# Query NotebookLM From Your AI Assistant

Research across uploaded documents can require repeated browser switching. This skill lets Claude, Codex, and Claude Code query NotebookLM through local browser automation.

## Install

```bash
npx skillstore add sickn33/notebooklm
```

## Metadata

- Status: approved
- Slug: sickn33-notebooklm
- Skillstore revision: r2
- Version status: missing
- Tree hash: 509b5350c54f3b5e44dcfe4f97be5e118e2db38df0dc48f3434fce09c2f067aa
- Author: sickn33
- GitHub username: sickn33
- License: MIT
- Repository: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/notebooklm
- Ref: 81e05e636292629114b76cbb3922fbe57672fc02
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: confirmation\_required
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: filesystem, network, env\_access, external\_commands
- Quality score: 38
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/sickn33-notebooklm
- Manifest: https://skillstore.pages.dev/api/skills/sickn33-notebooklm/manifest

## Capabilities

- Opens NotebookLM in a local Chrome session and submits natural-language questions.
- Supports manual Google authentication and restores saved browser session state.
- Stores notebook links with names, descriptions, topics, tags, and active selection.
- Lists, searches, activates, removes, and summarizes locally saved notebook metadata.
- Extracts the latest NotebookLM answer and returns it to the calling assistant.
- Creates a local Python environment and installs pinned runtime dependencies.

## Use Cases

- Check Technical Documentation: Ask implementation questions against uploaded product manuals, API guides, or internal engineering documents.
- Synthesize Research Sources: Compare findings across papers, reports, and notes already organized in a NotebookLM notebook.
- Retrieve Operational Guidance: Find policies, procedures, and exact specifications from a maintained organizational knowledge notebook.

## Prompt Templates

### List My Notebooks

```
List my saved NotebookLM notebooks and show which notebook is active.
```

### Ask One Source Question

```
Ask my [notebook name] notebook: [specific question]. Return the answer with its source context.
```

### Add and Describe a Notebook

```
Query this NotebookLM link, summarize its scope, and add it with a clear name, description, and topic tags: [NotebookLM URL].
```

### Compare Evidence Across Topics

```
Use my [notebook name] notebook to compare [topic A] and [topic B]. Identify agreements, conflicts, missing evidence, and practical implications.
```

## Limitations

- Requires local network access, Chrome, and an interactive Google login.
- Users must create notebooks and upload source documents in NotebookLM.
- Each question opens a fresh browser flow, so chat context is not persistent.
- Google rate limits and NotebookLM interface changes can interrupt automation.

## Best Practices

- Use a dedicated Google account and protect the local browser-state directory.
- Ask narrow questions that name the notebook, topic, and required level of detail.
- Verify important claims against NotebookLM citations before acting on them.

## Anti Patterns

- Do not provide notebook links from untrusted sources or unrelated domains.
- Do not share, commit, or back up saved browser cookies to insecure locations.
- Do not assume NotebookLM answers are complete when source documents lack evidence.

## Security Audit

- Audited at: 2026-08-04T16:14:49.377\+00:00
- Summary: The skill performs legitimate NotebookLM research automation, and most static matches are documentation or fixed commands. Confirmed risks include behavior-steering prompt injection, anti-detection automation, persistent Google session cookies, unsafe URL navigation with a disabled browser sandbox, automatic installers, and an arbitrary local Python execution path.

## Stats

- Views: 228
- Downloads: 17
- Favorites: 0
- Popularity score: 0