# How to Build a Free AI Research and Content System Using NotebookLM and Claude Code
By using NotebookLM and Claude Code you can make a powerful AI workflow. It feels like these two tools were made to work
You can set up a system in a few minutes that does the following things:
- Gets real data from YouTube
- Breaks down sets of data
- Turns insights into slides or infographics or summaries that sound like podcasts
- Runs without needing paid API tokens
is a simple step-by-step guide that anyone can follow.
Step 1: Create a YouTube Data Tool in Claude
First you need to set up Claude to get YouTube data.
Open Claude Code and type in the following:
"Create a Python-based YouTube search tool using yt-dlp. It should get video titles, channel names, view counts and extract captions."
Claude will make a working script that can handle data extraction.
Step 2: Add a Bridge for NotebookLM
NotebookLM does not work with tools on its own so you need a free bridge.
You need to do a things:
- Open a new terminal window. Keep Claude running in the one.
- Go to this GitHub repository:
https://github.com/teng-lin/notebooklm-py.git
- Follow the installation instructions in the repository.
- Run the setup command to install the tool.
This will allow NotebookLM and your local workflow to talk to each other.
Step 3: Link Your Google Account to NotebookLM
Next you need to connect the tool to NotebookLM.
In the terminal window type:
notebooklm login
A browser window will open. Sign in with your Google account.
You only need to do this
Step 4: Give Claude Control Instructions for NotebookLM
Now Claude needs instructions on how to work with NotebookLM.
You need to do a things:
- Go back to the same GitHub repository.
- Find the "Claude Skill" section.
- Copy the instruction block.
- Paste it into Claude Code.
This will allow Claude to make notebooks and manage outputs on its own.
Step 5: Run Everything with One Prompt in Claude Code
After you set everything up you can control the system with one prompt.
For example:
"Search YouTube for the 5 videos about AI tools, for online business and content creation. Add them to a NotebookLM notebook extract key insights identify trends and make a sketch-style infographic that summarizes everything."
Final Result of Using NotebookLM and Claude Code
This setup creates a mini AI research and content engine that can:
- Get data
- Look at the information
- Summarize insights
- Make audio-style outputs
All of this happens without needing complex infrastructure or paying ongoing costs. NotebookLM and Claude Code work together to make this possible.
