NotebookLM API | Export Conversations, Sources & Citations

Google doesn't let you export your NotebookLM data. This Actor fixes that - get your conversations, sources, and citations out as JSON, markdown (LLM ready) or CSV.

• Full conversation history with Q&A threads
• Source metadata (titles, URLs)
• Citation mapping - see exactly which sources informed each AI response
• Bulk export or pick specific notebooks

🎉 Free during beta! No limits, no payment required. This won't last forever, so try it out while you can.

Built this after seeing people complain on Reddit about the lack of exports. If you need a feature that's not here yet, just ask - I'm actively building this out.

---

Quick Start

Basic - Export Recent Notebooks

{
  "email": "your-email@gmail.com",
  "appPassword": "xxxx xxxx xxxx xxxx"
}

Exports up to 10 most recent notebooks (default limit).

Export All Notebooks

{
  "email": "your-email@gmail.com",
  "appPassword": "xxxx xxxx xxxx xxxx",
  "limit": 0
}

Set limit: 0 to export all notebooks without restriction.

Export Specific Notebooks

{
  "email": "your-email@gmail.com",
  "appPassword": "xxxx xxxx xxxx xxxx",
  "projectIds": [
    "6509c6d4-58fc-4066-8aa9-b691423e5ca7",
    "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
  ]
}

With Source Summaries & Tags

{
  "email": "your-email@gmail.com",
  "appPassword": "xxxx xxxx xxxx xxxx",
  "includeSourceSummaries": true
}

Note: Enabling includeSourceSummaries adds ~3 seconds per source to avoid rate limiting. For notebooks with many sources, runs will take longer.

---

Input Parameters

Parameter	Type	Required	Description
email	String	Yes	Your Google account email address
appPassword	String	Yes	16-character App Password (NOT your Google password!)
projectIds	Array	No	UUIDs of specific notebooks to export. If empty, exports all up to limit.
limit	Integer	No	Max notebooks to export. Default: 10. Set to 0 for unlimited.
includeSourceSummaries	Boolean	No	Fetch AI-generated summary and tags for each source. Default: false. Adds ~3s per source.
proxyConfiguration	Object	No	Proxy settings. Default: US residential proxy.

Proxy tip: If you're outside the US, change the proxy country to match your Google account location. This prevents Google from flagging the login as suspicious and asking for verification.

How to Get Project IDs

Run the Actor once to see all your notebooks. Each output item includes a projectId you can use for selective exports later.

---

Authentication Setup

This Actor uses Google App Passwords for secure authentication. This is NOT your regular Google password.

Step 1: Enable 2-Factor Authentication

App Passwords require 2FA to be enabled on your Google account:

1. Go to myaccount.google.com/security
2. Under "How you sign in to Google", click 2-Step Verification
3. Follow the prompts to enable 2FA if not already active

Step 2: Create an App Password

1. Go to myaccount.google.com/apppasswords
2. Click Select app → choose Other (Custom name)
3. Enter a name like NotebookLM Export or Apify
4. Click Generate
5. Copy the 16-character password (spaces are optional)

Step 3: Use in Actor Input

{
  "email": "your-email@gmail.com",
  "appPassword": "abcd efgh ijkl mnop"
}

App Passwords

This Actor authenticates using Google App Passwords, not your main Google password. App Passwords are:

• Scoped – They only grant access to specific services, not your full Google account
• Revocable – Delete them anytime from myaccount.google.com/apppasswords without changing your main password
• Encrypted – Stored as a secret input on Apify, never exposed in logs or metadata

I don't store your credentials or notebook content. Data streams directly to your Apify dataset.

---

Output

Each notebook is exported as a separate dataset item. All text content is formatted as markdown with bold highlights, bullet points, and structure - ready to feed directly into LLMs, RAG pipelines, or content workflows.

Output Fields

Field	Type	Description
projectId	String	Unique notebook UUID
projectTitle	String	Notebook name
projectEmoji	String	Notebook icon emoji
notebookSummary	String	AI-generated overview of the notebook contents
suggestedQuestions	Array	Suggested questions to explore the notebook
sources	Array	List of source documents
conversations	Array	All chat sessions with turns

Source Fields

Field	Type	Description
id	String	Source UUID
title	String	Source title
url	String	Original source URL
summary	String	AI-generated source summary (only with includeSourceSummaries: true)
tags	Array	Auto-generated topic tags (only with includeSourceSummaries: true)

Output Example

{
  "projectId": "6509c6d4-58fc-4066-8aa9-b691423e5ca7",
  "projectTitle": "Market Research Q1 2025",
  "projectEmoji": "📊",
  "notebookSummary": "These sources chronicle the rapid growth of **renewable energy markets** in 2025. The texts highlight a shift toward **distributed solar** and **battery storage**, with key players including Tesla, Enphase, and emerging Chinese manufacturers. Collectively, the documents provide a comprehensive look at **market sizing**, competitive dynamics, and regulatory tailwinds driving adoption.",
  "suggestedQuestions": [
    {
      "question": "What are the key market trends for 2025?",
      "promptTemplate": "Create a detailed briefing document..."
    }
  ],
  "sources": [
    {
      "id": "cb16c741-0698-4590-b2a0-94f5e1f3afd5",
      "title": "BloombergNEF Solar Market Outlook 2025",
      "url": "https://about.bnef.com/solar-market-outlook/",
      "summary": "This report projects **global solar installations** to reach 580 GW in 2025, driven by falling module prices and supportive policies in the US and EU. Key barriers include **grid interconnection delays** and polysilicon supply constraints.",
      "tags": ["Solar energy", "Market forecast", "Clean energy policy"]
    },
    {
      "id": "15969df9-c08e-4bcb-af2b-363a983b294b",
      "title": "Tesla Q3 2025 Earnings Call Transcript",
      "url": "https://ir.tesla.com/q3-2025-transcript",
      "summary": "Tesla's energy storage division reported **record deployments** of 12.4 GWh, up 180% YoY. Management highlighted **Megapack demand** outpacing production capacity through 2026.",
      "tags": ["Battery storage", "Tesla", "Energy earnings"]
    }
  ],
  "conversations": [
    {
      "sessionId": "079ffb57-e2f9-4e7a-bf5b-ed81d6a0aaab",
      "turns": [
        {
          "turnId": "bfd3f131-b306-4fa6-b90c-8aa91be0210e",
          "timestamp": 1765731768,
          "role": "user",
          "content": "What are the key market trends for 2025?",
          "citations": null
        },
        {
          "turnId": "c984b3b4-4ee3-4e76-a5b3-9d737c607a1d",
          "timestamp": 1765731770,
          "role": "assistant",
          "content": "Based on the sources, the key trends shaping the **renewable energy market** in 2025 include:\n\n*   **Solar dominance:** Global installations projected to hit 580 GW, with falling module prices driving adoption [1].\n*   **Storage surge:** Tesla's Megapack deployments up **180% YoY**, with demand outpacing supply through 2026 [2].\n*   **Grid bottlenecks:** Interconnection delays remain the primary barrier to faster deployment [1].\n\nThe overall trajectory suggests **accelerating electrification** despite supply chain challenges.",
          "citations": [
            {
              "sourceId": "cb16c741-0698-4590-b2a0-94f5e1f3afd5",
              "startPos": 95,
              "endPos": 180,
              "citedText": "Global installations projected to hit 580 GW",
              "sourceExcerpt": "BloombergNEF projects global solar PV installations to reach 580 GW in 2025, representing a 25% increase from 2024..."
            },
            {
              "sourceId": "15969df9-c08e-4bcb-af2b-363a983b294b",
              "startPos": 220,
              "endPos": 295,
              "citedText": "Tesla's Megapack deployments up **180% YoY**",
              "sourceExcerpt": "Our energy storage deployments reached 12.4 GWh this quarter, up 180% year-over-year, with Megapack demand continuing to exceed production capacity..."
            }
          ]
        }
      ]
    }
  ]
}

Note: summary and tags fields on sources only appear when includeSourceSummaries: true is set in the input.

Markdown output: All AI-generated content (notebookSummary, content, summary) uses markdown formatting with bold highlights, bullet points, and numbered references. This structured text is ready to pipe into LLMs, vector databases, or content workflows without preprocessing.

Citation Structure

Each AI response includes citation data showing exactly which sources informed the answer:

Field	Type	Description
sourceId	String	UUID of the cited source
startPos	Number	Character position where citation starts in response
endPos	Number	Character position where citation ends
citedText	String	The exact text that was cited
sourceExcerpt	String	Relevant excerpt from the original source

---

Key Features

Complete Data Extraction

• Notebook metadata - Title, emoji, and unique project ID
• Source library - All sources with titles and original URLs
• Full conversations - Every Q&A exchange in your notebooks
• Citation data - Which sources informed each AI response, with exact text excerpts

Flexible Export Options

• Export all notebooks - One-click backup of your entire NotebookLM library
• Selective export - Choose specific notebooks by project ID
• Multiple formats - JSON for programmatic use, CSV for spreadsheets

---

Use Cases

• Backup your research - Export before you lose access (graduating, switching jobs, etc.)
• Feed RAG pipelines - Use conversation history and citations as retrieval context
• Automate with n8n - Weekly digests, Notion sync, Slack alerts when topics appear
• Repurpose content - Turn NotebookLM conversations into blog posts with proper citations
• Fine-tune models - Export Q&A pairs for training data
• Audit AI responses - Review which sources informed each answer at scale

---

API Integration

Python

from apify_client import ApifyClient

client = ApifyClient("your_api_token")

run = client.actor("clearpath/notebooklm-api").call(
    run_input={
        "email": "your-email@gmail.com",
        "appPassword": "xxxx xxxx xxxx xxxx"
    }
)

# Process exported notebooks
for notebook in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(f"📓 {notebook['projectEmoji']} {notebook['projectTitle']}")
    print(f"   Sources: {len(notebook['sources'])}")
    print(f"   Conversations: {len(notebook['conversations'])}")

JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'your_api_token' });

const run = await client.actor('clearpath/notebooklm-api').call({
    email: 'your-email@gmail.com',
    appPassword: 'xxxx xxxx xxxx xxxx'
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();

for (const notebook of items) {
    console.log(`📓 ${notebook.projectEmoji} ${notebook.projectTitle}`);
    console.log(`   Sources: ${notebook.sources.length}`);
    console.log(`   Conversations: ${notebook.conversations.length}`);
}

---

n8n Integration Example

Combine NotebookLM Export with n8n to build powerful automated workflows:

Use Case: Weekly Research Digest

Create an automated pipeline that exports your NotebookLM research, processes it with an LLM, and sends a weekly summary email.

Workflow:

1. Schedule Trigger - Run every Monday at 9 AM
2. Apify Actor Node - Export all notebooks
3. Code Node - Extract key insights from conversations
4. OpenAI Node - Summarize findings into a digest
5. Gmail Node - Send the digest to your team

Apify Actor Node Configuration:

{
  "actorId": "clearpath/notebooklm-api",
  "input": {
    "email": "{{ $credentials.googleEmail }}",
    "appPassword": "{{ $credentials.googleAppPassword }}"
  }
}

Code Node - Extract Insights:

const notebooks = $input.all();
const insights = [];

for (const notebook of notebooks) {
  for (const conv of notebook.json.conversations) {
    for (const turn of conv.turns) {
      if (turn.role === 'assistant' && turn.citations?.length > 0) {
        insights.push({
          notebook: notebook.json.projectTitle,
          insight: turn.content.substring(0, 500),
          sourceCount: turn.citations.length
        });
      }
    }
  }
}

return insights.slice(0, 10); // Top 10 insights

OpenAI Node Prompt:

You are a research analyst. Summarize these key insights from our NotebookLM research into a concise weekly digest:

{{ $json.insights }}

Format as a professional email with:
- Executive summary (2-3 sentences)
- Key findings (bullet points)
- Sources referenced
- Recommended actions

Other n8n Use Cases

• Backup to Google Drive - Export → Format → Upload to Drive
• Slack Notifications - Alert when new insights are captured
• Notion Database - Sync notebooks to a Notion knowledge base
• RAG Pipeline - Feed exports into a vector database for semantic search
• Content Calendar - Extract topics → Schedule blog posts

---

Data Export Formats

Export your data in multiple formats from the Apify Console:

• JSON - Full structured data with nested objects
• CSV - Flattened for spreadsheet analysis
• Excel - Direct import to Microsoft Excel
• XML - For legacy system integration

---

FAQ

Is this official? Nope, just a side project. It uses your credentials to access your own notebooks - same as you would in a browser.

Is my data secure? Your App Password is encrypted and only used to authenticate. I don't store your notebook content - it streams straight to your dataset.

What's an App Password? A 16-character code that only gives access to NotebookLM, not your Gmail or Drive. You can revoke it anytime without changing your main password.

Why do I need 2FA? Google requires it for App Passwords. No way around it.

Can I export specific notebooks? Yes - pass the projectIds array. Run once without it to see all your notebook IDs.

Lots of notebooks? Handles large libraries fine. Built-in rate limiting and retries. Usually finishes in minutes.

Can I schedule exports? Yes, use Apify's scheduler for daily/weekly runs.

---

Roadmap

This Actor is under active development. Planned features:

• Create notebooks - Programmatically create new NotebookLM projects
• Add sources - Upload documents and URLs to notebooks via API
• Generate podcasts - Trigger Audio Overview generation
• Chat interface - Send queries to notebooks and get AI responses

Have a feature request? Open an issue or reach out!

---

Support

Email me at max@mapa.slmail.me or open an issue. I usually respond same-day.

---

Legal

This exports data from your own account using your credentials. You're responsible for complying with Google's ToS and any relevant data regulations (GDPR, etc.).