| name | notebooklm-management |
| description | NotebookLM MCP server management — query notebooks, add from share links, handle auth, reset sessions. Use when working with Google NotebookLM notebooks for conversational research tasks. |
| license | MIT |
NotebookLM MCP Management
Advanced toolkit for managing NotebookLM MCP server integration with GitHub Copilot. This skill enables effective use of NotebookLM's conversational AI research capabilities through MCP tools.
When to Use This Skill
Activate this skill when:
- User mentions "NotebookLM", "notebook.lm", or "Google's AI research notebooks"
- Asked to query information from existing NotebookLM notebooks
- Requested to add a new notebook from a Google share link
- Need to manage notebook library (list, search, select notebooks)
- Troubleshooting NotebookLM authentication or session issues
- Working with knowledge bases stored in NotebookLM
- Performing conversational research or RAG-based queries
- Managing session context and chat history
- Updating notebook metadata (topics, descriptions, tags)
Capabilities
Core Operations
- Notebook Discovery: List library notebooks, search by topics/tags
- Query Interface: Ask questions about notebook content with contextual understanding
- Notebook Integration: Add notebooks via Google share URLs
- Session Management: Reset sessions, manage authentication state
- Metadata Management: Update notebook information (name, description, topics, use cases)
Authentication Handling
- Initial Google authentication setup via browser
- Session recovery and refresh
- Cookie management for persistent access
- Rate limit handling and troubleshooting
Prerequisites
Required Accounts
- Google account with NotebookLM access (free tier or subscription)
- Existing NotebookLM notebooks or share links
MCP Server Configuration
Ensure NotebookLM MCP server is properly configured in your MCP settings.
Browser Access
- Google Chrome or compatible browser for OAuth flow
Understanding of Limits
- Free tier: 100 notebooks, 50 sources each, 500k words, 50 daily queries
- Pro/Ultra: 5x higher limits
Quick Reference Workflows
For detailed step-by-step workflows, see workflows.md.
Quick Actions
| Action | Key Steps | Details |
|---|
| Add Notebook | Gather info → Propose → Confirm | Full flow in workflows.md |
| Query Notebook | Select → Formulate → Execute | Full flow in workflows.md |
| Manage Library | List → Search → Update | Full flow in workflows.md |
| Update Metadata | Identify → Verify → Apply | Full flow in workflows.md |
| Troubleshoot Auth | Diagnose → Repair → Verify | Full guide in troubleshooting.md |
| Reset Session | Identify → Confirm → Reset | Full guide in workflows.md |
Common Patterns
Add Notebook: See examples/add-notebook.py for usage patterns
Query Notebook: See examples/simple-query.py for basic queries
Multi-Turn: See examples/multi-turn-conversation.py for conversation flow
Library Management: See examples/library-management.py for management tasks
Ad-Hoc Queries: See examples/adhoc-query.py for temporary queries
MCP Tool Reference
For complete tool documentation, see MCP Tool Reference.
Quick Reference
| Tool | Purpose | Required Params | Key Options |
|---|
add_notebook | Add from share link | url, name, description, topics | content_types, use_cases, tags |
ask_question | Query notebooks | question | notebook_id, session_id, notebook_url |
list_notebooks | Show all notebooks | None | - |
search_notebooks | Search library | query | - |
update_notebook | Update metadata | id | name, description, topics, etc. |
reset_session | Clear chat history | session_id | - |
See mcp-tool-reference.md for detailed parameters, examples, and best practices.
Interactive Workflows
Common Scenarios
First-Time User: User mentions NotebookLM → Check Library → Decide needed action
Conversational Research: Select notebook → Query → Explore with follow-ups → Reset if needed
Library Management: Audit library → Organize topics → Update metadata → Remove outdated notes
For detailed implementation of these flows, see workflows.md
Best Practices Quick Guide
| Area | Best Practice | Details |
|---|
| Query Formulation | Be specific and contextual | Use follow-up questions, leverage sessions |
| Metadata Quality | Descriptive names, good topics | 3-5 topics, clear use cases |
| Authentication | Check before queries | Handle rate limits gracefully |
| Library Organization | Group by topic, use tags | Update regularly, remove outdated |
Troubleshooting Quick Reference
For comprehensive troubleshooting guide, see troubleshooting.md
| Issue | Quick Fix | Details |
|---|
| Auth Required | notebooklm.auth-setup | See Auth Issues |
| Query Timeout | Check network, simplify query | See Query Issues |
| Session Expired | reset_session or new session | See Session Issues |
| Rate Limit | Wait or upgrade account | See Rate Limit |
Common Code Patterns
For full examples with context, see examples/ folder.
Simple Query
response = ask_question(
question="How do I use useEffect?",
notebook_id="react-docs-123"
)
Details: simple-query.py
Add Notebook
add_notebook(
url="https://notebooklm.google/share/abc",
name="React Guide",
description="Complete React guide",
topics=["React", "Frontend", "JavaScript"]
)
Details: add-notebook.py
Additional Resources
Documentation
Code Examples
Utility Scripts
External References
Script Quick Start
python scripts/notebooklm-helper.py list
python scripts/notebooklm-helper.py search --query "react"
python scripts/notebooklm-helper.py report
python scripts/notebooklm-helper.py export --output backup.json
See scripts/README.md for complete usage guide.
Encoding
All files are UTF-8 encoded without BOM. Ensure all modifications maintain this encoding standard.
Related Skills
