NotebookLM CLI & MCP Server — Enterprise + Personal

NotebookLM MCP Header

This is an enterprise-focused fork of [jacob-bd/notebooklm-mcp-cli](https://github.com/jacob-bd/notebooklm-mcp-cli). It adds full support for NotebookLM Enterprise (notebooklm.cloud.google.com) via the official Discovery Engine REST API, while keeping all personal-mode features intact. If you only use personal NotebookLM, the upstream repo is the right choice. If you have a Google Workspace enterprise account, you're in the right place.

What this fork adds over upstream

All personal-mode features (chat, video, reports, flashcards, research, sharing, etc.) are fully preserved.

Programmatic access to Google NotebookLM — via command-line interface (CLI) or Model Context Protocol (MCP) server. Supports both Personal and Enterprise accounts.

📺 Watch the Demos

Note: The demos below are from the upstream project (jacob-bd/notebooklm-mcp-cli) and show personal mode features. All commands and MCP tools work identically in this fork — enterprise mode adds on top of everything shown here.

Latest

Codex Setup + Cinematic Video & Slides

MCP Demos

General Overview	Claude Desktop	Perplexity Desktop	MCP Super Assistant

CLI Demos

CLI Overview	CLI, MCP & Skills	Setup, Doctor & mcpb	Infographics Support

Two Ways to Use

🖥️ Command-Line Interface (CLI)

Use nlm directly in your terminal for scripting, automation, or interactive use:

nlm notebook list                              # List all notebooks
nlm notebook create "Research Project"         # Create a notebook
nlm source add <notebook> --url "https://..."  # Add sources
nlm audio create <notebook> --confirm          # Generate podcast
nlm download audio <notebook> <artifact-id>    # Download audio file
nlm share public <notebook>                    # Enable public link

Run nlm --ai for comprehensive AI-assistant documentation.

🤖 MCP Server (for AI Agents)

Connect AI assistants (Claude, Gemini, Cursor, etc.) to NotebookLM:

# Automatic setup — picks the right config for each tool
nlm setup add claude-code
nlm setup add gemini
nlm setup add cursor
nlm setup add cline
nlm setup add antigravity

# Generate JSON config for any other tool
nlm setup add json

Then use natural language: "Create a notebook about quantum computing and generate a podcast"

Features

📚 More Documentation:

• [CLI Guide](docs/CLI_GUIDE.md) — Complete command reference
• [MCP Guide](docs/MCP_GUIDE.md) — All 35 MCP tools with examples
• [Authentication](docs/AUTHENTICATION.md) — Setup and troubleshooting
• [API Reference](docs/API_REFERENCE.md) — Internal API docs for contributors

Important Disclaimer

This MCP and CLI use internal APIs that:

• Are undocumented and may change without notice
• Require cookie extraction from your browser (I have a tool for that!)

Use at your own risk for personal/experimental purposes.

Installation

🆕 Claude Desktop users: Download the extension (.mcpb file) → double-click → done! One-click install, no config needed.

Install from PyPI. This single package includes both the CLI and MCP server:

Using uv (Recommended)

uv tool install notebooklm-enterprise-mcp

Using uvx (Run Without Install)

uvx --from notebooklm-enterprise-mcp nlm --help
uvx --from notebooklm-enterprise-mcp notebooklm-mcp

Using pip

pip install notebooklm-enterprise-mcp

Using pipx

pipx install notebooklm-enterprise-mcp

After installation, you get:

• nlm — Command-line interface
• notebooklm-mcp — MCP server for AI assistants

<details> <summary>Alternative: Install from Source</summary>

# Clone the repository
git clone https://github.com/Robiton/notebooklm-mcp-cli.git
cd notebooklm-mcp

# Install with uv
uv tool install .

</details>

Upgrading

# Using uv
uv tool upgrade notebooklm-enterprise-mcp

# Using pip
pip install --upgrade notebooklm-enterprise-mcp

# Using pipx
pipx upgrade notebooklm-enterprise-mcp

After upgrading, restart your AI tool to reconnect to the updated MCP server:

• Claude Code: Restart the application, or use /mcp to reconnect
• Cursor: Restart the application
• Gemini CLI: Restart the CLI session

Switching from Upstream (jacob-bd/notebooklm-mcp-cli)

If you were using the upstream package and want to switch to this fork for enterprise support:

Step 1: Uninstall the upstream package

uv tool uninstall notebooklm-mcp-cli
# or: pip uninstall notebooklm-mcp-cli

Step 2: Install this fork's package

uv tool install notebooklm-enterprise-mcp

Step 3: Verify

uv tool list | grep notebooklm
# Should show: notebooklm-enterprise-mcp v1.0.0
#                - nlm
#                - notebooklm-mcp

Step 4: Re-authenticate (if needed)

Your existing cookies from the upstream package should still work. If you hit auth errors:

nlm login

MCP configuration doesn't change — the executable is still called notebooklm-mcp, so your Claude Code, Cursor, or other tool configs need no edits.

Uninstalling

To completely remove the MCP:

# Using uv
uv tool uninstall notebooklm-enterprise-mcp

# Using pip
pip uninstall notebooklm-enterprise-mcp

# Using pipx
pipx uninstall notebooklm-enterprise-mcp

# Remove cached auth tokens and data (optional)
rm -rf ~/.notebooklm-mcp-cli

Also remove from your AI tools:

nlm setup remove claude-code
nlm setup remove cursor
# ... or any configured tool

Why This Fork?

The upstream project targets personal NotebookLM accounts only. Enterprise NotebookLM (notebooklm.cloud.google.com) uses a completely different authentication system (GCP OAuth2) and a separate official REST API — it's not just a different URL.

This fork adds:

• Enterprise REST API client — official Discovery Engine API, not reverse-engineered batchexecute
• Persistent config — nlm config set enterprise.mode enterprise persists across restarts (no env var editing)
• `configure_mode` MCP tool — switch modes from within Claude with auth pre-checks
• Paywall detection — URL sources are checked for login/subscription walls before adding
• Per-URL bulk results — one bad URL in a batch doesn't fail the whole batch
• Standalone Podcast API — generate podcasts from raw text, no notebook needed

The enterprise REST API (v1alpha) covers notebooks, sources, and audio. Chat, video, reports, and other features remain personal-only — they have no documented REST endpoints. The hope is that Google promotes the API to v1 stable and expands coverage over time.

See docs/AUTHENTICATION.md for full enterprise setup instructions.

Enterprise Mode

If you use NotebookLM Enterprise (notebooklm.cloud.google.com), configure enterprise mode:

Via CLI

nlm config set enterprise.mode enterprise
nlm config set enterprise.project_id YOUR_PROJECT_NUMBER
nlm config set enterprise.location global    # or "us" or "eu"

Via MCP (Claude Desktop)

Ask Claude to call:

configure_mode(mode="enterprise", project_id="YOUR_PROJECT_NUMBER", location="global")

Enterprise Authentication

Enterprise uses GCP OAuth2 instead of browser cookies:

gcloud auth login

Switch Back to Personal

nlm config set enterprise.mode personal

Enterprise Feature Support

Environment variables (NOTEBOOKLM_MODE, NOTEBOOKLM_PROJECT_ID, NOTEBOOKLM_LOCATION) override config.toml when set.

Paywall / Login-Required URLs

When adding URL sources, the MCP checks if the URL is behind a paywall or login wall. If your organisation hosts content on an internal domain or a subscription site you have access to, add it to your approved list to skip the check:

# Approve a single domain
nlm config set sources.approved_domains '["yourcompany.com"]'

# Approve multiple domains
nlm config set sources.approved_domains '["yourcompany.com", "ft.com", "wsj.com"]'

Or in the MCP, call source_add with skip_paywall_check=True to bypass for a single URL.

Authentication

Before using the CLI or MCP, you need to authenticate with NotebookLM:

CLI Authentication (Recommended)

# Auto mode: launches your browser, you log in, cookies extracted automatically
nlm login

# Check if already authenticated
nlm login --check

# Use a named profile (for multiple Google accounts)
nlm login --profile work
nlm login --profile personal

# Manual mode: import cookies from a file
nlm login --manual --file cookies.txt

# External CDP provider (e.g., OpenClaw-managed browser)
nlm login --provider openclaw --cdp-url http://127.0.0.1:18800

Profile management:

nlm login --check                    # Show current auth status
nlm login switch <profile>           # Switch the default profile
nlm login profile list               # List all profiles with email addresses
nlm login profile delete <profile>   # Delete a profile
nlm login profile rename <old> <new> # Rename a profile

Each profile gets its own isolated browser session, so you can be logged into multiple Google accounts simultaneously.

Standalone Auth Tool

If you only need the MCP server (not the CLI):

nlm login              # Auto mode (launches browser)
nlm login --manual     # Manual file mode

How it works: Auto mode launches a dedicated browser profile (supports Chrome, Arc, Brave, Edge, Chromium, and more), you log in to Google, and cookies are extracted automatically. Your login persists for future auth refreshes.

Prefer a specific browser? Set it with nlm config set auth.browser chromium (or brave, arc, edge, chrome, etc.). Falls back to auto-detection if the preferred browser is not found.

For detailed instructions and troubleshooting, see [docs/AUTHENTICATION.md](docs/AUTHENTICATION.md).

MCP Configuration

⚠️ Context Window Warning: This MCP provides 35 tools. Disable it when not using NotebookLM to preserve context. In Claude Code: @notebooklm-mcp to toggle.

Automatic Setup (Recommended)

Use nlm setup to automatically configure the MCP server for your AI tools — no manual JSON editing required:

# Add to any supported tool
nlm setup add claude-code
nlm setup add claude-desktop
nlm setup add gemini
nlm setup add cursor
nlm setup add windsurf

# Generate JSON config for any other tool
nlm setup add json

# Check which tools are configured
nlm setup list

# Diagnose installation & auth issues
nlm doctor

Install AI Skills (Optional)

Install the NotebookLM expert guide for your AI assistant to help it use the tools effectively. Supported for Cline, Antigravity, OpenClaw, Codex, OpenCode, Claude Code, and Gemini CLI.

# Install skill files
nlm skill install cline
nlm skill install openclaw
nlm skill install codex
nlm skill install antigravity

# Update skills
nlm skill update

Remove from a tool

nlm setup remove claude-code

Using uvx (No Install Required)

If you don't want to install the package, you can use uvx to run on-the-fly:

# Run CLI commands directly
uvx --from notebooklm-enterprise-mcp nlm setup add cursor
uvx --from notebooklm-enterprise-mcp nlm login

For tools that use JSON config, point them to uvx:

{
  "mcpServers": {
    "notebooklm-mcp": {
      "command": "uvx",
      "args": ["--from", "notebooklm-enterprise-mcp", "notebooklm-mcp"]
    }
  }
}

<details> <summary>Manual Setup (if you prefer)</summary>

Tip: Run nlm setup add json for an interactive wizard that generates the right JSON snippet for your tool.

Claude Code / Gemini CLI support adding MCP servers via their own CLI:

claude mcp add --scope user notebooklm-mcp notebooklm-mcp
gemini mcp add --scope user notebooklm-mcp notebooklm-mcp

Cursor / Windsurf resolve commands from your PATH, so the command name is enough:

{
  "mcpServers": {
    "notebooklm-mcp": {
      "command": "notebooklm-mcp"
    }
  }
}

Claude Desktop / VS Code may not resolve PATH — use the full path to the binary:

{
  "mcpServers": {
    "notebooklm-mcp": {
      "command": "/full/path/to/notebooklm-mcp"
    }
  }
}

Find your path with: which notebooklm-mcp

</details>

📚 Full configuration details: MCP Guide — Server options, environment variables, HTTP transport, multi-user setup, and context window management.

What You Can Do

Simply chat with your AI tool (Claude Code, Cursor, Gemini CLI) using natural language. Here are some examples:

Research & Discovery

• "List all my NotebookLM notebooks"
• "Create a new notebook called 'AI Strategy Research'"
• "Start web research on 'enterprise AI ROI metrics' and show me what sources it finds"
• "Do a deep research on 'cloud marketplace trends' and import the top 10 sources"
• "Search my Google Drive for documents about 'product roadmap' and create a notebook"

Adding Content

• "Add this URL to my notebook: https://example.com/article"
• "Add this YouTube video about Kubernetes to the notebook"
• "Add my meeting notes as a text source to this notebook"
• "Import this Google Doc into my research notebook"

AI-Powered Analysis

• "What are the key findings in this notebook?"
• "Summarize the main arguments across all these sources"
• "What does this source say about security best practices?"
• "Get an AI summary of what this notebook is about"
• "Configure the chat to use a learning guide style with longer responses"

(All queries sent from CLI or MCP automatically persist in your NotebookLM web UI chat history!)

Content Generation

• "Create an audio podcast overview of this notebook in deep dive format"
• "Generate a video explainer with classic visual style"
• "Make a briefing doc from these sources"
• "Create flashcards for studying, medium difficulty"
• "Generate an infographic in landscape orientation with professional style"
• "Build a mind map from my research sources"
• "Create a slide deck presentation from this notebook"

Smart Management

• "Check which Google Drive sources are out of date and sync them"
• "Show me all the sources in this notebook with their freshness status"
• "Delete this source from the notebook"
• "Check the status of my audio overview generation"

Sharing & Collaboration

• "Show me the sharing settings for this notebook"
• "Make this notebook public so anyone with the link can view it"
• "Disable public access to this notebook"
• "Invite [email protected] as an editor to this notebook"
• "Add a viewer to my research notebook"

Pro tip: After creating studio content (audio, video, reports, etc.), poll the status to get download URLs when generation completes.

Authentication Lifecycle

v1.0.0+: The server now automatically handles token expiration:

1. Refreshes CSRF tokens immediately when expired
2. Reloads cookies from disk if updated externally
3. Runs headless browser auth if profile has saved login

You can also call refresh_auth() to explicitly reload tokens.

If automatic refresh fails (Google login fully expired), run nlm login again.

Troubleshooting

uv tool upgrade Not Installing Latest Version

Symptoms:

• Running uv tool upgrade notebooklm-enterprise-mcp installs an older version than expected
• uv cache clean doesn't fix the issue

Why this happens: uv tool upgrade respects version constraints from your original installation. If you initially installed an older version or with a constraint, upgrade stays within those bounds by design.

Fix — Force reinstall:

uv tool install --force notebooklm-enterprise-mcp

This bypasses any cached constraints and installs the absolute latest version from PyPI.

Verify:

uv tool list | grep notebooklm
# Should show: notebooklm-enterprise-mcp v1.0.0 (or latest)

Limitations

• Rate limits: Free tier has ~50 queries/day
• No official support: API may change without notice
• Cookie expiration: Need to re-extract cookies every few weeks

Reporting Issues

Use the [GitHub Issues](https://github.com/Robiton/notebooklm-mcp-cli/issues/new/choose) tab — you'll be prompted to pick the right template:

Personal mode bugs that also exist in the upstream project should be filed at jacob-bd/notebooklm-mcp-cli first — fixes accepted there get cherry-picked here automatically.

Contributing

See CONTRIBUTING.md for setup instructions, PR process, and how to add new features.

A Note on This Fork

This fork started as an internal tool to fill a real gap: NotebookLM Enterprise had no programmatic access, and the upstream project (which is excellent) is scoped to personal accounts only. The enterprise implementation was built with AI assistance — I'm not a professional Python developer — so if you see something that could be done better, PRs are genuinely welcome.

The upstream project's original "vibe coding" spirit applies here too: the goal was to solve a problem, not to write a textbook. The code works, passes tests, and is production-stable enough for daily use. But experienced developers will likely see room for improvement, and that input is valuable.

See CONTRIBUTING.md for how to get involved — especially if you have enterprise NotebookLM access and can test end-to-end.

Credits

Special thanks to:

• Le Anh Tuan (@latuannetnam) for contributing the HTTP transport, debug logging system, and performance optimizations.
• David Szabo-Pele (@davidszp) for the source_get_content tool and Linux auth fixes.
• saitrogen (@saitrogen) for the research polling query fallback fix.
• devnull03 (@devnull03) for multi-browser CDP authentication support (Arc, Brave, Edge, Chromium, Vivaldi, Opera).
• VooDisss (@VooDisss) for multi-browser authentication improvements.
• codepiano (@codepiano) for the configurable DevTools timeout for the auth CLI.
• Tony Hansmann (@997unix) for contributing the nlm setup and nlm doctor commands and CLI Guide documentation.
• Fabiana Furtado (@fabianafurtadoff) for batch operations, cross-notebook query, pipelines, and smart select/tagging (PR #90).

License

MIT License