notebooklmlisted
Install: claude install-skill asynchronos/operator47-plugins
# NotebookLM report + infographic generator
Drive Google NotebookLM from the command line to turn local content into an
AI **report** (Markdown) and **infographic** (PNG), and to run **grounded Q&A**
over those sources. All logic lives in two portable scripts under
`${CLAUDE_PLUGIN_ROOT}/skills/notebooklm/scripts/` — nothing is tied to any
particular project; sources are just text/Markdown.
> ⚠️ **Unofficial + ToS risk.** This uses `notebooklm-py`, which drives
> NotebookLM's private internal API. Use a **throwaway Google account**, keep
> usage light, and only send non-sensitive content. Confirm with the user
> before the first live run. See [references/legal.md](references/legal.md).
## When to use / not use
- **Use** when the user wants a NotebookLM report/infographic from their docs or
data, wants to load files into a notebook, or wants grounded answers from sources.
- **Don't use** for ordinary summarization you can do directly, or when the user
has no NotebookLM/Google account to spare — offer a local alternative instead.
## Prerequisites (one-time)
Check setup first; if it fails, walk the user through
[references/setup.md](references/setup.md).
```bash
pip install -r "${CLAUDE_PLUGIN_ROOT}/skills/notebooklm/scripts/requirements.txt"
notebooklm login # opens a browser; sign in with a THROWAWAY account
notebooklm auth check --test --json # expect {"status": "ok", ...}
```
On Windows use `py -m pip ...` and `py` instead of `python`. `no