Back to Skills
Research

NotebookLM Query

Get better results from NotebookLM through Claude Code. A skill that improves on the default MCP/CLI experience with smarter query handling and follow-up chains.

March 8, 2026
notebooklmresearchquerying

The Problem With the Default Experience

NotebookLM has an MCP server and a CLI tool (nlm). You can connect them to Claude Code and query your notebooks. In theory, this is great — your curated knowledge base, accessible from the terminal.

In practice, the out-of-the-box experience is underwhelming. Claude tends to restructure your natural language question into something more "formal" before sending it to NotebookLM, and the backend actually responds worse to that. Follow-up queries are treated as isolated questions instead of building on previous context. Source filtering is possible but not obvious.

What This Skill Fixes

Three things:

Pass queries verbatim. The single biggest improvement. NotebookLM's backend is tuned for natural language. When Claude rewrites "What does the McKinsey report say about AI adoption?" into a structured query, the results get worse. This skill instructs Claude to pass your question as-is.

Chain follow-up queries. Instead of treating each query as independent, the skill builds context from previous answers into follow-up questions. "Tell me more about the adoption barriers" becomes "Based on the McKinsey report's mention of three adoption barriers — skills gaps, data quality, and integration complexity — which of these does the Gartner source also discuss?" Much better results.

Filter by source. When a notebook has many sources and you want information from a specific one, the skill knows how to target it. Mention the source by name and the query narrows appropriately.

The Tool Exists; This Skill Makes It Useful

This is a pattern worth recognising: many MCP tools and CLI integrations work technically but produce mediocre results because Claude doesn't know the quirks of the backend. A skill file that encodes those quirks — "pass queries verbatim", "chain follow-ups", "the API rate-limits rapid queries" — turns a functional tool into a genuinely useful one.

The skill doesn't replace the nlm CLI. It teaches Claude how to use it well.

Setup

  1. Install the nlm CLI tool and authenticate
  2. Download the skill file below
  3. Create a directory: .claude/skills/notebooklm-query/
  4. Save the file as SKILL.md in that directory
  5. Ensure you have at least one NotebookLM notebook with sources added

Ready to use this skill?

Drop it into your .claude/skills/ folder and you're ready.

Download Skill File

This guide was my gift to you. I want everyone to be able to punch above their weight class by leveraging AI to do more with what they've got.

If this helped and you want to know how I help companies through AI consulting, mentoring, or workshops — sign up for my email list or reach out below.

Or schedule a conversation →