notebooklm-mcp-server
Provides seamless integration with Google NotebookLM, enabling AI agents to manage notebooks, add sources (including YouTube, Google Drive, and local files), query notebook content, and generate audio overviews and mind maps.
Allows adding Google Drive files (Docs, Slides, etc.) as sources to notebooks and syncing them to keep content up-to-date.
Enables adding YouTube videos as sources to notebooks, allowing agents to query video content and use it for grounded answers.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@notebooklm-mcp-server Ask my notebook about quantum computing breakthroughs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
π‘ What is this?
Google's Gemini Notebook (renamed from NotebookLM in July 2026) is the best grounded-answer engine on the market: it only answers from the sources you give it, with citations. This MCP server hands that power to your AI agents β Claude, Gemini, Antigravity, Cursor and any other MCP client β so they can build notebooks, run Deep Research, generate podcasts, videos, quizzes, flashcards, reports and mind maps, take notes, and share the results with your team. All programmatically, with zero hallucinations.
NotebookLM is now Gemini Notebook. Google renamed the product on July 16, 2026 (see notebook.google). Same product, same notebooks, same API surface β this server keeps working unchanged, and the npm package keeps its notebooklm-mcp-server name.
Related MCP server: notebooklm-py-diet-mcp
π Installation
Option 1 β Global install (recommended)
npm install -g notebooklm-mcp-serverThe server checks for updates on startup and keeps itself current automatically.
Option 2 β Zero-install with NPX
npx -y notebooklm-mcp-server auth # authenticate
npx -y notebooklm-mcp-server start # run the serverπ Authentication
Run the interactive login (opens a Chromium window):
npx notebooklm-mcp-server authSign in with your Google account. When the notebook list appears, the session cookies (including Google's rotating
__Secure-1PSIDTStoken) are captured and saved to~/.notebooklm-mcp/auth.json.Done β the server loads the session automatically and refreshes the rotating token by itself.
If the session ever expires, runnpx notebooklm-mcp-server auth again in a terminal, then call the refresh_auth MCP tool (or just restart your client) to pick up the new cookies without reconfiguring anything.
β‘ Connect Your AI Client
π€ Claude Code
claude mcp add notebooklm -- npx -y notebooklm-mcp-server start㪠Claude Desktop
Add to claude_desktop_config.json (Settings β Developer β Edit Config):
{
"mcpServers": {
"notebooklm": {
"command": "npx",
"args": ["-y", "notebooklm-mcp-server", "start"]
}
}
}π Antigravity CLI
Antigravity's CLI manages MCP servers through a JSON config file:
Global:
~/.gemini/config/mcp_config.jsonPer-workspace:
.agents/mcp_config.jsonin your project
{
"mcpServers": {
"notebooklm": {
"command": "npx",
"args": ["-y", "notebooklm-mcp-server", "start"]
}
}
}Then type /mcp in the Antigravity prompt panel to open the MCP Manager: check the connection status ring, reload the config, or inspect logs β no restart needed. Your agents can now cite real sources in every mission.
π₯οΈ Antigravity IDE
The IDE reads the same config files as the CLI (~/.gemini/config/mcp_config.json or .agents/mcp_config.json). Alternatively, open the MCP Store panel, choose Add custom server, and paste the same JSON snippet. Reload the server list and the NotebookLM tools appear in the agent's toolbox.
π Gemini CLI
gemini mcp add notebooklm --scope user -- npx -y notebooklm-mcp-server startβ¨οΈ Cursor
Add to .cursor/mcp.json in your project (or ~/.cursor/mcp.json globally):
{
"mcpServers": {
"notebooklm": {
"command": "npx",
"args": ["-y", "notebooklm-mcp-server", "start"]
}
}
}π Windsurf
Add the same mcpServers block to ~/.codeium/windsurf/mcp_config.json.
π§© VS Code (Cline)
In Cline: MCP Servers β Configure and add the same mcpServers block shown above.
π Tool Reference
47 tools, grouped by what they do. Full parameter signatures live in docs/TOOLS.md.
π Notebooks
Tool | Description |
| List all notebooks with sources and metadata |
| Create a new notebook |
| Get one notebook's details and source IDs |
| Rename a notebook |
| Delete a notebook (requires |
| AI notebook guide: summary + suggested questions |
| AI-suggested prompts to ask about the sources |
π Sharing
Tool | Description |
| Toggle public link and/or manage collaborators by email |
| Read the current sharing configuration |
ποΈ Sources
Tool | Description |
| Add a website or YouTube video as a source |
| Add pasted text as a source |
| Add a Google Drive document as a source |
| Upload a local PDF / TXT / Markdown file |
| AI guide for one source: summary + key topics |
| Rename a source |
| Check if a URL/Drive source has newer content |
| Re-sync a Drive source to the latest content |
| Delete a source (requires |
π¬ Chat
Tool | Description |
| Ask questions about the sources, with citations |
| Read the latest conversation's Q&A turns |
| Clear the chat history (requires |
| Set the chat persona (default / learning guide / custom) and response length |
π Research
Tool | Description |
| Launch web or Drive research (fast β30s / deep β5min) |
| Check research progress and discovered sources |
| Import the discovered sources into the notebook |
π¨ Studio
Tool | Description |
| Podcast-style Audio Overview (formats: deep dive, brief, critique, debate) |
| Video Overview (9 visual styles, from whiteboard to anime) |
| Written report from the sources |
| Flashcards (quantity and difficulty options) |
| Interactive quiz (quantity and difficulty options) |
| Infographic (landscape / portrait / square) |
| Slide deck |
| Structured data table |
| Check generation status of all artifacts |
| Delete a studio artifact |
π¦ Artifacts
Tool | Description |
| Fetch generated content (quiz/flashcards HTML, mind map JSON) |
| Rename an artifact |
| Export an artifact to Google Drive (Docs or Sheets) |
π Notes & Mind Maps
Tool | Description |
| Create a note |
| List all notes |
| Update a note's content/title |
| Delete a note (requires |
| Generate a mind map JSON from sources |
| Save a mind map to the notebook |
| List saved mind maps |
| Delete a mind map |
βοΈ System
Tool | Description |
| Reload session cookies from disk after re-running |
π§ͺ Recipes
Real prompts you can paste into any connected agent β it picks the right tools by itself.
ποΈ From research question to podcast
"Research the latest on solid-state batteries with deep research, build a notebook called 'Solid State 2026' with the best sources, and generate a brief audio overview in English. Give me the link when it's done."
The agent chains notebook_create β research_start(mode: deep) β research_poll β research_import β audio_overview_create(format: brief) β studio_poll, and returns the notebook URL with the finished podcast.
π Study kit from your PDFs
"Take the three PDFs in ./lectures, put them in a new notebook, and make me a hard quiz plus flashcards on the key concepts. Show me the quiz questions here."
The agent runs notebook_add_local_file per PDF, then quiz_create(difficulty: hard) + flashcards_create, polls with studio_poll, and pulls the questions out with artifact_content_get.
π₯ Team briefing, shared
"Build a briefing notebook from these five links about our competitor, write a report, export it to Google Docs, and share the notebook with ana@example.com as a viewer."
The agent chains notebook_add_url Γ5 β report_create β studio_poll β artifact_export(format: docs) β notebook_share(user_email, user_role: viewer) β and your teammate finds the notebook and the Doc waiting.
π οΈ Troubleshooting
Symptom | Fix |
| Your Google account enforces Device Bound Session Credentials (DBSC) β default-on for Workspace and most personal accounts. DBSC ties the session to a device-held key, so cookies exported from the browser are refused by the API. Cookie-extraction clients cannot bypass this; use a Google account without DBSC enforcement. |
| Update to β₯ 3.0.8 β older versions missed Google's rotating |
| Google sessions eventually expire. Run |
Studio generation stuck on | Long sources take a while β keep polling |
RPC errors after a Google update | Google occasionally rotates internal build labels. Update the package; if it persists, open an issue. |
π§βπ» Development
git clone https://github.com/moodRobotics/notebooklm-mcp-server.git
cd notebooklm-mcp-server
npm install
npm run build
npm run typecheckArchitecture notes live in docs/ARCHITECTURE.md, the roadmap in docs/ROADMAP.md.
π Localization
This README is available in English, EspaΓ±ol, FranΓ§ais, PortuguΓͺs and Deutsch. If you edit one, please keep the others in sync (npm run docs:check).
π License
MIT β Developed with β€οΈ by moodRobotics.
This is an unofficial community project. It is not affiliated with or endorsed by Google. It relies on reverse-engineered internal APIs that may change without notice; use it with a Google account you are comfortable automating.
Available Tools
29 toolsaudio_overview_createB
Generate an Audio Overview (podcast) for the notebook
Inputs: notebook_id (required), language, source_ids, focus_prompt.
chat_configureB
Configure notebook chat settings (goal, response length)
Inputs: notebook_id (required), goal, custom_prompt, response_length.
data_table_createC
Generate a data table from notebook sources
Inputs: notebook_id (required), language, source_ids, focus_prompt.
flashcards_createC
Generate flashcards from notebook sources
Inputs: notebook_id (required), language, source_ids, focus_prompt.
infographic_createC
Generate an infographic from notebook sources
Inputs: notebook_id (required), language, source_ids, focus_prompt.
mind_map_deleteC
Delete a Mind Map from a notebook
Inputs: mind_map_id (required), notebook_id (required).
mind_map_generateB
Generate a Mind Map JSON from notebook sources
Inputs: source_ids (required).
mind_map_listC
List all Mind Maps in a notebook
Inputs: notebook_id (required).
mind_map_saveA
Save a generated Mind Map to a notebook
Inputs: source_ids (required), notebook_id (required), mind_map_json (required), title.
notebook_add_driveC
Add a Google Drive document as a source
Inputs: title (required), document_id (required), notebook_id (required), doc_type.
notebook_add_local_fileB
Add a local PDF or text/markdown file as a source
Inputs: path (required), notebook_id (required).
notebook_add_textB
Add pasted text as a source to a notebook
Inputs: title (required), content (required), notebook_id (required).
notebook_add_urlC
Add a URL (website or YouTube) as a source to a notebook
Inputs: url (required), notebook_id (required).
notebook_createB
Create a new notebook
Inputs: title (required).
notebook_deleteA
Delete a notebook permanently. IRREVERSIBLE.
Inputs: confirm (required), notebook_id (required).
notebook_listA
List all notebooks with their sources and metadata
Inputs: no named parameters.
notebook_queryA
Ask AI about EXISTING sources in a notebook. NOT for finding new sources.
Inputs: query (required), notebook_id (required), source_ids, conversation_id.
notebook_renameD
Rename a notebook
Inputs: title (required), notebook_id (required).
refresh_authA
Reload authentication cookies from disk. Run 'notebooklm-mcp-server auth' in a terminal first if cookies are expired, then call this tool to pick up the new cookies.
Inputs: no named parameters.
report_createC
Generate a written report from notebook sources
Inputs: notebook_id (required), language, source_ids, focus_prompt.
research_importB
Import discovered research sources into a notebook
Inputs: sources (required), task_id (required), notebook_id (required).
research_pollC
Poll for research status and results
Inputs: notebook_id (required).
research_startA
Start web or Drive research to find NEW sources. Workflow: research_start β research_poll β research_import
Inputs: query (required), notebook_id (required), mode, source.
slide_deck_createC
Generate a slide deck from notebook sources
Inputs: notebook_id (required), language, source_ids, focus_prompt.
source_deleteA
Delete a source permanently. IRREVERSIBLE.
Inputs: confirm (required), source_id (required).
source_syncB
Sync a Google Drive source to get the latest content
Inputs: source_id (required).
studio_deleteC
Delete a studio artifact
Inputs: artifact_id (required), notebook_id (required).
studio_pollC
Check status of studio artifacts (Audio, Video, Report, etc.)
Inputs: notebook_id (required).
video_overview_createC
Generate a Video Overview for the notebook
Inputs: notebook_id (required), language, source_ids, focus_prompt.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
TDQS
Scored across 29 tools
Each tool targets a distinct operation or resource. Overlaps exist (e.g., multiple 'create' tools) but are disambiguated by their specific output types (audio, video, flashcards, etc.) and clear descriptions.
Most tools follow a <resource>_<action> pattern (e.g., notebook_create, mind_map_delete). The outlier 'refresh_auth' uses verb_noun instead, which is a minor inconsistency.
29 tools cover a wide range of NotebookLM features (notebooks, sources, multiple outputs, research, mind maps, studio). While slightly high, each tool serves a distinct purpose within the scope.
The tool surface covers the full lifecycle: CRUD for notebooks and sources, generation of various outputs, research workflow, and studio management. No obvious gaps for the intended use cases.
Maintenance
Related MCP Connectors
Google NotebookLM via natural language: create notebooks, add sources (PDF, URL, YouTube) and ask gr
Google Gemini Notebook (NotebookLM): notebooks, sources, cited chat, podcasts, video, slides
Persistent memory and knowledge management for AI agents with semantic search and 50+ tools.
Access Pollinations models and API capabilities through agent tools.
Related MCP Servers
- FlicenseAqualityDmaintenanceEnables AI agents to query and interact with Google NotebookLM notebooks to retrieve citation-backed information. It provides tools for listing notebooks, accessing source data, and asking natural language questions.11-
- AlicenseAqualityDmaintenanceConnects AI coding agents to Google NotebookLM, enabling querying notebooks, adding sources, and generating artifacts like reports, podcasts, and slide decks.141MIT
- AlicenseAqualityDmaintenanceEnables AI assistants to programmatically access and control Google NotebookLM, supporting operations like notebook management, source addition, audio generation, and more via natural language.39MIT
- AlicenseNot gradedqualityDmaintenanceEnables interaction with Google NotebookLM to create notebooks, add sources (PDF, URL, YouTube), and ask questions with citations.1MIT