deep-research-notebooklm — Claude Code skill: research via NotebookLM
Runs multi-source research on NotebookLM and writes a structured brief to a local file, with optional slides, podcasts or reports.
npx claude-code-templates@latest --skill ai-research/deep-research-notebooklmWhat deep-research-notebooklm does
The skill lays out the research as steps. It picks a research type first — market, competitive intel, prospect, trend, proposal or academic — confirms the angle with you, then creates a NotebookLM notebook, seeds it with your URLs and documents, starts research and imports the sources it finds. After that the agent asks a few targeted questions, pulls the answers together and saves the brief to research/[topic-slug]-[YYYY-MM-DD].md using research-brief-template.md. The research/ folder is created if missing.
You need the NotebookLM MCP server configured: nlm setup add claude-code. Without it the tools the skill calls — notebook_create, research_start, notebook_query — are not there, so check the setup before you run anything. Fast mode is the recommended default; deep mode goes longer and can stall at zero sources, so the skill suggests it only when you ask for exhaustive coverage.
Once the brief is saved you can request studio artifacts: a slide deck, audio podcast, video, infographic, report or mind map. They are generated with studio_create, tracked through studio_status and saved with download_artifact. Audio and video take longer to finish than slides.
What it helps with
- build a market brief with cited sources
- map competitors by positioning and features
- prep a prospect dossier before a call
- turn finished research into a podcast or slide deck
- surface risks and counterarguments on a topic
Example prompts for the agent
- research the CI/CD market for small teams
- run competitive intel on three rivals and save the brief
- generate an audio podcast from this research
How to install deep-research-notebooklm
1claude-code-templates
npx claude-code-templates@latest --skill ai-research/deep-research-notebooklm2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill deep-research-notebooklm -a claude-code3Manually: put the files in place
~/.claude/skills/deep-research-notebooklm/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/deep-research-notebooklm/
Manually: put the files in place
.agents/skills/deep-research-notebooklm/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/deep-research-notebooklm/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates deep-research-notebooklm2Manually: put the files in place
.github/skills/deep-research-notebooklm/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/deep-research-notebooklmManually: put the files in place
.devin/skills/deep-research-notebooklm/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about deep-research-notebooklm
How do I install deep-research-notebooklm in Claude?
Use the templates package: npx claude-code-templates@latest --skill ai-research/deep-research-notebooklm. Alternatively run npx skills add davila7/claude-code-templates --skill deep-research-notebooklm -a claude-code.
What does deep-research-notebooklm do?
It runs structured research on a topic through the NotebookLM MCP server and saves the findings as a brief file. It can also produce slides, podcasts, videos, infographics, reports and mind maps from that research.
What has to be set up first?
The NotebookLM MCP server, via nlm setup add claude-code. If the NotebookLM tools are missing from the session, the skill tells you to run that command and restart.
