: Ingests external content via source_ops.sh add (--content) and source_ops.sh research (--query).\n
Boundary markers: No delimiters or explicit safety warnings are used when passing external content to the CLI tool to mitigate obedience to embedded instructions.\n
Capability inventory: The skill can write to the local filesystem (via download) and execute the notebooklm tool.\n
Sanitization: No input validation or sanitization is performed on external content before it is passed to the underlying CLI for processing.

notebooklm-exporter