[Cotlet](https://cotlet.dev/index.md) · [Integrations](https://cotlet.dev/integrations.md) · [How it works](https://cotlet.dev/how-it-works.md) · [Security](https://cotlet.dev/security.md) · [For developers](https://cotlet.dev/for-developers.md) · [For automation agencies](https://cotlet.dev/for-agencies.md) · [Blog](https://cotlet.dev/blog.md) · [API docs](https://cotlet.dev/docs.md) · [Help center](https://cotlet.dev/help.md) · [Changelog](https://cotlet.dev/changelog.md) · [Contact](https://cotlet.dev/contact.md) · [Pricing](https://cotlet.dev/pricing.md) · [Sign up](https://cotlet.dev/signup) # NotebookLM MCP server: Let your AI agent fill and question your notebooks. > Google has no MCP server for NotebookLM and no API for personal accounts (checked 6 Nov 2026). Its one official API is for NotebookLM Enterprise, in v1alpha preview. Cotlet adds 14 NotebookLM tools for the account you already have: no API key, no Google Cloud project, hosted 24/7, inside the ChatGPT and Claude apps, with one Cotlet account for many apps. Create NotebookLM notebooks, add sources, ask questions and generate Audio Overviews from Claude, ChatGPT, n8n or a script. Sign in to Google once in your Cloud Browser. Nothing to install, nothing to host. [MCP setup](https://cotlet.dev/integrations/notebooklm-api-mcp-server.md#connect) [REST API guide](https://cotlet.dev/integrations/notebooklm-api-mcp-server.md#rest-api) - Free 7-day trial, no card - Unofficial, not affiliated with Google - Page: https://cotlet.dev/integrations/notebooklm-api-mcp-server ## What can your AI agent do in NotebookLM? - Build a notebook from a list of links - Find sources on a topic and import them - Ask a notebook, with citations - Save an answer as a note - Generate and download an Audio Overview - Share a notebook with your team ## Does NotebookLM have an MCP server or an API? - The Enterprise API creates, lists, shares and deletes notebooks, manages sources, and creates or deletes one Audio Overview per notebook. It needs an Enterprise license and a Google Cloud role. Its developer pages show no chat or notes methods. - Google's help center now calls the product Gemini Notebook. The app still opens at notebooklm.google.com. - Zapier lists NotebookLM as "Not Yet Supported". The n8n integrations directory has no NotebookLM node. | What you need | Enterprise API | GitHub MCP server | Cotlet | |---|---|---|---| | A free or personal Google account | No, Enterprise license | Yes | Yes | | Create notebooks, add sources, share | Yes, in preview | Yes | Yes, sharing after your OK | | Ask a notebook, with citations | Not documented | Yes | Yes | | Save and read notes | Not documented | Yes | Yes | | Generate and download an Audio Overview | Create and delete only | Yes | Yes | | No API key, no install, runs 24/7 | Google Cloud project and login | Runs on your computer | Hosted, in Claude and ChatGPT | Checked 6 Nov 2026. Sources: [NotebookLM Enterprise: notebooks API](https://docs.cloud.google.com/gemini/enterprise/notebooklm-enterprise/docs/api-notebooks), [NotebookLM Enterprise: sources API](https://docs.cloud.google.com/gemini/enterprise/notebooklm-enterprise/docs/api-notebooks-sources), [NotebookLM Enterprise: Audio Overview API](https://docs.cloud.google.com/gemini/enterprise/notebooklm-enterprise/docs/api-audio-overview), [Zapier: Google NotebookLM not yet supported](https://zapier.com/apps/google-notebooklm-upcoming/integrations), [Gemini Notebook Help: plans](https://support.google.com/notebooklm/answer/16213268) ## How does Cotlet connect to NotebookLM? 1. **Sign in to NotebookLM.** Use your own Google account in your Cloud Browser. 2. **Connect your AI agent.** Add the Cotlet address, or make an API key. 3. **Ask in plain words.** Build notebooks, ask them, make audio. We never keep your Google password, and a notebook is not shared or deleted until you approve it. ## Connect NotebookLM to your AI agent Add one address to Claude, ChatGPT, Cursor, Codex or n8n. NotebookLM appears next to your other apps once you enable it. 1. **Enable NotebookLM for agents.** Sign up, sign in to your Google account in the Cloud Browser, then press Enable for agents on NotebookLM. 2. **Add the Cotlet address.** Add https://cotlet.dev/mcp/v1 as an MCP server or connector in your agent. 3. **Approve the connection.** Chat apps ask you to sign in to Cotlet. Editors, n8n and scripts use an API key from the console. One shared address, `https://cotlet.dev/mcp/v1`, also carries NotebookLM next to your other apps. If you enable NotebookLM after connecting, refresh the connector so the new tools show up. Claude: ``` Settings > Connectors > Add custom connector URL: https://cotlet.dev/mcp/v1/connect Claude Code: claude mcp add --transport http cotlet \ https://cotlet.dev/mcp/v1 \ --header "Authorization: Bearer $COTLET_API_KEY" ``` ChatGPT: ``` Settings > Apps & Connectors > Create MCP server URL: https://cotlet.dev/mcp/v1/connect Authentication: OAuth ``` Cursor: ``` // ~/.cursor/mcp.json { "mcpServers": { "cotlet": { "url": "https://cotlet.dev/mcp/v1", "headers": { "Authorization": "Bearer ${env:COTLET_API_KEY}" } } } } ``` Codex: ``` # ~/.codex/config.toml [mcp_servers.cotlet] url = "https://cotlet.dev/mcp/v1" bearer_token_env_var = "COTLET_API_KEY" ``` n8n: ``` MCP Client Tool (inside an AI Agent) Transport: HTTP Streamable URL: https://cotlet.dev/mcp/v1 Auth: Bearer, ``` Make, Zapier: ``` Make: MCP Client > Call a tool Zapier: MCP Client by Zapier > Run Tool Server URL: https://cotlet.dev/mcp/v1 Token: ``` ## How to use the NotebookLM REST API ### Create your Cotlet API key Sign in to Google in your Cloud Browser and enable NotebookLM for agents. What a call can do follows your access to each notebook: owner, editor or viewer. Create a Cotlet API key with access to NotebookLM and its read operations. Store it as `COTLET_API_KEY` in your environment or your automation platform's credential store. ### Make your first NotebookLM API request cURL: list_notebooks: ``` curl https://cotlet.dev/api/v1/notebooklm/notebooks \ -H "Authorization: Bearer $COTLET_API_KEY" ``` ### Read notebooks and continue the list A successful REST call puts the tool output in `result`. This illustrative response excerpt omits call IDs, messages and credit details. Illustrative response excerpt: ``` { "status": "success", "app": "notebooklm", "tool": "list_notebooks", "result": { "notebooks": [ { "notebook_id": "nb_7f3a91c2", "title": "Customer interviews", "sources": 12, "role": "owner", "last_opened": "2026-11-03" }, { "notebook_id": "nb_2c8d40e1", "title": "Weekly reading", "sources": 7, "role": "owner", "last_opened": "2026-11-04" }, { "notebook_id": "nb_91b6d3a7", "title": "Q4 research", "sources": 5, "role": "editor", "last_opened": "2026-10-28" } ], "next_cursor": null } } ``` Read `result.notebooks` and pass `result.next_cursor` unchanged as `cursor` until it is null. Keep notebook IDs in your workflow so later runs update the same notebook. Source text is not returned by `list_sources`. An Audio Overview takes a couple of minutes, so call `generate_audio_overview`, poll `list_studio_outputs` and fetch the link with `download_audio_overview` when the status is ready. ### NotebookLM workflows for agents and automations | Tool | Connect through | Example workflow | |---|---|---| | n8n | HTTP Request | Each Monday, call `add_source` for the links in your saved-articles sheet, then `generate_audio_overview` and post the download link to Slack. | | Python or Node.js | REST API | Call `ask_notebook` with a fixed question per notebook and write the answers and citations to a report. | | Make | HTTP > Make a request | Call `list_notebooks` and `list_sources` to build a sheet of notebooks and their failed imports. | Send `Authorization: Bearer ` in request headers. You configure the schedule, destination and comparisons. Cotlet reads and writes NotebookLM; your workflow steps write to the destination you choose. These HTTP workflows need no native Cotlet connector and no prebuilt template. The [API and MCP reference](https://cotlet.dev/integrations/notebooklm-api-mcp-server.md#tools) lists operations and inputs. The [REST API guide](https://cotlet.dev/docs/api.md) explains authentication, errors and approvals. ## NotebookLM API endpoints and MCP tools ### Notebooks Only the owner can delete a notebook. Sharing gives viewer or editor access by email. - **List notebooks** (`list_notebooks`, read): Lists the notebooks in your NotebookLM home. - **Get a notebook** (`get_notebook`, read): Returns one notebook with its sources, notes and Studio outputs. - **Create a notebook** (`create_notebook`, write): Creates an empty notebook. - **Share a notebook** (`share_notebook`, write, asks first): Gives people access to a notebook by email. - **Delete a notebook** (`delete_notebook`, write, asks first): Deletes a notebook you own. ### Sources and research Web pages, public YouTube videos with captions and pasted text. Files and Google Drive items are not added by these tools. - **List sources** (`list_sources`, read): Lists the sources of a notebook. - **Search for sources** (`search_sources`, read): Runs NotebookLM's Fast Research to find sources for a topic. - **Add a source** (`add_source`, write): Adds a web page, a YouTube video or pasted text to a notebook. ### Chat and notes Answers come only from the notebook's sources. Nothing becomes a note until you save it. - **Ask a notebook** (`ask_notebook`, read): Asks a question in the notebook chat and returns the answer with its citations. - **List notes** (`list_notes`, read): Lists the notes saved in a notebook. - **Save a note** (`save_note`, write): Saves a note in a notebook. ### Audio Overviews and Studio Generating needs edit access. An Audio Overview finishes within a couple of minutes. - **List Studio outputs** (`list_studio_outputs`, read): Lists what NotebookLM generated from the sources of a notebook. - **Generate an Audio Overview** (`generate_audio_overview`, write): Starts an Audio Overview from the notebook's sources. - **Download an Audio Overview** (`download_audio_overview`, read): Returns a download link for a finished Audio Overview. Inputs, REST path and example response of every tool: https://cotlet.dev/integrations/notebooklm-api-mcp-server#tools ## Who can see my NotebookLM session, and how do I revoke it? - **What we store.** Your signed-in Google profile, encrypted with AES-256-GCM, in Finland. Not the password. - **What your agent gets.** The 14 NotebookLM tools. Never the cookie. - **What the session sees.** The notebooks you can open: your own and those shared with you. - **What no tool touches.** Google plan, password, two-factor and your other Google apps. - **If Google asks to verify.** A new device may ask for a code. You type it in, we never do. - **How to cut access.** Disable for agents, reset the browser, or sign out in your Google account. Our site scores A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory ([live results](https://cotlet.dev/security.md)). There is no SOC 2 report or third-party pentest yet. ## What can my agent reach, and what are the NotebookLM limits? | Plan | Notebooks | Sources per notebook | |---|---|---| | Standard (free) | 100 | 50 | | Google AI Plus | 200 | 100 | | Google AI Pro | 500 | 300 | | Google AI Ultra (20 TB) | 500 | 500 | | Google AI Ultra (30 TB) | 500 | 600 | What stays the same with an agent: - Your Google account and your plan, billed by Google - Owner, editor or viewer rights on each notebook - Edit access to generate or delete an Audio Overview - Answers only from the notebook's sources > Google marks these limits "Subject to Change". Sharing a notebook does not change the source limit for anyone you share it with. ## What happens when NotebookLM changes? - **12 h** between tests of the NotebookLM tools - **$0** for a call to a NotebookLM tool paused for repair Google has renamed the product once already: its help center now says Gemini Notebook. When NotebookLM moves a button or renames a field, the fix happens on our side and your prompts keep working. [status.cotlet.dev](https://status.cotlet.dev) watches the site, console and API every 3 minutes. ## Approvals, undo and mistakes 1. **Creates the notebook.** A new, empty notebook that you own. 2. **Adds your links.** Web pages and public YouTube videos, one by one. 3. **Checks the imports.** Lists each source with its status and reason if it failed. 4. **Waits for your OK.** Sharing the notebook shows up in your Cotlet console first. | Action | Undo | |---|---| | Create a notebook | Delete it; the delete tool asks first | | Add a source | Remove it in the Sources panel | | Save a note | It sits in the Sources panel next to your sources; remove it there | | Generate an Audio Overview | Delete it in the Studio panel, which needs edit access | | Share a notebook | Change access in NotebookLM; the invitation is already sent | | Delete a notebook | No, not through Cotlet | 2 of 14 tools ask first: share_notebook and delete_notebook. Creating a notebook, adding a source, saving a note and generating audio reach no one, so they run at once. Switch the wait off per tool for a scheduled run. > Deleting a notebook removes its sources, notes and Studio outputs, and links shared from it stop working. MCP call: ``` { "name": "generate_audio_overview", "arguments": { "notebook_id": "nb_2c8d40e1", "format": "brief", "length": "default", "prompt": "Focus on what changed since last week." } } ``` REST: ``` curl -X POST https://cotlet.dev/api/v1/notebooklm/notebooks/nb_2c8d40e1/audio-overview/generate \ -H "Authorization: Bearer $COTLET_API_KEY" \ -H "Content-Type: application/json" \ -d '{"format":"brief","length":"default","prompt":"Focus on what changed since last week."}' ``` Tool output: ``` { "output_id": "out_5e1", "notebook_id": "nb_2c8d40e1", "format": "brief", "status": "generating" } ``` Scripts and cron jobs reach the same tools over REST with the same key and price. A failed call costs nothing. The tools table lists the REST path of each tool. ## How much does the NotebookLM MCP server cost? **About $18 a month** for a researcher building 4 notebooks and one weekly Audio Overview, 3,060 credits. Your Google plan is billed by Google as before. | Job | Calls | Credits | |---|---|---| | Build a notebook from 10 links | 12 × 4 | 480 | | Weekly Audio Overview of the reading | 9 × 4 | 360 | | Ask a notebook and save the answer | 2 × 20 | 400 | | Find and import sources on a topic | 7 × 4 | 280 | | Share a notebook with the team | 2 × 2 | 40 | | Cloud Browser, 30 days | | 1,500 | | Total | | 3,060 | - Free 7-day trial, no card - 10 credits per successful call, MCP or REST - 50 credits a day for the Cloud Browser - Works on the free NotebookLM plan too The month is our example, not a measured average: Google publishes no typical account size. It fits inside the free plan's 100 notebooks and 50 sources per notebook. Dollars at the $29 pack rate. ## NotebookLM MCP server on GitHub, Google or hosted: which to use | Option | What it does | Where it stops | |---|---|---| | [jacob-bd/gemini-notebook-mcp-cli](https://github.com/jacob-bd/gemini-notebook-mcp-cli) | MCP server and CLI, 55 tools, MIT, 6,468 stars, last commit 5 Nov 2026. | You install and run it on your computer. Its sign-in cookies last about 2 to 4 weeks. | | [teng-lin/notebooklm-py](https://github.com/teng-lin/notebooklm-py) | Python library and CLI with agent skills, MIT, 20,133 stars, last commit 5 Nov 2026. | A library for developers, run on your own machine. | | [roomi-fields/notebooklm-mcp](https://github.com/roomi-fields/notebooklm-mcp) | MCP server plus a local HTTP REST API for n8n, Zapier and Make, 221 stars, last commit 30 Oct 2026. | Runs on your own machine. | | [PleasePrompto/notebooklm-mcp](https://github.com/PleasePrompto/notebooklm-mcp) | MCP server that drives Chrome, 3,443 stars. | Archived, last commit 10 Sep 2026. | | [Google NotebookLM Enterprise API](https://docs.cloud.google.com/gemini/enterprise/notebooklm-enterprise/docs/api-notebooks) | Official REST API for notebooks, sources and Audio Overviews. | Enterprise license and Google Cloud role, v1alpha preview, no chat or notes methods documented. | | [Kortex MCP](https://www.kortex-notebooklm.com/blog/gemini-notebook-in-claude) | Hosted MCP address that relays calls through a Chrome extension; $9.99 a month on Kortex Pro. | Chrome must stay open with the extension signed in. | A GitHub server is free and fine on your own laptop. Google's Enterprise API suits a company with licences and a Cloud project. Cotlet is the hosted option for the account you already have: it keeps working when your computer is off and runs inside the ChatGPT and Claude apps. ## Who runs Cotlet Take Group sp. z o.o. runs Cotlet. It is a software company registered in Poland, EU (VAT PL6342984273), working since 2019, founded by Bart Krawczyk (Co-founder, CTO) and Jake Slugocki (Co-founder, SEO, sales and marketing). Write to contact@cotlet.dev with questions about the NotebookLM tools. ## NotebookLM MCP server: questions ### How do I connect NotebookLM to ChatGPT or Claude? Sign up at cotlet.dev, log in to NotebookLM once in its Cloud Browser and add Cotlet to ChatGPT or Claude, following the steps in your console. No API key is needed. ChatGPT and Claude then work in your NotebookLM account through Cotlet, in their phone apps too. ### What can ChatGPT do in NotebookLM? With Cotlet, ChatGPT, Claude and other AI agents handle jobs like these in NotebookLM: build a notebook from a list of links; turn weekly reading into an Audio Overview; ask a notebook and save the answer. You ask in plain words, and anything that reaches other people waits for your OK first. ### How do I get access? Create a free account at https://cotlet.dev, open NotebookLM under Integrations in the console and sign in to Google in the Cloud Browser. The trial starts with your first app, no card asked. ### Does this work with a personal Google account? Yes. Cotlet works in the NotebookLM you open with your Google account, so it does not need the Enterprise license that the official API requires (Google Cloud docs, checked 5 Nov 2026). Your access to each notebook still counts: generating or deleting an Audio Overview needs edit access, and only the owner can delete a notebook. ### Which source types can my agent add? Web pages, public YouTube videos and pasted text. NotebookLM imports only the text of a web page and only the transcript of a YouTube video, and a video needs captions (Google's help center). Uploading files and picking Google Drive items are not covered by these tools. ### Is NotebookLM the same as Gemini Notebook? Yes. Google's help center now calls NotebookLM "Gemini Notebook" (checked 6 Nov 2026). The app is the same, and Cotlet uses it the way you do in the browser. ### Can I use the NotebookLM API without an AI agent? Yes. Call Cotlet's REST API from a script or an HTTP step in n8n, Make or Zapier with your Cotlet key. MCP is the alternative for AI agents; you do not need to connect an agent to use REST. ### Is the MCP server different from the REST API? Only the door is different. Both expose the same 14 tools, inputs and answers, with one API key and one price. Agents use MCP; scripts and HTTP steps use REST. ### Is Cotlet affiliated with Google? No. Cotlet is independent and not affiliated with Google. --- Give your AI agents your signed-in browser. **Product:** [Integrations](https://cotlet.dev/integrations.md) · [How it works](https://cotlet.dev/how-it-works.md) · [Pricing](https://cotlet.dev/pricing.md) · [Security](https://cotlet.dev/security.md) **Resources:** [Blog](https://cotlet.dev/blog.md) · [API docs](https://cotlet.dev/docs.md) · [Help center](https://cotlet.dev/help.md) · [Changelog](https://cotlet.dev/changelog.md) · [GitHub](https://github.com/cotletdev/cotlet) **Solutions:** [For developers](https://cotlet.dev/for-developers.md) · [For automation agencies](https://cotlet.dev/for-agencies.md) **Company:** [About](https://cotlet.dev/about.md) · [AI info: company facts](https://cotlet.dev/ai-info.md) · [Contact](https://cotlet.dev/contact.md) **Legal:** [Terms of service](https://cotlet.dev/terms) · [Privacy policy](https://cotlet.dev/privacy) · [Refund policy](https://cotlet.dev/refunds) · [Acceptable use](https://cotlet.dev/acceptable-use) · [Cookie policy](https://cotlet.dev/cookies) · [DPA](https://cotlet.dev/dpa) · [Subprocessors](https://cotlet.dev/subprocessors) **Elsewhere:** [Status page](https://status.cotlet.dev) · [Cotlet on GitHub](https://github.com/cotletdev) · [Cotlet on X](https://x.com/cotletdev) · [Cotlet on LinkedIn](https://www.linkedin.com/company/cotletdev) · [Cotlet on Product Hunt](https://www.producthunt.com/products/cotlet) · [Cotlet on Crunchbase](https://www.crunchbase.com/organization/cotlet) · [Cotlet on SaaSHub](https://www.saashub.com/cotlet) Cotlet.dev is built by Bart Krawczyk and Jake Slugocki, the founders of Take Group sp. z o.o., ul. Wincentego Janasa 4, 40-855 Katowice, Poland. KRS 0000871398, NIP 6342984273, VAT EU: PL6342984273. [About us](https://cotlet.dev/about.md) Cotlet is an independent service. It is not affiliated with, endorsed by or sponsored by any of the products listed on this site. Product names are trademarks of their respective owners.