notebooklm-research - SKILL.md Agent Skill

name: notebooklm-research description: >- Run web and Google Drive research through NotebookLM -- start research queries, poll progress, and import discovered sources into notebooks. Use when the user asks to "research a topic in NotebookLM", "start NLM research", "web research and import sources", "Drive research", "deep research in NLM", "find sources for notebook", "discover sources", "import research sources", "research and build notebook", "NLM deep research", "poll research status", "NLM 리서치", "노트북LM 연구", "웹 리서치 시작", "소스 탐색", "리서치 시작", "딥 리서치", "드라이브 검색", "소스 발견", "NLM 웹 리서치", "리서치 결과 가져오기", or any NotebookLM research/discovery workflow. Do NOT use for notebook/source/note CRUD or querying -- use notebooklm. Do NOT use for content generation (audio, video, reports) -- use notebooklm-studio. Do NOT use for general web search without NotebookLM -- use parallel-web-search. Do NOT use for finance-specific web search -- use alphaear-search. metadata: author: thaki version: 1.0.0 category: research

NotebookLM Research: Web & Drive Discovery

Run research queries through Google NotebookLM to discover relevant sources from the web or Google Drive, then import them into notebooks. Uses the notebooklm-py CLI.

Prerequisites

notebooklm-py CLI installed and authenticated (see notebooklm skill)
A target notebook to import discovered sources into

Available Tools

Tool	Description	Key Parameters
`research_start`	Start web or Drive research	`notebook_id`, `query`, `mode` (web/drive/deep)
`research_status`	Poll research progress	`notebook_id`, `task_id`, `max_wait`
`research_import`	Import discovered sources	`notebook_id`, `task_id`, `source_ids` (optional)

Workflow

Standard research pipeline

Start research:

research_start(notebook_id, query="semiconductor supply chain trends 2026", mode="web")

Returns a task_id for tracking.

Poll until complete:
```
research_status(notebook_id, task_id, max_wait=300)
```
Returns discovered source list when done. max_wait is in seconds.
Import sources:
```
research_import(notebook_id, task_id)
```
Imports all discovered sources. Optionally pass source_ids to import a subset.

Deep research mode

For comprehensive research covering more sources:

research_start(notebook_id, query="AI trends in financial markets", mode="deep")

Deep research takes longer (2-5 min) but returns more comprehensive results.

Google Drive research

Search your Google Drive for relevant documents:

research_start(notebook_id, query="quarterly earnings report", mode="drive")

Selective import

After reviewing discovered sources, import only the relevant ones:

research_status(notebook_id, task_id) -- review source list
research_import(notebook_id, task_id, source_ids=["src_1", "src_3", "src_5"])

Polling Strategy

Research discovery typically takes 1-3 minutes (5+ for deep mode):

Call research_start to begin
Call research_status with max_wait=300 -- it will block until complete or timeout
If timed out, call research_status again
Once complete, call research_import

Stock Analytics Integration

Research-driven stock analysis notebook

Create notebook: notebooklm create "NVIDIA Supply Chain Analysis"
Start research: research_start(notebook_id, query="NVIDIA supply chain semiconductor 2026", mode="deep")
Wait: research_status(notebook_id, task_id, max_wait=300)
Import all: research_import(notebook_id, task_id)
Query: notebooklm chat <notebook_id> "What are the key supply chain risks?")
Generate podcast: notebooklm studio create <notebook_id> --type audio

Market event research

When a significant market event occurs:

Create notebook: notebooklm create "Fed Rate Decision March 2026"
Research: research_start(notebook_id, query="Federal Reserve rate decision March 2026 market impact", mode="deep")
Import sources and query for investment implications
Generate a briefing doc: notebooklm studio create <notebook_id> --type report

Troubleshooting

Symptom	Fix
Research returns no results	Try a broader or different query; check auth status
Import fails	Verify research completed via `research_status` first
Timeout on deep research	Increase `max_wait` to 600 seconds
Drive research empty	Ensure Google Drive is connected to the same Google account

CLI Reference

notebooklm research start <notebook_id> --query "search terms" --mode deep
notebooklm research status <notebook_id> --task-id <task_id>
notebooklm research import <notebook_id> --task-id <task_id>

Related Skills

notebooklm -- notebook/source/note CRUD and querying
notebooklm-studio -- content generation from research notebooks
alphaear-search -- finance-specific web search (Jina/DDG/Baidu)
alphaear-news -- real-time financial news aggregation

Examples

Example 1: Standard usage

User says: "Research a topic in NotebookLM"

Actions:

Gather necessary context from the project and user
Execute the skill workflow as documented above
Deliver results and verify correctness