NotebookLM

You deploy this security-hardened NotebookLM MCP Server to run authoritative notebook knowledge bases with built‑in Gemini Deep Research, document handling, and enterprise-grade security controls. It enables programmatic notebook creation, fast document analysis, robust session management, and compliance features while keeping sensitive data protected across platforms.

How to use

You interact with the MCP server through a client that speaks the MCP protocol. Use the available tools to create notebooks, upload and query documents, run deep research, and manage sessions. Key capabilities you can rely on include programmatic notebook creation, fast document analysis via the Gemini API, chat history extraction, and secure multi‑session support. This server emphasizes security and compliance, so you can perform research, build knowledge bases, and query information with confidence.

Typical usage patterns include: creating notebooks from code, uploading documents for targeted querying, performing deep research with citations, and exporting chat histories for auditing. You can also enable multi‑session operation with isolated browser profiles to keep user sessions separate.

How to install

Prerequisites: you need Node.js and npm or a compatible runtime to install the MCP server client. Ensure your environment can run npm or npx commands.

# Quick install via Claude Code workflow (no authentication shown here)
claude mcp add notebooklm -- npx @pan-sec/notebooklm-mcp@latest

If you require authentication and Gemini integration, you can pass environment variables to enable authentication and Gemini access when starting the MCP client.

claude mcp add notebooklm \
  --env NLMCP_AUTH_ENABLED=true \
  --env NLMCP_AUTH_TOKEN=$(openssl rand -base64 32) \
  --env GEMINI_API_KEY=your-gemini-api-key \
  -- npx @pan-sec/notebooklm-mcp@latest

Configuration and environment

All security features are enabled by default. You can override settings with environment variables to tailor authentication, encryption, session lifetimes, and monitoring.

# Example environment configuration for starting the MCP server client
export NLMCP_AUTH_ENABLED=true
export NLMCP_AUTH_TOKEN=YOUR_SECURE_TOKEN
export GEMINI_API_KEY=YOUR_GEMINI_API_KEY
export GEMINI_DEFAULT_MODEL=gemini-2.5-flash
export GEMINI_DEEP_RESEARCH_ENABLED=true
export GEMINI_TIMEOUT_MS=30000

Notes on security and compliance

The server includes post‑quantum encryption, certificate pinning, memory scrubbing, audit logging with tamper‑evident hashes, and GDPR/SOC2/CSSF readiness. It supports per‑session isolation and redacts credentials in outputs. Review the security settings to ensure they align with your organization's policies.

Troubleshooting and tips

If you encounter authentication or connectivity issues, verify that the GEMINI_API_KEY and NLMCP_AUTH_ENABLED tokens are correctly set and that the environment variables are visible to the MCP client process.

For issues with multi‑session profiles, ensure you have isolated profiles configured and that any required cloning behavior is enabled as described in the configuration notes.

Available tools

ask_question

Query your NotebookLM notebooks with grounding to your uploaded sources via the browser-based UI or API.

deep_research

Perform comprehensive research with citations using the Gemini Deep Research backend; returns sources and structured answers.

gemini_query

Fast, grounded queries with Google Search grounding, code execution, and URL context support.

get_research_status

Check the progress of ongoing background deep research tasks.

add_notebook

Add a new notebook to your library programmatically.

list_notebooks

List all notebooks in your library.

get_notebook

Retrieve details for a specific notebook.

update_notebook

Update notebook metadata such as title or description.

remove_notebook

Remove a notebook from your library.

select_notebook

Set the active notebook for subsequent queries.

search_notebooks

Search notebooks by query to locate relevant knowledge bases.

manage_sources

Add, remove, or list sources within a notebook.

generate_audio

Generate an Audio Overview podcast for a notebook.

sync_notebook

Sync notebook sources from a local directory.

get_query_history

Review past MCP queries with filtering and search.

get_notebook_chat_history

Extract and paginate notebook chat history with optional export.

setup_auth

Initial authentication setup for the MCP server.

re_auth

Re-authenticate to refresh session credentials.

cleanup_data

Deep cleanup utility to remove sensitive data.

get_health

Server health check with deep UI verification capability.

get_library_stats

Retrieve statistics about your notebook library.

get_quota

Check usage quotas and remaining resources.