Annoyed by NotebookLM's copy-paste hell, dev builds a 25k-user Web Clipper

Google's NotebookLM is dope, we all know that. But feeding it data? Absolute nightmare. When you're in the zone doing research, copy-pasting URLs one by one and losing video transcripts is enough to make anyone flip their desk. One annoyed dev decided he had enough of this BS, fired up his IDE, and built a Web Clipper to save us all.

From copy-paste hell to a 25k-user tool

Stéphane, the maker behind this tool, pitched his story on Product Hunt. He admitted this all started from a "dumb, repetitive pain." He just wanted to drop YouTube videos and web pages into NotebookLM to study them, but the native workflow was hot garbage.

He tried a few existing tools, but they were either clunky, intrusive, or just didn't work right. So, like any self-respecting dev, he built his own.

It started as a simple 1-click wonder: hit a button, and the web page, PDF, Reddit thread, or YouTube playlist magically lands in NotebookLM. But the more he used it, the more he realized NotebookLM was missing basic quality-of-life features. So, he turbocharged it:

• Jailbreaking data (Export): NotebookLM loves trapping your notes. Stéphane added 1-click exports to Anki (flashcards), Obsidian (mind maps), and Markdown/Word (full chats).
• Duplicate notebooks: A feature Google forgot to add.
• Bulk YouTube imports: Scrape whole channels or playlists while skipping dupes.
• Auto-sync Drive sources: Keeps your data fresh in the background.

He threw it out there as a freemium tool, and a few months later, it's sitting at over 25,000 users. Not bad for scratching your own itch.

What the Product Hunt hivemind is saying

The launch thread was mostly high-fives, but the community definitely had some analytical takes:

• The Productivity Nerds: The overwhelming consensus is that the ecosystem desperately needed this. Fixing the copy-paste friction is cool, but the ability to export out to Anki and Obsidian is the real MVP here. It essentially bridges NotebookLM with other powerhouse ai tools.
• The Data Skeptics: A user named Jim Jeffers brought up a big-brain point: freshness. When you export or sync a Drive file, how do you know if the data is from today or three weeks ago? A timestamp would build a lot of trust. Stéphane agreed and said he's waiting on Google's API updates to potentially grab that sync date.
• The API Watchdogs: One vigilant dev raised an eyebrow at the Reddit thread clipping feature. Given that Reddit has been nuking scrapers and locking down their API lately, how is this tool surviving? Are they scraping the DOM or using the API? Stéphane smoothly replied that they use Reddit's public API endpoints respectfully, staying well below rate limits since it's a personal clipper, not a mass scraper.

C4F Takeaway: The power of a pissed-off dev

The biggest takeaway here? The best software is built by lazy, annoyed developers. Instead of complaining about manual labor, write code to automate it.

In the current AI gold rush, too many startups are trying to build massive, world-changing foundation models and failing. Meanwhile, the real money and user adoption often lie in building "janitor" tools for Big Tech's platforms. Giants like Google or OpenAI build incredible core tech, but their UI/UX can be severely lacking. Find the friction, build a slick bridge, and you've got yourself a winning product.

Now, go find an app that makes you want to smash your keyboard and build a tool to fix it.

Source: Product Hunt