NotebookLM MCP Tools: Complete Reference Guide

The NotebookLM MCP server registers 39 tools that any MCP-compatible AI assistant can call on your behalf. This page documents every tool: what it does, the parameters it accepts, and a usage example. Tools are organized by category — use the accordions below to explore each group. For setup instructions, see the MCP Overview.

Several tools are unified — a single tool that handles multiple related operations via a source_type, artifact_type, or action parameter. Prefer these over chaining multiple single-purpose calls; they save context and reduce round-trips.

Notebooks

notebook_list — List all notebooks

List all NotebookLM notebooks in your account.

max_results

integer

default:"100"

Maximum number of notebooks to return.

notebook_list()
notebook_list(max_results=50)

notebook_create — Create a notebook

Create a new, empty notebook.

title

string

Optional display title for the new notebook.

notebook_create(title="AI Strategy Research")

notebook_get — Get notebook details and sources

Retrieve full details for a notebook, including its ID, title, URL, source count, and a list of all sources.

notebook_id

string

required

Notebook UUID.

notebook_get(notebook_id="abc123")

notebook_describe — AI summary and suggested topics

Ask NotebookLM to generate an AI-written summary of the notebook’s contents and return a list of suggested discussion topics.

notebook_id

string

required

Notebook UUID.

Returns summary (markdown) and suggested_topics (list of strings).

notebook_describe(notebook_id="abc123")

notebook_rename — Rename a notebook

Change the display title of an existing notebook.

notebook_id

string

required

Notebook UUID.

new_title

string

required

New display title.

notebook_rename(notebook_id="abc123", new_title="Q3 Competitive Research")

notebook_delete — Delete a notebook (irreversible)

Permanently delete a notebook and all its contents. This action cannot be undone. The AI must ask the user to confirm before setting confirm=True.

notebook_id

string

required

Notebook UUID.

confirm

boolean

default:"false"

Must be True after explicit user approval. The tool returns an error if this is False.

notebook_delete(notebook_id="abc123", confirm=True)

Sources

source_add — Add a source (unified)

Add any type of source to a notebook. This single tool replaces separate URL, text, file, and Drive add operations.

notebook_id

string

required

Notebook UUID.

source_type

string

required

Type of source: url, text, file, or drive.

url

string

Web page or YouTube URL. Used when source_type=url.

urls

array

List of URLs to add in bulk. Alternative to url for batch URL imports.

text

string

Raw text content to paste as a source. Used when source_type=text.

title

string

Optional display title (primarily for text sources).

file_path

string

Absolute path to a local file. Used when source_type=file. Supported formats: PDF, TXT, MD, DOCX, CSV, EPUB, MP3, M4A, WAV, AAC, OGG, OPUS, MP4, JPG, JPEG, PNG, GIF, WEBP.

document_id

string

Google Drive document ID. Used when source_type=drive.

doc_type

string

default:"doc"

Drive document type: doc, slides, sheets, or pdf. Only used when source_type=drive.

wait

boolean

default:"false"

If True, block until the source finishes processing before returning.

wait_timeout

number

default:"120.0"

Maximum seconds to wait when wait=True.

# Add a URL
source_add(notebook_id="abc123", source_type="url", url="https://example.com/article")

# Add multiple URLs at once
source_add(notebook_id="abc123", source_type="url", urls=["https://a.com", "https://b.com"])

# Add a URL and wait for processing to complete
source_add(notebook_id="abc123", source_type="url", url="https://example.com", wait=True)

# Add pasted text
source_add(notebook_id="abc123", source_type="text", text="Key findings from the meeting...", title="Meeting Notes")

# Upload a local PDF
source_add(notebook_id="abc123", source_type="file", file_path="/home/user/report.pdf", wait=True)

# Add a Google Drive document
source_add(notebook_id="abc123", source_type="drive", document_id="1BxiMVs0XRA5nFMdKvBdBZjgmUUqptlbs74OgVE2upms", doc_type="doc")

source_list_drive — List sources with Drive freshness

List all sources in a notebook and their Drive sync status (fresh or stale). Call this before source_sync_drive to identify which sources need updating.

notebook_id

string

required

Notebook UUID.

source_list_drive(notebook_id="abc123")

source_sync_drive — Sync stale Drive sources

Re-sync one or more Google Drive sources so they reflect the latest document content. Call source_list_drive first to get the IDs of stale sources.

source_ids

array

required

List of source UUIDs to sync.

confirm

boolean

default:"false"

Must be True after user approval.

source_sync_drive(source_ids=["src_uuid_1", "src_uuid_2"], confirm=True)

source_rename — Rename a source

Change the display title of an existing source in a notebook.

notebook_id

string

required

Notebook UUID containing the source.

source_id

string

required

Source UUID to rename.

new_title

string

required

New display title for the source.

source_rename(notebook_id="abc123", source_id="src_uuid_1", new_title="Q3 Earnings Report")

source_delete — Delete a source (irreversible)

Permanently delete one or more sources. This action cannot be undone.

source_id

string

Single source UUID to delete.

source_ids

array

List of source UUIDs for bulk deletion. Alternative to source_id.

confirm

boolean

default:"false"

Must be True after explicit user approval.

source_delete(source_id="src_uuid_1", confirm=True)
source_delete(source_ids=["src_uuid_1", "src_uuid_2"], confirm=True)

source_describe — AI summary and keywords

Get an AI-generated summary of a source, including extracted keyword chips. Returns summary (markdown with bold keywords) and keywords (list).

source_id

string

required

Source UUID.

source_describe(source_id="src_uuid_1")

source_get_content — Raw text content

Retrieve the raw indexed text of a source without any AI processing. Much faster than notebook_query for content export. Works for PDFs, web pages, pasted text, and YouTube transcripts.Returns content, title, source_type, and char_count.

source_id

string

required

Source UUID.

source_get_content(source_id="src_uuid_1")

Querying

notebook_query — Ask AI about notebook sources

Ask the NotebookLM AI a question about the sources already in a notebook. Responses are grounded in the notebook’s sources and include citations. Note: This tool is for querying existing content — use research_start to find new sources on the web or in Drive.All queries sent through this tool automatically persist in your NotebookLM web UI chat history.

notebook_id

string

required

Notebook UUID.

query

string

required

The question to ask.

source_ids

array

Restrict the query to specific sources. Defaults to all sources.

conversation_id

string

Pass the conversation_id from a previous response to ask follow-up questions in the same thread.

timeout

number

default:"120.0"

Request timeout in seconds. Overrides NOTEBOOKLM_QUERY_TIMEOUT env var.

# Basic query
notebook_query(notebook_id="abc123", query="What are the three most important takeaways?")

# Follow-up in the same conversation
notebook_query(notebook_id="abc123", query="Can you expand on point two?", conversation_id="conv_xyz")

# Query specific sources only
notebook_query(notebook_id="abc123", query="What does this source say about security?", source_ids=["src_1", "src_2"])

chat_configure — Configure chat behavior

Set the AI chat goal and response length for a notebook. Changes affect all future notebook_query calls on that notebook.

notebook_id

string

required

Notebook UUID.

goal

string

default:"default"

Chat goal preset: default, learning_guide, or custom.

custom_prompt

string

Custom system instructions. Required when goal=custom. Maximum 10,000 characters.

response_length

string

default:"default"

Desired response length: default, longer, or shorter.

chat_configure(notebook_id="abc123", goal="learning_guide", response_length="longer")
chat_configure(notebook_id="abc123", goal="custom", custom_prompt="Always respond in bullet points.")

notebook_query_start — Start an async query

Start a notebook query asynchronously. Use this instead of notebook_query when querying notebooks with many sources (50+) where the response may exceed 60 seconds. Returns immediately with a query_id. Poll notebook_query_status until the result is ready.

notebook_id

string

required

Notebook UUID.

query

string

required

The question to ask.

source_ids

array

Restrict the query to specific sources. Defaults to all sources.

conversation_id

string

Pass the conversation_id from a previous response to ask follow-up questions in the same thread.

timeout

number

default:"120.0"

Request timeout in seconds. Overrides NOTEBOOKLM_QUERY_TIMEOUT env var.

result = notebook_query_start(notebook_id="abc123", query="Summarize all findings")
# result["query_id"] → poll with notebook_query_status

notebook_query_status — Check async query result

Check the status of an async query started with notebook_query_start. Poll every few seconds until status is completed or error.

query_id

string

required

The query ID returned by notebook_query_start.

notebook_query_status(query_id="qry_abc123")

Studio

studio_create — Create any artifact (unified)

Create any type of NotebookLM studio artifact. This is the unified creation tool that replaces nine separate create operations. Requires user confirmation before proceeding.

notebook_id

string

required

Notebook UUID.

artifact_type

string

required

Type of artifact to create. One of: audio, video, report, quiz, flashcards, mind_map, slide_deck, infographic, data_table.

confirm

boolean

default:"false"

Must be True after user approval. Without confirmation the tool returns a preview of the settings.

source_ids

array

Source IDs to include. Defaults to all sources in the notebook.

Audio options (artifact_type=audio):

audio_format

string

default:"deep_dive"

deep_dive, brief, critique, or debate.

audio_length

string

default:"default"

short, default, or long.

Video options (artifact_type=video):

video_format

string

default:"explainer"

explainer, brief, or cinematic.

visual_style

string

default:"auto_select"

auto_select, custom, classic, whiteboard, kawaii, anime, watercolor, retro_print, heritage, or paper_craft.

video_style_prompt

string

Custom style description when visual_style=custom.

Infographic options (artifact_type=infographic):

orientation

string

default:"landscape"

landscape, portrait, or square.

detail_level

string

default:"standard"

concise, standard, or detailed.

infographic_style

string

default:"auto_select"

auto_select, sketch_note, professional, bento_grid, editorial, instructional, bricks, clay, anime, kawaii, or scientific.

Slide deck options (artifact_type=slide_deck):

slide_format

string

default:"detailed_deck"

detailed_deck or presenter_slides.

slide_length

string

default:"default"

short or default.

Report options (artifact_type=report):

report_format

string

default:"Briefing Doc"

Briefing Doc, Study Guide, Blog Post, or Create Your Own.

custom_prompt

string

Custom instructions. Used when report_format=Create Your Own.

Quiz options (artifact_type=quiz):

question_count

integer

default:"2"

Number of questions to generate.

Shared options (most artifact types):

difficulty

string

default:"medium"

For flashcards and quiz: easy, medium, or hard.

language

string

BCP-47 language code (e.g. en, es, fr, de, ja). Defaults to NOTEBOOKLM_HL env var or en.

focus_prompt

string

Optional free-text focus instruction applied to all artifact types.

# Audio podcast (deep dive format)
studio_create(notebook_id="abc123", artifact_type="audio", audio_format="deep_dive", confirm=True)

# Video with cinematic style
studio_create(notebook_id="abc123", artifact_type="video", video_format="cinematic", visual_style="watercolor", confirm=True)

# Briefing doc report
studio_create(notebook_id="abc123", artifact_type="report", report_format="Briefing Doc", confirm=True)

# Study quiz with 10 questions
studio_create(notebook_id="abc123", artifact_type="quiz", question_count=10, difficulty="hard", confirm=True)

# Flashcards
studio_create(notebook_id="abc123", artifact_type="flashcards", difficulty="medium", confirm=True)

# Professional infographic
studio_create(notebook_id="abc123", artifact_type="infographic", orientation="landscape", infographic_style="professional", confirm=True)

# Slide deck
studio_create(notebook_id="abc123", artifact_type="slide_deck", slide_format="detailed_deck", confirm=True)

After calling studio_create, poll studio_status until the artifact’s status shows completed. Generation typically takes 1–5 minutes depending on artifact type.

studio_status — Check generation progress

Check the status of all studio artifacts in a notebook. Also supports renaming an artifact or listing all supported artifact types.

notebook_id

string

required

Notebook UUID.

action

string

default:"status"

status (default), rename, or list_types.

artifact_id

string

Required when action=rename.

new_title

string

Required when action=rename.

Returns artifacts list (each with artifact_id, title, type, status, and url) and a summary with counts.

# Poll until all artifacts are completed
studio_status(notebook_id="abc123")

# Rename an artifact
studio_status(notebook_id="abc123", action="rename", artifact_id="art_xyz", new_title="Weekly Podcast Ep 12")

# List all artifact types and their options
studio_status(notebook_id="abc123", action="list_types")

studio_delete — Delete an artifact (irreversible)

Permanently delete a studio artifact. Call studio_status first to list artifact IDs.

notebook_id

string

required

Notebook UUID.

artifact_id

string

required

Artifact UUID (from studio_status).

confirm

boolean

default:"false"

Must be True after explicit user approval.

studio_delete(notebook_id="abc123", artifact_id="art_xyz", confirm=True)

studio_revise — Revise slides in a slide deck

Apply targeted revisions to individual slides in an existing slide deck. Creates a new artifact — the original is not modified. Only slide decks support revision. Poll studio_status after calling to wait for the new deck.

notebook_id

string

required

Notebook UUID.

artifact_id

string

required

UUID of the existing slide deck to revise (from studio_status).

slide_instructions

array

required

List of revision instructions. Each item must have a slide (1-based slide number) and instruction (description of the change).

confirm

boolean

default:"false"

Must be True after user approval.

studio_revise(
    notebook_id="abc123",
    artifact_id="art_xyz",
    slide_instructions=[
        {"slide": 1, "instruction": "Make the title larger and bolder"},
        {"slide": 3, "instruction": "Replace the bullet list with a comparison table"}
    ],
    confirm=True
)

Downloads

download_artifact — Download any artifact (unified)

Download a completed studio artifact to a local file. This unified tool replaces nine separate download commands. Ensure studio_status shows completed before downloading.

notebook_id

string

required

Notebook UUID.

artifact_type

string

required

Type of artifact: audio, video, report, mind_map, slide_deck, infographic, data_table, quiz, or flashcards.

output_path

string

required

Local file path where the artifact should be saved (e.g. podcast.mp3, slides.pdf).

artifact_id

string

Specific artifact UUID. Uses the most recent completed artifact of artifact_type if not provided.

output_format

string

default:"json"

For quiz and flashcards only: json, markdown, or html.

slide_deck_format

string

default:"pdf"

For slide_deck only: pdf or pptx.

# Download audio podcast
download_artifact(notebook_id="abc123", artifact_type="audio", output_path="podcast.mp3")

# Download slide deck as PowerPoint
download_artifact(notebook_id="abc123", artifact_type="slide_deck", output_path="slides.pptx", slide_deck_format="pptx")

# Download quiz as HTML
download_artifact(notebook_id="abc123", artifact_type="quiz", output_path="quiz.html", output_format="html")

# Download a specific artifact by ID
download_artifact(notebook_id="abc123", artifact_type="video", artifact_id="art_xyz", output_path="overview.mp4")

Exports

export_artifact — Export to Google Docs or Sheets

Export a completed studio artifact to Google Docs or Google Sheets. Useful for reports, data tables, and other text-based artifacts. Returns the URL to the created Google Doc or Sheet.

notebook_id

string

required

Notebook UUID.

artifact_id

string

required

Artifact UUID (from studio_status).

export_type

string

required

Destination format: docs (Google Docs) or sheets (Google Sheets). Data tables export to sheets; reports export to docs.

title

string

Optional title for the exported document.

# Export a report to Google Docs
export_artifact(notebook_id="abc123", artifact_id="art_xyz", export_type="docs", title="My Report")

# Export a data table to Google Sheets
export_artifact(notebook_id="abc123", artifact_id="art_xyz", export_type="sheets")

Research

research_start — Start web or Drive research

Launch a research task that searches the web or your Google Drive for sources relevant to a query. Returns immediately with a task_id. Poll research_status to wait for results, then call research_import to add the discovered sources to the notebook.

query

string

required

What to search for (e.g. "quantum computing breakthroughs 2025").

source

string

default:"web"

Where to search: web or drive.

mode

string

default:"fast"

Research depth: fast (~30 seconds, ~10 sources) or deep (~5 minutes, ~40 sources, web only).

notebook_id

string

Existing notebook to attach results to. Creates a new notebook if not provided (requires title).

title

string

Title for the new notebook to create. Used when notebook_id is not provided.

# Start deep web research
research_start(query="enterprise AI ROI metrics", mode="deep", notebook_id="abc123")

# Fast research with auto-created notebook
research_start(query="cloud marketplace trends", mode="fast", title="Cloud Research")

research_status — Poll research progress

Block until a research task completes or the timeout is reached. Returns a summary report and list of discovered sources.

notebook_id

string

required

Notebook UUID.

max_wait

integer

default:"300"

Maximum seconds to wait. Set to 0 for a single non-blocking poll.

poll_interval

integer

default:"30"

Seconds between status checks.

task_id

string

Specific task ID to poll (from research_start).

query

string

Optional query string for fallback task matching when task_id is unavailable (deep research).

compact

boolean

default:"true"

Truncate the report and limit sources shown to save tokens. Set to False for full details.

research_status(notebook_id="abc123", max_wait=300)
research_status(notebook_id="abc123", task_id="task_xyz", max_wait=600, compact=False)

research_import — Import discovered sources

Import sources discovered by a completed research task into the notebook. Call after research_status returns status="completed".

notebook_id

string

required

Notebook UUID.

task_id

string

required

Research task ID (from research_start or research_status).

cited_only

boolean

default:"false"

If True, import only the sources that were actually cited in the research report. Overrides source_indices.

source_indices

array

Specific source indices to import. Defaults to all discovered sources. Ignored when cited_only=True.

timeout

number

default:"300.0"

Import timeout in seconds. Increase for large notebooks.

# Import only cited sources (recommended for quality filtering)
research_import(notebook_id="abc123", task_id="task_xyz", cited_only=True)

# Import all discovered sources
research_import(notebook_id="abc123", task_id="task_xyz", timeout=600)

# Import specific sources by index
research_import(notebook_id="abc123", task_id="task_xyz", source_indices=[0, 1, 2, 5])

Notes

note — Manage notes (unified)

Create, list, update, and delete notes in a notebook. All note operations go through this single unified tool.

notebook_id

string

required

Notebook UUID.

action

string

required

Operation to perform: create, list, update, or delete.

note_id

string

Note UUID. Required for update and delete.

content

string

Note body text. Required for create, optional for update.

title

string

Note display title. Optional for create and update.

confirm

boolean

default:"false"

Must be True for action=delete.

note(notebook_id="abc123", action="list")
note(notebook_id="abc123", action="create", content="Key insight from today's meeting.", title="Meeting Insight")
note(notebook_id="abc123", action="update", note_id="note_xyz", content="Updated insight text.")
note(notebook_id="abc123", action="delete", note_id="note_xyz", confirm=True)

notebook_share_status — Get sharing settings

Retrieve the current sharing configuration for a notebook, including whether it is public, the access level, the public link (if any), and the list of collaborators.

notebook_id

string

required

Notebook UUID.

notebook_share_status(notebook_id="abc123")

notebook_share_public — Enable or disable public link

Toggle public link access for a notebook. When enabled, anyone with the link can view the notebook.

notebook_id

string

required

Notebook UUID.

is_public

boolean

default:"true"

True to enable a public link, False to disable it.

notebook_share_public(notebook_id="abc123", is_public=True)   # Enable
notebook_share_public(notebook_id="abc123", is_public=False)  # Disable

notebook_share_invite — Invite a collaborator

Invite someone to collaborate on a notebook by email address.

notebook_id

string

required

Notebook UUID.

string

required

Email address of the person to invite.

role

string

default:"viewer"

Collaboration role: viewer or editor.

notebook_share_invite(notebook_id="abc123", email="colleague@example.com", role="editor")

notebook_share_batch — Invite multiple collaborators

Invite multiple collaborators to a notebook in a single request. Requires user confirmation before proceeding.

notebook_id

string

required

Notebook UUID.

recipients

array

required

List of recipient objects, each with an email key (string) and an optional role key (viewer or editor, defaults to viewer).

confirm

boolean

default:"false"

Must be True after explicit user approval.

notebook_share_batch(
    notebook_id="abc123",
    recipients=[
        {"email": "alice@example.com", "role": "editor"},
        {"email": "bob@example.com", "role": "viewer"}
    ],
    confirm=True
)

Auth

refresh_auth — Reload auth tokens

Reload authentication tokens from disk. Call this after running nlm login in a terminal so the MCP server picks up the fresh cookies without a restart. Also attempts headless browser re-authentication if a saved Chrome profile exists.

refresh_auth()

refresh_auth() reloads tokens from disk but cannot revive expired credentials. If your Google session has fully expired, run nlm login in a terminal first.

save_auth_tokens — Save cookies manually (fallback)

Save NotebookLM cookies directly from a browser cookie header. This is a fallback method — prefer nlm login (automated CLI) first.

string

required

Cookie header string copied from Chrome DevTools Network tab.

request_body

string

Optional raw POST body from a NotebookLM network request. Used to auto-extract the CSRF token.

request_url

string

Optional request URL from a NotebookLM network request. Used to auto-extract the session ID and build label.

save_auth_tokens(
    cookies="SID=xxx; HSID=xxx; SSID=xxx; APISID=xxx; SAPISID=xxx; ...",
    request_body="at=ABC123&...",
    request_url="https://notebooklm.google.com/...?f.sid=XYZ&bl=ABC..."
)

Server

server_info — Version and auth status

Return the currently installed version, the latest version on PyPI, whether an update is available, and the current authentication status. The AI assistant will suggest updating if update_available is True.

server_info()

Returns:

version — installed version string
latest_version — latest PyPI version (or null if check failed)
update_available — true if a newer version exists
auth_status — configured, stale, unverified, not_configured, or error
update_command — uv tool upgrade notebooklm-mcp-cli

Batch & Cross-Notebook

batch — Batch operations across notebooks

Perform the same operation across multiple notebooks at once. Target notebooks by name, by tag, or by using all=True for all notebooks.

action

string

required

Operation to perform: query, add_source, create, delete, or studio.

notebook_names

string

Comma-separated notebook names or IDs to target.

Pipelines

pipeline — List or run multi-step workflows

Run automated multi-step workflows that chain multiple tools together. Built-in pipelines cover common end-to-end tasks.

action

string

required

list to see available pipelines, or run to execute one.

notebook_id

string

Target notebook UUID. Required for action=run.

pipeline_name

string

Name of the pipeline to run. Required for action=run.

input_url

string

Input URL for the pipeline (used by ingest-and-podcast and research-and-report).

Built-in pipelines:

ingest-and-podcast — Add a URL source and generate an audio podcast
research-and-report — Run deep research and generate a briefing doc
multi-format — Generate audio, slide deck, and briefing doc in one pass

# List available pipelines
pipeline(action="list")

# Run ingest-and-podcast end-to-end
pipeline(action="run", notebook_id="abc123", pipeline_name="ingest-and-podcast", input_url="https://example.com/article")

# Run research-and-report
pipeline(action="run", notebook_id="abc123", pipeline_name="research-and-report", input_url="https://example.com")

Labels

label — Source label management (unified)

Organize sources in a notebook into thematic categories using AI-generated or manually managed labels. Requires at least 5 sources for auto-labeling. Sources can belong to multiple labels simultaneously.

notebook_id

string

required

Notebook UUID.

action

string

required

Operation to perform: auto, list, reorganize, create, rename, set_emoji, move_source, or delete.

label_id

string

Label UUID. Required for rename, set_emoji, move_source, and delete.

label_ids

array

List of label UUIDs for batch delete. Alternative to label_id for action=delete.

name

string

Label display name. Required for create and rename.

emoji

string

Emoji character to set on a label (e.g. "📊"). Use "" to clear. Required for set_emoji.

source_id

string

Source UUID to assign to a label. Required for move_source.

unlabeled_only

boolean

default:"false"

For reorganize: if True, only label sources not yet assigned to any label (no confirmation needed). If False, replaces all existing labels from scratch (requires confirm=True).

confirm

boolean

default:"false"

Required for delete and for reorganize with unlabeled_only=False.

# AI auto-label all sources
label(notebook_id="abc123", action="auto")

# List current labels
label(notebook_id="abc123", action="list")

# Re-categorize everything from scratch
label(notebook_id="abc123", action="reorganize", confirm=True)

# Only label sources not yet in any category
label(notebook_id="abc123", action="reorganize", unlabeled_only=True)

# Create a new label with an emoji
label(notebook_id="abc123", action="create", name="Research", emoji="📚")

# Rename a label
label(notebook_id="abc123", action="rename", label_id="lbl_xyz", name="Core Research")

# Set an emoji on a label
label(notebook_id="abc123", action="set_emoji", label_id="lbl_xyz", emoji="🎯")

# Assign a source to a label
label(notebook_id="abc123", action="move_source", label_id="lbl_xyz", source_id="src_abc")

# Delete a label (sources are preserved)
label(notebook_id="abc123", action="delete", label_id="lbl_xyz", confirm=True)

action=delete permanently removes the label. Sources assigned to the label are not deleted — they remain in the notebook.

Tags

tag — Tag management and smart select

Add tags to notebooks for organization, list tagged notebooks, and use smart select to find notebooks relevant to a query by tag matching.

action

string

required

Operation: add, remove, list, or select.

notebook_id

string

Notebook UUID. Required for add and remove.

Get Started

CLI Reference

MCP Server

Guides

NotebookLM MCP Tools: Complete Reference Guide

Notebooks

Sources

Querying

Studio

Downloads

Exports

Research

Notes

Auth

Server

Batch & Cross-Notebook

Pipelines

Labels

Tags

Build docs developers (and LLMs) love

Get Started

CLI Reference

MCP Server

Guides

Documentation Index

​Notebooks

​Sources

​Querying

​Studio

​Downloads

​Exports

​Research

​Notes

​Sharing

​Auth

​Server

​Batch & Cross-Notebook

​Pipelines

​Labels

​Tags

Build docs developers (and LLMs) love

Notebooks

Sources

Querying

Studio

Downloads

Exports

Research

Notes

Sharing

Auth

Server

Batch & Cross-Notebook

Pipelines

Labels

Tags