NotebookLM integration patterns for external RAG, research synthesis, studio content generation (audio, cinematic video, slides, infographics, mind maps), and knowledge management. Use when creating notebooks, adding sources, generating audio/video, or querying NotebookLM via MCP.
Referencemedium
Auto-activated — this skill loads automatically when Claude detects matching context.
NotebookLM = external RAG engine that offloads reading from your context window. Uses the notebooklm-mcp-cli MCP server (PyPI, v0.5.25+) to create notebooks, manage sources, generate content, and query with grounded AI responses. Supports batch operations across notebooks, pipelines, and multilingual content generation.
Disclaimer: Uses internal undocumented Google APIs via browser authentication. Sessions last ~20 minutes. API may change without notice.
BEFORE doing ANYTHING else, create tasks to track progress:
# 1. Create main task IMMEDIATELYTaskCreate( subject="NotebookLM: {operation}", description="Managing notebooks, sources, and content generation", activeForm="Managing NotebookLM resources")# 2. Create subtasks for the notebook workflowTaskCreate(subject="Notebook setup", activeForm="Creating/configuring notebook")TaskCreate(subject="Source management", activeForm="Adding sources to notebook")TaskCreate(subject="Content generation", activeForm="Generating studio content")# 3. Set dependencies for sequential stepsTaskUpdate(taskId="3", addBlockedBy=["2"])TaskUpdate(taskId="4", addBlockedBy=["3"])# 4. Before starting each task, verify it's unblockedtask = TaskGet(taskId="2") # Verify blockedBy is empty# 5. Update status as you progressTaskUpdate(taskId="2", status="in_progress") # When startingTaskUpdate(taskId="2", status="completed") # When done
# 1. Create a notebook for your projectnotebook_create(title="Auth Refactor Research")# 2. Add sources (docs, articles, existing code analysis)source_add(notebook_id="...", type="url", url="https://oauth.net/2.1/")source_add(notebook_id="...", type="text", content="Our current auth uses...")source_add(notebook_id="...", type="file", path="/docs/auth-design.md")# 3. Query with grounded AI responsesnotebook_query(notebook_id="...", query="What are the key differences between OAuth 2.0 and 2.1?")# 4. Generate a deep dive audio overview (supports language param)studio_create(notebook_id="...", artifact_type="audio", audio_format="deep_dive", language="he", confirm=True)studio_status(notebook_id="...") # Poll until complete# 5. Generate a cinematic video overview (Plus/Ultra, English)studio_create(notebook_id="...", artifact_type="video", video_format="cinematic", visual_style="classic", confirm=True)studio_status(notebook_id="...") # Poll — takes 3-8 minutes# 6. Capture insights as notesnote(notebook_id="...", action="create", content="Key takeaway: PKCE is mandatory in 2.1")
Forgetting auth expiry — Sessions last ~20 min. Always check with nlm login --check before long workflows. Re-auth with nlm login.
One giant notebook — Split by project/topic. One notebook with 50 sources degrades retrieval quality.
Huge single sources — Split documents >50K characters into logical sections for better chunking and retrieval.
Not polling studio_status — Studio content generation takes 2-5 minutes. Poll studio_status instead of assuming instant results.
Ignoring source types — Use type=url for web pages (auto-extracts), type=file for local files. Using type=text for a URL gives you the URL string, not the page content.
Deleting notebooks without checking — notebook_delete is irreversible. List contents with source_list_drive and note(action=list) first.
Skipping research_import — research_start discovers content but does not add it. Use research_import to actually add findings as sources.
Raw queries on empty notebooks — notebook_query returns poor results with no sources. Add sources before querying.
Ignoring language param — studio_create supports BCP-47 language codes (e.g., he, ar, ja). Defaults to English if omitted.
Batch without purpose — batch and cross_notebook_query are powerful but add latency. Use for multi-project synthesis, not single-notebook tasks.
Authenticate with Google via nlm login, then register the MCP server with Claude Code. Auth sessions expire after ~20 minutes of inactivity. Supports any Chromium-based browser (Chrome, Arc, Brave, Edge, Vivaldi, Opera).
Incorrect -- manually editing .mcp.json with wrong server path:
# 1. Authenticate (opens browser for Google OAuth)nlm login# 2. Register MCP server with Claude Codenlm setup add claude-code# 3. Verify auth is activenlm login --check
Manual .mcp.json fallback (if nlm setup is unavailable):
# Set preferred browser (auto-detected by default)nlm config set auth.browser arc # arc | chrome | brave | edge | vivaldi | opera | auto# Or via environment variableexport NLM_BROWSER=arc
Key rules:
Always authenticate with nlm login before first use -- browser OAuth flow required
Auth sessions last ~20 minutes; re-run nlm login if tools start failing
Use nlm login --check to verify session status before long workflows
Prefer nlm setup add claude-code over manual .mcp.json editing
If setup command fails, use the manual .mcp.json fallback with "command": "nlm"
Any Chromium-based browser works: Chrome, Arc, Brave, Edge, Chromium, Vivaldi, Opera
Set a preferred browser with nlm config set auth.browser <name> or NLM_BROWSER env var
v0.5.25 batch operations across multiple notebooks, cross-notebook queries with aggregated answers, and multi-step pipelines. Use for multi-project synthesis — not single-notebook tasks.
Incorrect — querying notebooks one at a time for synthesis:
# Slow: sequential queries across 5 notebooksfor nb_id in notebook_ids: result = notebook_query(notebook_id=nb_id, query="What are the security risks?") results.append(result)# Manual aggregation required
Correct — cross-notebook query for aggregated answers:
# Single call: queries all notebooks, returns aggregated answer with per-notebook citationsresult = cross_notebook_query( notebook_ids=["nb1", "nb2", "nb3"], query="What are the security risks across all projects?")# result includes: aggregated_answer, per_notebook_citations[]
Batch operations — multi-notebook bulk actions:
# Add the same source to multiple notebooks at oncebatch( action="add-source", notebook_ids=["nb1", "nb2", "nb3"], source_type="url", url="https://owasp.org/Top10/")# Create studio content across notebooksbatch( action="studio", notebook_ids=["nb1", "nb2"], artifact_type="audio", audio_format="brief", language="he", confirm=True)
Pipelines — multi-step workflows in a single call:
# Ingest sources and immediately generate a podcastpipeline( action="run", pipeline_name="ingest-and-podcast", notebook_id="...", sources=[{"type": "url", "url": "https://example.com/article"}], audio_format="deep_dive", language="en", confirm=True)# Research a topic, add findings, generate a reportpipeline( action="run", pipeline_name="research-and-report", notebook_id="...", query="Latest trends in LLM safety", report_format="Briefing Doc", confirm=True)# Generate multiple artifact types from the same sourcespipeline( action="run", pipeline_name="multi-format", notebook_id="...", artifact_types=["audio", "infographic", "mind_map"], language="he", confirm=True)
Tagging notebooks for organization:
# Tag notebooks for smart filteringtag(notebook_id="...", action="add", tags="security,q1-2026")# Use tags to select notebooks for batch operationsbatch( action="query", tag_filter="security", query="What vulnerabilities were identified?")
Key rules:
Use cross_notebook_query for synthesis across projects — avoids sequential query overhead
batch operations run in parallel server-side — faster than sequential MCP calls
Pipelines combine ingest + generation — use when source addition and content creation are a single intent
Tag notebooks early — enables tag-based batch operations later
All batch/pipeline operations support the language parameter for multilingual output
confirm=True is required on destructive or generative batch actions
Build dedicated notebooks as curated knowledge bases for debugging, security, and onboarding. Add incident reports, advisories, and runbooks as sources for grounded, verified answers.
Incorrect -- re-investigating a known issue from scratch:
User: "Production is throwing OOM errors again"Claude: "Let me research possible causes..."# Wastes time if this was already diagnosed and documented
Correct -- query the debugging knowledge base:
# 1. Create dedicated KB notebooksnotebook_create(name="Debugging KB")notebook_create(name="Security Handbook")# 2. Add incident reports and advisories as sourcessource_add(notebook_id="debug_kb", content="INC-042: OOM caused by unbounded cache. Fix: add TTL...")source_add(notebook_id="security_kb", content="SEC-007: SQL injection in search endpoint. Fix: parameterize...")# 3. Query for grounded answersnotebook_query(notebook_id="debug_kb", query="What causes OOM errors and how were they fixed?")notebook_query(notebook_id="security_kb", query="Known SQL injection patterns in our codebase")
Key rules:
Create separate notebooks for debugging, security, and onboarding domains
Add incident reports, post-mortems, and security advisories as sources
Query KB notebooks before re-investigating known issues
Keep sources current -- add new incidents as they are resolved
Use for onboarding: new team members query the KB instead of asking around
Use the research API for automated web and Google Drive discovery. The flow is async: start a research task, poll for status, then import discovered sources into your notebook.
Incorrect -- manually searching and adding URLs one by one:
# Tedious and misses relevant contentsource_add(notebook_id="...", url="https://example.com/article1")source_add(notebook_id="...", url="https://example.com/article2")source_add(notebook_id="...", url="https://example.com/article3")# Missed 20 other relevant articles
Correct -- automated research flow:
# 1. Start research (searches web and/or Google Drive)task = research_start( notebook_id="...", topic="Latest developments in WebAssembly component model", sources=["web", "drive"])# 2. Poll for completion (uses Google API quota)status = research_status(task_id=task.id)# status: "searching" | "analyzing" | "completed"# 3. Import discovered sources into notebook (timeout configurable, default 300s)research_import(task_id=task.id, notebook_id="...")# Adds the most relevant discovered sources automatically
Deep research mode (v0.5.1+):
# Deep research for comprehensive multi-source synthesisresearch_start(notebook_id="...", topic="...", mode="deep")# Transient API errors now auto-retry with RPCError handling
Key rules:
Use research_start for broad topic discovery instead of manual URL hunting
Always poll with research_status -- research takes 1-3 minutes
Research uses Google API quota -- avoid running many parallel research tasks
Import results with research_import to add discovered sources to your notebook
For notebooks with many sources, increase timeout on research_import (default 300s, was 120s)
Use mode="deep" for comprehensive synthesis -- transient errors are now auto-retried (v0.5.1+)
Combine web and Drive sources for comprehensive coverage
Follow up with notebook_query to synthesize the newly imported sources
Add large documents, codebases, and references as notebook sources instead of pasting them into chat. Use notebook_query for targeted synthesis without consuming context window.
Incorrect -- pasting large content directly into chat:
User: "Here's our entire codebase (100K chars)... now explain the auth flow"# Wastes context, may hit token limits, loses nuance in truncation
Correct -- add as source, query for synthesis:
# 1. Add large docs as sources (use RepoMix for codebases)source_add(notebook_id="...", url="https://docs.example.com/api-reference")source_add(notebook_id="...", content=repomix_output)# 2. Query for specific synthesisnotebook_query(notebook_id="...", query="How does the authentication middleware chain work?")# 3. Follow up with targeted questionsnotebook_query(notebook_id="...", query="What error codes does the auth endpoint return?")
Key rules:
Add large documents as sources rather than pasting into chat context
Use RepoMix to bundle codebases into a single source for onboarding
Query the notebook for synthesis -- NotebookLM reads the full source each time
Multiple targeted queries are cheaper than one massive context load
Combine with second-brain pattern to build persistent project knowledge
Create a dedicated notebook per project to capture decisions, design docs, and insights. Query the notebook for grounded answers instead of relying on ephemeral chat context.
Incorrect -- relying on Claude's memory for past decisions:
User: "What did we decide about the auth architecture last week?"Claude: "I don't have context from previous sessions..."
Correct -- add decisions as sources, query later:
# 1. Create a project notebooknotebook_create(name="Project Alpha Decisions")# 2. Add decision documents as sourcessource_add(notebook_id="...", content="ADR-001: Use JWT for auth because...")source_add(notebook_id="...", content="ADR-002: PostgreSQL over MongoDB for...")# 3. Capture new insights with notesnote(notebook_id="...", content="Perf test showed 2x latency with Redis cache miss")# 4. Query for grounded answersnotebook_query(notebook_id="...", query="What auth approach did we choose and why?")
Key rules:
One notebook per project or domain -- avoid mixing unrelated topics
Add decision records, design docs, and meeting notes as sources
Use note tool to capture in-session insights for future retrieval
Use notebook_query for grounded answers backed by actual sources
Periodically prune outdated sources to keep answers relevant
Control notebook access with sharing tools. Always check current sharing status before modifying access -- public links expose all notebook content.
Incorrect -- sharing publicly without reviewing content:
# Dangerous: makes everything in the notebook publicly accessiblenotebook_share_public(notebook_id="...", enabled=true)# Notebook contains internal security advisories -- now exposed
Correct -- check status, review, then share deliberately:
# 1. Check current sharing settingsstatus = notebook_share_status(notebook_id="...")# Shows: public link (on/off), list of collaborators, permission levels# 2. Review notebook content for sensitive materialnotebook_query(notebook_id="...", query="Does this notebook contain credentials or internal secrets?")# 3a. Share with specific collaborators (preferred)notebook_share_invite( notebook_id="...", email="colleague@company.com", role="reader" # or "editor")# 3b. Or enable public link (use with caution)notebook_share_public(notebook_id="...", enabled=true)
Batch sharing — invite multiple collaborators at once:
# Single call for multiple invites with mixed rolesnotebook_share_batch( notebook_id="...", invitations=[ {"email": "dev@company.com", "role": "editor"}, {"email": "pm@company.com", "role": "reader"}, {"email": "lead@company.com", "role": "editor"} ], confirm=True)
Key rules:
Always call notebook_share_status before modifying sharing settings
Prefer notebook_share_invite (or notebook_share_batch for multiple) over public links
Review notebook content for sensitive material before enabling public access
Use role="reader" by default -- only grant "editor" when collaboration is needed
Use notebook_share_batch for 3+ collaborators -- single API call vs. multiple invites
Disable public links when no longer needed: notebook_share_public(enabled=false)
NotebookLM Studio generates 10 artifact types from notebook sources. All generation is async -- create, poll status, then download.
Incorrect -- calling studio_create and waiting synchronously:
# Blocks for 2-5 minutes with no feedbackresult = studio_create(notebook_id="...", type="audio_overview")# User sees nothing until completion or timeout
Correct -- create, poll, download:
# 1. Create the artifact (returns immediately with artifact ID)artifact = studio_create(notebook_id="...", type="audio_overview")# 2. Poll for completionstatus = studio_status(artifact_id=artifact.id)# status: "pending" | "processing" | "completed" | "failed"# 3. Download when completeddownload_artifact(artifact_id=artifact.id, path="./output/podcast.mp3")
Visual summary with narration (slides + voiceover). Use video_format="explainer"
video
MP4 video
3 formats: explainer, brief, cinematic. 9 visual styles. Cinematic requires Plus/Ultra, English only, max 20/day
mind_map
SVG/PNG
Visual topic relationships
quiz
JSON
Test comprehension of sources
flashcards
JSON
Study aid from source material
slide_deck
PDF/PPTX
Presentation from sources
infographic
PNG
Visual data summary (11 visual styles available)
data_table
CSV/JSON
Structured data extraction
report
PDF/Markdown
Comprehensive written summary
Infographic visual styles (via infographic_style param on studio_create):
Style
Description
auto_select
Let NotebookLM choose (default)
sketch_note
Hand-drawn sketch style
professional
Clean corporate layout
bento_grid
Modular grid layout
editorial
Magazine-style design
instructional
Step-by-step educational
bricks
Block/brick composition
clay
3D clay/plasticine look
anime
Anime-inspired visuals
kawaii
Cute Japanese style
scientific
Technical/scientific format
Revising slide decks with studio_revise:
# 1. Get the artifact_id from studio_statusstatus = studio_status(notebook_id="...")# 2. Revise specific slides (creates a NEW deck, original is preserved)studio_revise( notebook_id="...", artifact_id="existing-deck-id", slide_instructions=[ {"slide": 1, "instruction": "Make the title more concise"}, {"slide": 3, "instruction": "Add a comparison table"} ], confirm=True)# 3. Poll studio_status for the new revised deckstudio_status(notebook_id="...")
Exporting artifacts to Google Workspace with export_artifact:
# Export a report to Google Docsexport_artifact(notebook_id="...", artifact_id="...", export_type="docs")# Export a data table to Google Sheetsexport_artifact(notebook_id="...", artifact_id="...", export_type="sheets")
Multilingual content generation:
# Generate a Hebrew infographicstudio_create( notebook_id="...", artifact_type="infographic", infographic_style="bento_grid", language="he", confirm=True)# Generate a Japanese audio overviewstudio_create( notebook_id="...", artifact_type="audio", audio_format="deep_dive", language="ja", confirm=True)
Supported language values (BCP-47): en, he, ar, es, fr, de, ja, ko, pt, zh, ru, hi, and more. Defaults to en or NOTEBOOKLM_HL env var.
Create a dedicated NotebookLM notebook for each OrchestKit release to preserve release context, changelog details, and key skill diffs as a queryable knowledge base.
# Dump everything into one shared notebooksource_add(notebook_id="shared", type="text", title="v7 + v6 + v5 notes", content="<all changelogs mixed together>")
Correct:
# One notebook per minor version with focused sourcesnotebook_create(title="OrchestKit v7.0 Release Notes")source_add(notebook_id="v7", type="text", title="CHANGELOG v7.0.0", content="<v7.0.0 changelog section only>")
# Query specific release contextnotebook_query(notebook_id="...", query="What changed in the implement skill for v7.0?")# Compare across releases (query the relevant notebook)notebook_query(notebook_id="...", query="What breaking changes were introduced?")